JOURNAL

OF

APPLIED

PHYSIOLOGY

Vol. 41, No. 6, December

1976.

Printed

in U.S.A.

Effect of altitude thermoregulatory C. M. BLA’ITEIS US Army Research

exposure on response of man to cold AND L. 0. LUTHERER Institute of Environmental

BLAWEIS, C. M., AND L. 0. LUTHERER. I3fict of c&&de exposure on thermoregulatory response of man to cold. J. Appl. Physiol. 41(6): 848-858. 1976. -The thermoregulatory responses to 10°C (for 3 h) were investigated in 1) 12 natives from sea level (lowlanders) at 150 m, and on arrival at 3,350 and 4,340 m; 2) 6 of these during a 6-wk sojourn at 4,360 m, and on return to sea level; and 3) 5 natives from each of the two altitudes (highlanders) in their respective habitat, and after descent to 150 m. The cold-induced increase in the rate of 0, consumption (00,) of the lowlanders was significantly smaller at both altitudes than at sea level. It did not recover substantially during the 6 wk at altitude, but was restored to its initial rate on return to sea level. By contrast, visible shivering activity was augmented on arrival at altitude. It persisted throughout the 6 wk there, but was greatly depressed on return to sea level, despite the increased vo,. Mean skin temperatures (Y&J stabilized in the cold at significantly higher values at altitude. Rectal temperature (Tr,) decreased similarly at all altitudes. vo, of the highlanders in the cold was significantly greater at sea level than at their resident altitudes, although shivering activity was less intense; Tsk stabilized at significantly lower levels at 150 m than at either altitude. These results indicate that altitude exposure reduces the calorigenic response of man to cold, and that this effect is not moderated by acclimatization to altitude, yet is reversible immediately on descent to sea level. The component of cold thermogenesis which appeared to be reduced by altitude exposure was nonshivering thermogenesis rather than visible shivering. shivering and nonshivering thermogenesis; body temperature; hypoxic hypoxia; mountain sickness; cardiopulmonary function; altitude acclimatization; oxygen consumption

IT IS NOW WELL ESTABLISHED that acute, moderate hypoxic hypoxia reduces in most mammals the increase in oxygen consumption normally induced by cold exposure (3). Attempts to confirm this effect in man, however, have not yielded consistent results, some workers reporting a depressed metabolic response to cold under hypoxia (19, 33, 39), and others finding no effect (1, 7, 9, 14,23,40,58>. Both types of response are in fact compatible with the presumed mode of action of hypoxia in other species. Thus, it has been suggested that moderate hypoxia may depress thermoregulatory heat production by selective inhibition of nonshivering thermogenesis (NST) (Z-6, 19,21, 56, 57). Since the capacity for NST may be inversely related to body size (31), its contribution to cold-induced thermogenesis in man, one of the

Medicine,

Natick,

Massachusetts

01760

larger species, may be insignifi cant; accordingly, its reducti .on by hypoxia might not be detectable. On the other hand, the significance of this NST may be greater than heretofore realized, in which case hypoxia might significantly depress it and thereby unmask the full extent of its contribution (4, 21). Unfortunately, shivering and nonshivering thermogenesis were not dissociated in any but one (19) of the earlier studies. Moreover, meaningful comparisons among previous reports are difficult because of vast differences in the experimental designs used, e.g., the manner of induction and the type of the hypoxic &imulu s, the degree and duration of both the hypoxic and cold stresses, the methods of assessing the thermoregulatory response, and, in some instances, the diverse anthropological origins of individual subjects within a given study. Additiona lly , no data have been reported on the thermoregulatory respon se to cold of sea-level nati ves acclimatized to altitude or, conversely, of altitude natives descended. to sea level, as possible measures of the persistence of the hypoxic effect on cold thermogenesis. Although the adaptive changes in the internal body and cutaneous thermal responses of man to altitude and cold have been investigated previously in different ethnic groups (1, 19, 20, 23, 24, 30, 4246, 49), the course of these changes and the associated adj ustments in the calorigenic response of sea level natives during altitude acclimatization also have not yet been reported. The present report, therefore, is concerned with a) the thermoregulatory response to cold of sea-level and altitude natives exposed acutely to various altitudes, and b) the effect on this response in sea level natives during 6-wk residence at altitude. METHODS

These experiments were conducted in the Central Andes of Peru, during the local win .ter season. The subjects included 12 Peruvian soldiers, born and reared at sea level (lowlanders), and 10 Andean farm workers or miners, indigenous to two di.fferent high altitudes (highlanders). Their physical characteristics are summarized in Table 1. None of the subjects had any recent (less than 1 yr) or long-term (more than 3 days) exposure to any other than his own resident altitude. Histories were taken and physical examinations performed on all the subjects to insure that they were healthy. Two series of experiments were conducted, in se-

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

MAN,

ALTITUDE,

849

COLD

ment period. Inspired air volumes (VI) were determined by reading from the continuously cumulating dial of a Lowlanders Highlanders low-resistance mass flowmeter (see ADDENDUM)incorpoLima Pachacayo Cochas rated into the intake side of a double-J valve. Expired 150 m 3.330 m 4,360 m air was collected in Douglas bags, and the fraction of 0, N 12 5 5 in the samples was analyzed using a Beckman E-2 20 (17-22)* 21 (18-24) 22 (18-25) Age, yr analyzer. The rate of oxygen consumption (Vo& was Weight, kg 62.8 (59.8-66.3) 50.9 (47.9-56.5) 56.4 (51.9-64.7) 156.4 (152.5-164.0) Height, cm 160.5 (154.5-167.0) 151.9 (150.5-154.0) calculated from these values. Body temperatures were 1.66 (1.62-1.71) 1.45 (1.40-1.54) 1.55 (1.46-1.70) Surface area, m’ measured by means of appropriate thermistor probes Hematocrit, % 47.0 f 1.0t 51.2 + 1.2 54.7 + 1.8 and telethermometers (Yellow Springs Instrument Co.). * Values are means 2 ranges in parentheses. t Hematocrit values are means 5 SE. Probes were located 10 cm in the rectum (T,,) and on the skin (forehead, abdomen, thigh, calf, toe, forearm, and quence. In the first, the lowlanders were tested initially finger). Mean skin temperature (Tsk) was derived acat sea level (Lima: 150 m, PB = 748 Torr, mean winter cording to the formula of Mitchell and Wyndham (48) temp = 17°C) and, subsequently, within 6 h after arrival for seven measuring sites. Shivering activity was monifrom sea level at each of two altitudes (Rio Blanco: 3,350 tored visually and quantified by observers (one per pair m, PB = 504 Torr; and Casapalca: 4,340 m, B = 442 of subjects) according to an arbitrary scale (see legend, Torr). The highlanders, in turn, were tested initially in Fig. 2). Alveolar PO, and Pco2 (PACTand PACES,by the their own habitat (Pachacayo: 3,330 m, PB = 506 Torr, end-expiratory method of Rahn and Otis (51)), pulse winter Ta = 11°C; and Cochas: 4,360 m, PB = 440 Torr, rate (HR, by palpation), arterial 0, saturation (SaO,, by winter T, = 8°C) and, subsequently, within 6 h after earlobe oximetry), and hematocrit (Hct, arterialized finarrival at sea level (Lima). A l-wk interval separated ger-capillary sample) were measured at the end of the each consecutive test. During this time, all the subjects control and cold exposure periods. Comments by the test lived at their habitual, resident altitude; the lowlanders subjects and qualitative evaluations by the observers were quartered in a military hospital in Lima, while the were recorded at any time. highlanders engaged in their usual activities and ate The data were evaluated statistically using analysis regular meals in their own homes. Transportation be- of variance models. All the analyses took one of four tween sites was by enclosed vehicle, with the subjects forms. The first was a simple one-way design, with only recumbent on air mattresses and provided with blantwo categories of classification (lowlanders and highlankets. ders). This design was used to compare a given response In the second series of experiments, which began 1 wk at 26°C of the lowlanders with that of the highlanders, after completing the first, six of the lowlanders, chosen under the same or different altitudinal conditions. The at random, were retested initially at sea level (SLi). One second was a one-way design in randomized blocks, with week later, they were transported to 4,360 m altitude only two categories of classification (sea level and alti(Cochas) by enclosed vehicle, while recumbent on air tude, or two different durations of exposure to the same mattresses and provided with blankets. They were altitude). This design was used to compare a given tested within 6 h after arrival (wk 0), and after 1, 2, 4, response at 26°C of either the lowlanders or the highlanand 6 (wk 6) wk of residence. They were returned to ders at two different altitudes, or of the lowlanders at Lima at the end of the 7th wk at altitude and tested two different durations of exposure to the same altitude; again within 6 h after their descent (SL,). Between each block consisted of measurements made on a single tests, the lowlanders enjoyed the freedom of Cochas. subject. The third was a two-factor factorial design with They dressed for thermal comfort, and consumed ad repeated measures on one factor. This design was used libitum their habitual diet (brought up twice weekly, to compare the course of a given response of the lowlanuncooked, from sea level and prepared daily in amounts ders with that of the highlanders during cold exposure as needed, by the subject themselves). At least two under the same or different altitudinal conditions. The observers supervised the lowlanders at all times. fourth was a two-way design in randomized blocks. This At all altitudes, two adjoining temperature-controlled design was used to compare the course of a given rerooms were used, one maintained at 26°C (control) and sponse of either the lowlanders or the highlanders at sea the other at 10°C (cold). After appropriate measuring level and at altitude during exposure to cold, or of the devices were attached, the subjects, clad only in shorts, lowlanders following two different durations of exposure were allowed to stabilize, reclining on webbed lawn to the same altitude; each block consisted of measurechairs, for 2.5 h in the 26OC environment before any ments made on a single subject. An IBM 360 model 40 measurements were begun. Three consecutive lo-min computer was programmed to perform these analyses. measurements were made in the 26°C room; the subjects Shivering scores were analyzed nonparametrically (52). then were carried in their own lawn chairs to the 10°C The 5% level of confidence was accepted as statistically room, where additional lo-min measurements were ob- significant in all the analyses. tained at 35-min intervals during the following 3 h. All the subjects were postabsorptive. They were trained in RESULTS the procedures used in this study prior to initiation of Effects of Acute Altitude Exposure the project. Thermoregulatory responses. RATE OF o2 CONSUMPRespiratory metabolism was measured by the opencircuit technique over the duration of each measure- TION (FIG. 1). Exposure of the lowlanders to 10°C proTABLE

1. Initial

AND

physical

characteristics

of subjects

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

850

C. M. LOWLANDERS from 150m

HIGHLANDERS from 3330m

HIGHLANDERS from 4360m

BLATTEIS

LOWLANDERS

AND from

L. 0.

LUTHERER

150m

ik.

iO,p ml/kg per min

T-I

1

0

I

1

,

4-

2

0 2 TIME IN COLD (hours) FIG. 1. Course of the rate of oxygen consumption (\joz) of 12 lowlanders (LOL) and two groups of 5 highlanders (HIL) during 3h in 10°C at sea level (SL) and two altitudes (ALT I and II, 3,350 and 4,340 m, and 3,330 and 4,360 m, respectively). Points are mean values; vertical lines indicate SE. First point (immediately preceding time 0) is the mean of three consecutive measurements in 26”C, just prior to transfer into the cold.

duced at sea level an immediate and significant increase with only slight abatement in Vo,, which continued throughout the 3 h of this exposure. At 3,350 m, the cold also caused a significant initial rise of Vo, in these subjects; however, the increase ended after only 35 min, and the maximum rate achieved was significantly less than at sea level. At 4,340 m, the lowlanders exhibited no significant increase in Vo, at 10°C although the same maximal rate was attained as that at 3,350 m; however, the initial value at this altitude at 26OC was significantly higher than that at the other two altitudes. The differences in the overall responses to cold of the lowlanders between sea level and either altitude were statistically significant. The maximum Voz attained at altitude was 20% smaller than that at sea level. The Vo, response to, 10°C of the highlanders from both the 3,330 and 4,360 m levels was significantly smaller (by 21% at m aximum response) at their resident altitudes than at sea level. Also, whereas at sea level there was from the beginning a progressive rise in Vo,, at altitude this rise was preceded by an initial fall. Although the values at both higher altitudes tended to be elevated, there were no statistically significant differences in the highlanders’ Vo2 between sea -level and either altitude at 26°C. Comparing the lowlanders with the highlanders, there were no significant differences at sea level among the Vo,‘s of these groups at both 26” and 10°C. However, at altitude, the patterns of their cold-induced increase in Vo, were significantly different. Thus, not only did the highlanders show an initial fall in Vo, upon exposure to cold, but also at 4,360 m (the highest altitude) the cold-induced increase in Vo, was significantly greater than that of the lowlanders at the same relative altitude. Nevertheless, the extent of the depression of the cold-induced increase in Vo, caused by altitude exposure was of the same order of magnitude for both lowlanders and highlan-ders (20 and 21%, respectively). SHIVERING (FIG. 2). The lowlanders, both at sea level and at altitude, generally exhibited rather vigorous, albeit discontinuous, shivering movements, evident

2G

0,

-

-

-

7 I

1 HIGHLANDERS

from

3330m

HIGHLANDERS

from

4360m

I

TIME IN COLD (hours) 2. Visible shivering activity, recorded at intervals in arbitrary units (0 = no shivering activity; 1 = mild shivering in bursts; 2 = generalized, but discontinuous, shivering; 3 = very marked and continued shivering movements), of 12 lowlanders and two groups of 5 highlanders during 3 h in 10°C at sea level (SL) and two altitudes (ALT I and II). FIG.

within 10 min after the onse t of cold exposure. At sea level, the intensity of these tremors tended to abate during the following hour, but then gradually grew again-until, 2 h after entry i nto the cold environment, shivering was very marked and continuous. At 3,350 and 4,340 m, on the other hand, the transient abatement of shivering activity during the first hour after the initial 10 min at 10°C tended to be brief or even absent, so that it appeared that the shivering of the lowlanders progressed at altitu .de to vigorous and continual activity earlier than at sea level. The highlanders shivered in the cold at their resident altitudes in a manner not appreciatively different from that of the lowlanders newly arrived at the same altitudes. By contrast, they did not exhibit at sea level (except for bursts of mild shivering on initial exposure to cold) any visible tremors until after the first hour at 10°C. These movements appeared then to grow in intensity more slowly than previously at altitude, and indeed did not attain the same maximum level as at altitude, even after 3 h in the cold.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

MAN, BODY

ALTITUDE,

AND

851

COLD

TEMPERATURES

(FIG.

3). i) Rectal temperature

(upper panels). At a 11aIt i t u des, including sea level, cold exposure resulted in a progressive and significant fall of the T,, of the lowlanders. At 26"C, T,, was not different at any altitude. Exposure to cold at their resident altitudes induced a gradual, but significant, fall of the T,, of both groups of highlanders. However, the overall response was not significantly different from that at sea level. At 26”C, there were no significant differences between these temperatures at sea level and at both altitudes. No differences also were observed between lowlanders and highlanders either for the T,& at 26°C or for the extent of their fall at 10°C. ii) Mean skin temperature (lower panels). The TSkof the lowlanders at all altitudes fell rapidly upon initial exposure to cold, then became stabilized after approximately 60 min. The total decline was significantly smaller at both higher altitudes than at sea level. Altitude had no significant effect on TSk at 26°C. The Tsk’s of both groups of highlanders fell during exposure to 10°C. In both groups, the fall at their resident altitudes was significantly smaller than that at sea level. Initial temperatures (at 26”C), however, were not different at both altitudes. At altitude, the initial T,,‘s of the highlanders were significantly higher at 26°C than those of the lowlanders at the equivalent altitude, but the degree of fall of these temperatures at 10°C was similar in all groups. Ventilatory responses (Fig. 4). The VI of the lowlanders at 26OCwas significantly increased proportionately with each successive altitude above sea level. At lO”C, their VI rose further under the three altitudinal conditions; however, the maximum VI'S attained were the same, so that the relative increments became progressively smaller at each higher altitude. The VI of the highlanders from 3,330 m was significantly higher at 26OC at their resident altitude than subsequently at sea level. It also was significantly LOWLANDERS from 150m

HIGHLANDERS from 4360m

HIGHLANDERS from 3330m

c

37.0 -

1 re,oC

0

35.0 34.0

’

’

’

’

’

’

HIGHLANDERS from 4360m

HIGHLANDERS from 3330m

LOWLANDERS from 150m

22

18 QE I /min (BTPS) 14 I

10 O

t !+

I

,

,~,

2

0

,

,

,r,

,

2

0 TIME

0

,

,

2

IN COLD (hours)

4. Course of the minute ventilation (VE) of 12 lowlanders (LOL) and two groups of 5 highlanders (HIL) during 3 h in 10°C at sea level (SL) and two altitudes (ALT I and II). Symbols and conditions same as in Fig. 1. FIG.

higher than that of the lowlanders at 3,350 m and at sea level, respectively. Upon exposure to cold, the VI of this group at 3,330 m increased significantly more than did that of the lowlanders at the same altitude. At sea level, however, the cold-induced increase was in the same order of magnitude as that of the lowlanders. By contrast to the other groups, the VI of the natives from 4,360 m was not different at 26°C at sea level and at altitude. The value at sea level, however, was significantly greater than the corresponding VI of the lowlanders, but equal to that of the highlanders from 3,330 m. The value at altitude, on the other hand, was significantly lower than that of the highlanders from 3,330 m at their resident altitude. The ventilatory increase induced by cold exposure in this group at altitude was significantly greater than that in the lowlanders at this altitude. At sea level, on the other hand, it was equal in magnitude to that of the other groups. Other cardiorespiratory responses (Table 2). At 26”C, the lowlanders exhibited a significant decrease in PACES, PACT, and Sao,, at both 3,350 and 4,340 m as compared to sea level. The PAN* and Sao, of the highlanders also were significantly less at their resident altitudes than at sea level. However, no difference in ~~~~~ was observed in the highlanders from 3,330 m between sea level and

2. Cardiorespiratory functions of 12 lowlanders and two groups of 5 highlanders at T, = 26”C, both at their resident altitudes and during acute exposure to opposite altitudinal condition TABLE

’

’

’ . .

’

’

’

:o

HR, beats/ min Lowlanders (150 m)

Tsk 9“C 26.0

I

0 TIME

IN COLD

I

I

1

2

(hours)

3. Rectal (T,,) and mean skin (Y&J temperature changes of 12 lowlanders (LOL) and two groups of 5 highlanders (HIL) during 3 h in 10°C at sea level (SL) and two altitudes (ALT I and II). Symbols and conditions same as in Fig. 1. FIG.

Highlanders (3,330 m) Highlanders (4,360 m)

SL 3,350 m 4,340 m SL ALT SL ALT

37.0 31.0 29.0 32.5 33.3 32.3 30.4

f + 2 2 +f t

0.9 0.6* 0.8* 2.2$ 3.7 0.73 0.4*

101.6 52.5 43.1 108.6 52.6 104.2 43.8

-t 2.1 2 1.3*-f 2 1.7* 2 2.8 -t 3.6* 2 2.1 _+ 2.1*

97.1 87.2 83.0 93.6 88.3 94.6 82.3

2 + + k 2 + 5

0.3 l.l* 1.4* 1.0 3.0* 1.0 1.3*

56 71 76 52 61 44 65

2 ?I + + + 2 2

2 4*t 4*t 2 2* 2$ 2*

Values are means 2 SE. SL = sea level; ALT = resident altitude; Pk.o, = alveolar CO, tension; P&,, = alveolar 0, tension; Sqq,, = arterial 0, saturation; HR = heart rate. * Value at altitude significantly different from corresponding value at sea level. t Value of lowlanders at altitude significantly different from value of highlanders at same altitude. $ Value of highlanders at sea level significantly different from value of lowlanders at sea level.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

852

C. M.

their resident altitude; on the other hand, that of the highlanders from 4,360 m was lower at altitude than at sea level. Also, the ~~~~~ of the lowlanders at both altitudes tended to be lower than the corresponding values of the highlanders; but at sea level it was higher. The resting HR’s of both the lowlanders and the highlanders were significantly higher at altitude than at sea level, the relative changes produced by exposure to the opposite altitude being similar in magnitude in all the groups. However, the HR of the lowlanders was significantly higher at both high altitudes than the corresponding rate of the resident highlanders, while that of the natives from 4,360 m was strikingly lower at sea level than that of the other groups. Exposure to 10°C (not shown in Table 2) induced different effects on these variables at sea level and at altitude. Thus, at sea level, it caused small rises of ~~~~~ and HR and falls of PA~~ and Sao,, whereas at both altitudes it produced a fall of PA but increases of PA~~, SG,, and HR, in all the subjects. Subjective responses. No lowlander reported that the 26°C environment felt any different at any altitude than it did at sea level; nevertheless, a few subjects complained of lightheadedness and mild headache upon arrival at altitude. During the exposure to cold, however, more subjects reported the development of headaches and, furthermore, complained of nausea, tingling of fingers and toes, and of leg cramps. The severity and incidence of these complaints were greatest at 4,340 m. All the lowlanders reported without solicitation that they felt colder during their exposure to 10°C at altitude than they had at sea level. No affective complaints were made by the highlanders during cold exposure at any altitude. They did report, however, that 10°C sea level felt warmer there than it had at their resident altitude. co29

Effects of Altitude

Acclimatization

Thermoregulatory responses. RATE OF o2 CONSUMP(FIG. 5). A small, but significant, increase in the basal (i.e., in 26°C) voz occurred in the lowlanders on arrival at 4,360 m ( wlz 0, left panel); during the subsequent 6 wk at altitude, however, it was reversed to beyond its original value at sea level. Cold exposure evoked in these subjects a significant increase in their voe both at sea level and at altitude (same panel). However, the magnitude of this increase (i.e., AOo,, computed as the difference between the basal and the third hour cold values) was significantly smaller throughout the lowlanders’ stay at altitude than at sea level (right panel). By comparison, the magnitude of the coldinduced increase in irO, of the highlanders (data replotted from Fig. 1, right panel) was significantly higher at 150 m than that of the lowlanders; at 4,360 m, it was higher only in comparison with that of the latter on their arrival (wk 0). SHIVERING (FIG. 6). Generalized, but discontinuous, shivering developed in the lowlanders within minutes after the onset of cold exposure, both at sea level originally (SLi) and on arrival at altitude (wk 0). However, at sea level, this initial response generally diminished TION

i(,,, ml/ kg per min

BLATTEIS

AND

L. 0.

LUTHERER

9-

i

7-

t

sOL

1

0

I

1

HOURS IN COLD

1

t

2

3

I

I

SLi 0 WEEKS

2

1

I- iI

4

1,

6 SLf

AT ALTITUDE

5. Left panel: course of the rate of oxygen consumption (vo,) of 6 lowlanders (LOL) during 3 h in 10°C initially at sea level (SLi), on arrival at 4,360 m (wk 0), after a 6-wk sojourn there (wk 6), and on return to sea level (SL,). Points are mean values; vertical lines indicate SE. First points (immediately preceding time 0) are the means of three consecutive measurements in 26”C, just prior to transfer into the cold. Right panel: magnitude of cold-induced increase in \io, (Avo,) at sea level initially (SLi), during 6 wk at 4,360 m, and on return to sea level (SL,), calculated as difference between value in 26OC and that after 3 h in 10°C. Corresponding responses of 5 highlanders from 4,360 m (HIL) at their resident altitude and at sea level are replotted from Fig. 1. Points are mean values; vertical lines indicate SE. FIG.

significantly within the next 15 min, remained low during the 1st h, then gradually increased again in intensity, reaching a sustained, maximum activity only after 2 h in the cold. By contrast, the transient abatement of shivering was not observed on arrival at altitude, and vigorous and continued movements occurred significantly earlier. But, after 6 wk of residence at 4,360 m (wk 6), the intensity of the initial shivering response to cold was smaller, and maximum activity-was not observed until just prior to the end of the exposure. Upon return to sea level (SL,), shivering activity was significantly lower than at any time previously. The course and intensity of the shivering response of the lowlanders newly arrived at 4,360 m and upon their return to 150 m conformed exactly to those of the highlanders at altitude and sea level, respectively. BODY TEMPERATURES (FIG. 7). i) Rectal temperature (upper panels). No significant effect of altitude as compared to sea level was apparent on the T,, of the lowlanders, both at 26 and 10°C (left panel). The tendency of T,, to decrease during the course of the 3-h cold exposure diminished gradually as the experimental weeks continued (right panel); this fall was significantly smaller at sea level after than before the 6 wk at altitude. The coldinduced fall of T,, of the highlanders (data replotted from Fig. 3, upper right panel) also was similar at sea level and altitude, although it was greater at sea level than that of the altitude-acclimatizing Jlowlanders. ii) Mean skin temperature (lower panels). Acute exposure of the lowlanders to 4,360 m (wk 0) produced no change in their Tsk in 26OC. However, after 6 wk at this altitude (wk 6), Tsk was significantly higher than originally at sea level (SLi); moreover, it remained elevated upon their return to sea level (SL,). Exposure to cold induced a fall, first rapid, then more gradual, of the Tsk of these subjects both at sea level and altitude, and the final temperatures attained were significantly higher at 4,360 than at 150 m (left panel). However, the magnitude of the fall of Tsk was not different during any week

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

MAN,

ALTITUDE,

AND

853

COLD 37.01

LOWLANDERS

I

\

-3

1rep “C

\

WEEK 0

4r

WEEK 6

kz

0

I

1

I

I

I

.

I

4r 2oD

’

-’

I 2 1 TIME IN COLD (hours)

FIG. 6. Visible shivering activity, recorded trary units (see legend, Fig. 2), of 6 lowlanders (SLi), on arrival at 4,360 m (wk O), after a 6-wk and on return to sea level (SL,). Corresponding landers from 4,360 m at their resident altitude replotted from Fig. 2.

I

t,

2 IN COLD

3 WEEKS

AT ALTITUDE

7. Left panels: Course of the rectal (T,,,) and mean skin (T,,> temperatures of 6 lowlanders (LOL) during 3 h in 10°C at sea level initially (SLi), on arrival at 4,360 m (wk O), after a 6-wk sojourn there (wk 6), and on return to sea level (SL,). Symbols and conditions same as in Fig. 5. Right panels: magnitude of the cold-induced changes in at sea level initially these variables (AT,, and ATsk, respectively) (/Xi), during 6 wk at 4,360 m, and on return to sea level (SL,). Corresponding changes of 5 highlanders from 4,360 m (HIL) at their resident altitude and at sea level are replotted from Fig. 3. Symbols and conditions same as in Fig. 5.

-

-

bi

FIG.

HIGHLANDERS -

1 HOURS

I

2.

1

t

bk, “c

SL

r

t,

0

OL

0

4

‘:, 0.S

0‘ ---

/

\

i/l\

ATre

36.0

4r

\

\

1 3

at intervals in arbiat sea level initially sojourn there (wk 6), responses of 5 highand at sea level are

of these experiments, excepting upon arrival at altitude when it was smaller than at any other time (right panel). The extent of fall of Tsk of the highlanders (data replotted from Fig. 3, lower right panel) was in the same order of magnitude as that of the lowlanders under both altitudinal conditions, excepting on ulk 0 and SLf, when it was greater. Cardiorespiratory responses (Table 3). On initial exposure to altitude (wk 0), the lowlanders at 26OC characteristically exhibited significant increases in VI and HR, and significant decreases in PACES, PACT, and Sao,, relative to sea level previously (SLi). Functional changes developed as their residence at 4,360 m continued, evidenced on the 6th week (wk 6) by a rise of Hct? a further fall of PACES, and a partial recovery of PACT; VI, however, was not changed further from its value on wk 0. On descent to sea level (SLJ, these variables all returned toward their original values. Small quantitative differences in these functions were apparent between the highlanders and the altitude-acclimatized

lowlanders, both at 150 and 4,360 m. Cold exposure (not shown in Table 3) had a variable effect on these functions. Thus, at sea level, both before and during altitude acclimatization, it caused modest falls of PA~)~and Sat),, but rises of PA COnand HR. At altitude, by contrast, it induced on all weeks increases of PA(), and Sao,, and decreases of PA coZ and HR. %h increased in 10°C relative to its value in 26°C but the magnitude of this increase was significantly smaller on arrival at altitude than on later weeks. Hct was not sigificantly affected under any of these conditions. Subjective responses. Upon arrival at altitude, the lowlanders reported lightheadedness and mild headaches in 26°C; these became more severe following induction of cold exposure. These symptoms, however, were not experienced again during the subsequent 6 wk at 4,360 m. The subjects complained throughout their stay at altitude that they felt considerably colder there during exposure to 10°C than they had at sea level; they stated sensing no appreciable relief in this regard as residence continued. To the contrary, they claimed upon return to sea level that 10°C now felt to them markedly warmer than at any time previously. As stated previously, the highlanders also reported that 10°C felt less cold at sea level than at their resident altitude. DISCUSSION

Three observations in this study are of particular interest. The first is the finding that the cold-induced increases in J?oZ of both the lowlanders and the highlanders were reduced at altitude as compared to those at sea level. The second is the fact that the shivering intensity of the lowlanders was not diminished at altitude as compared to that at sea level despite the observed decrease in their 00, at altitude, while that of

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

854

C. M.

TABLE 3. Cardiorespiratory functions of six altitude-ac’climatizing and five highlanders measured in 26°C at sea level and altitude Week

Lowlanders 150 m 4,360 m 4,360 m 150 m Highlanders 4,360 m 150 m

SL, Wk 0 Wk 6 SLf

‘7 l!min\: IhTPS

Hct, %

48.3 49.3 56.5 55.2

+ 2 + in

1.0 1.3b 1.1 1.2’l

54.7 t 1.8 53.8 2 1.7

37.8 30.6 26.7 31.3

11.29 11.76

30.4 5 o.4csf 32.3 t 0.7

+ 0.40 t 0.45”

+ + -e IL

1.9” OAb 0.6’ 0.6”

AND

L. 0.

LUTHERER

lowlanders _____-

PAc.O*, Torr

9.61 + 0.50” 11.91 + 0.38 11.97 t 0.56’ 10.20 + 0.60

BLATTEIS

Ph*, Torr

102.7 43.1 47.2 106.8

+ t k Ifi

%v %

2.1;’ 1.2” 0.8’ 2.1

43.8 2 2.P’ 104.2 + 2.1

96.9 85.4 88.7 95.4

2 + 2 k

HR, beatslmin

0.3” 2.3 4.1’ 1.0

82.3 2 1.3”’ 94.6 +_ 1.0

56 74 78 57

+ zk 2 ”

1” 8 5’ 3

65 2 2’s’ 44 k 2K

Values are means + SE. SLi = initial measurements at sea level; wk 0 = measurements on arrival at altitude; wk 6 = measurements after 6-wk residence at altitude; SLf = final measurements at sea level, upon return from altitude; Hct = hematocrit; VI = pulmonary ventilation. Other symbols as described in Table 2. The highlanders data are the same as in Table 2; they are shown again here to illustrate the relevant differences. Superior letters indicate significant differences among the lowlanders between: “SLi and wk 0; ‘)wk 0 and wk 6; ‘wk 6 and SL,; “SLi and SL,; among the highlanders between: ‘altitude and sea level; ‘at altitude between highlanders and lowlanders in wk 6; Rat sea level between highlanders and lowlanders in SL,.

the highlanders was lower at sea level than at altitude despite their higher Vo, at sea level. The third is that the altitude-induced reduction of the Vo, response to cold of the lowlanders was not reversed during the course of their 6-wk residence at 4,360 m. The latter two results, to our knowledge, have not been reported previously. The present finding of a hypoxic reduction of the calorigenic response to cold of man is in agreement with those of several previous workers (19, 33, 39), but in disagreement with those of some others (1, 7, 11, 23, 40, 58). However, these conflicting results probably may be ascribed to the different environmental conditions, experimental subjects, and protocols used in past studies. In the present study, by contrast, the physical and anthropological characteristics of the subjects were more closely matched than heretofore, the responses to cold were related to base-line values obtained on the same subjects during a control period at room temperature immediately preceding the cold exposure, and all the subjects, including the highlanders, were tested both at sea level and at altitude. This design allowed a more inclusive evaluation of the effect of altitude on man’s response to cold than had been possible previously. The occurrence at altitude of a reduction, as compared to sea level, of the cold-induced increase in Vo, of these lowlanders without a concomitant diminution of their shivering activity would suggest that the component of their thermogenic response to cold which was depressed by altitude exposure was independent of visible shivering, viz., it may have been NST. While the susceptibility of NST and the relative resistance of shivering to hypoxic suppression have been demonstrated previously in small and neonatal species (2, 3, 21, 56, 57), well known to possessNST, this is a surprising finding for man because the very existence of NST in adult man, and particularly in non-cold-acclimated man, has not yet been established; but, on the other hand, it also has not yet been ruled out. Thus, Heldmaier (31) has demonstrated that the amount of nonshivering thermogenic capacity is inversely proportional to the body weight of a species. Accordingly, a large species such as man would be expected to possessonly traces of NST, with shiver-

ing, therefore, the dominant thermogenic mechanism upon cold exposure. Indeed, the occurrence of shivering as a consistent event in the thermoregulatory response of man to cold has been shown repeatedly (32). However, a strict correlation between shivering and total metabolism in the cold has not been an invariable finding in any species (16-18, 32, 36). For instance, as for man, significant increases in oxygen consumption frequently occur without concomitant visible shivering on initial exposure to severe cold (16), during extended exposure to a moderately cool (12, 14, 34, 37, 59, or over the course of several weeks’ exposure to a cold (16-18) environment. Such results would implicate the participation under those conditions of heat-producing mechanisms which do not involve muscular contractions, i.e., NST. The possible existence of regulatory NST in man is further suggested by the reported increase in norepinephrine calorigenesis following, as compared to preceding, acclimatization to cold (8, 38). Instudies analogical to the present one, the rise in Vo, induced by exposure to cold was significantly reduced by moderate hypoxia in dogs (5) and in miniature pigs (4), two other large species thought to lack significant amounts of NST, also without any abatement of the shivering intensity. Additionally, the calorigenic action of infused norepinephrine was abolished in both dogs (6) and minipigs (4) when breathing lo-12% 0,. These results therefore were interpreted as indicating the presence in these species of a small, but significant, nonshivering, norepinephrine-mediated, thermogenic component of the response to cold, which is especially vulnerable to hypoxia and evidently not readily detectable by other means. Hence, on the same basis, it might be speculated that a small, but significant, hypoxia-sensitive, nonshivering thermogenic mechanism, which contributes along with shivering to cold thermogenesis, exists in man as well. According to the present data, this contribution could amount to as much as 21% of the total metabolism, a value within the range of those estimated by others (34, 37, 38). It should be noted, in this connection, that, as in other species (3), only the thermoregulatory component of the total Vo, of these men was affected by hypoxia, since at room temperature Vo, was the same or elevated at altitude as compared to its value at sea level, and in

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

MAN,

ALTITUDE,

AND

COLD

the cold it was not reduced below its basal rate. The finding that the cold-induced increases in the voZ of both groups of highlanders were significantly greater at altitude than those of the lowlanders might indicate that some recovery from the hypoxic depression of the metabolic response to cold occurs with prolonged exposure to altitude. Indeed, such an adaptation would be compatible with other functional adjustments to the stress of altitude (13, 25). However, although the lowlanders appropriately acclimatized to altitude during their 6-wk residence at 4,360 m (13, 25), the altitudeinduced reduction of their calorigenic response to cold was not reversed during the course of this process. On the other hand, their visible shivering intensity diminished as their exposure to altitude continued, though their total Vo, in the cold was not further decreased. It might be assumed, therefore, that NST replaced the reduced shivering thermogenesis, i.e., some recovery of NST may indeed have occurred during the 6 wk at altitude. The amount of this recovery, however, was very small, since on return to sea level the Vo, in the cold of these lowlanders was again increased to a rate approximating that at sea level originally. And, importantly, it was raised now without a concomitant increase in shivering activity; in fact, shivering was less intense under these conditions than under any others previously. Similarly, the Vo, of the highlanders in the cold was greater at sea level than at their resident altitudes, although their shivering activity was less intense. Hence, the persistence of shivering in the cold at altitude, and the continued, reduced Tjo, even upon extended residence would lend further support to the view that moderate hypoxic hypoxia preferentially may suppress NST in man as it does in other species (2-6, 19, 21, 56, 57). Furthermore, the immediate reversibility of this depression on descent to sea level would suggest that this effect of altitude is essentially nonadaptive. The absence of recovery from the altitude-induced depression of the calorigenic response to cold in these altitude-acclimatized lowlanders is at variance with the results of Lintzel (41) in rats and of Dikshit et al. (20) in man. However, in both those studies, the subjects were exposed to cold and al titude simultaneously. By contrast, in the present study, since the object was to investigate only the effect of exposure to chronic hypoxia on the thermoregulatory response to cold, these subjects were allowed to dress in accordance with their perception of the ambient thermal conditions. They were, therefore, cold-exposed only during the tests, each test thus representing an acute exposure to cold. Hence, it would appear that the recovery of Vo, in the cold at altitude reported by these other workers (20, 41) is not associated with altitude acclimatization per se, but rather with cold acclimation, or perhaps, in man, also with habituation (27, 28, 35). It is noteworthy, in support of this interpretation, that the calorigenic response to 10°C of the altitude-acclimatized lowlanders was significantly less upon their return to sea level than before their ascent to 4,360 m, and less also than that of the highlanders at sea level, when the latter were cold exposed only for the second time. Several investigators have reported small but signifi-

855 cant increases in the resting oxygen consumption of lowlanders newly come to altitude (10, 15, 26, 29) and of resident highlanders (45, 46, 50), as compared to their VoZ at sea level. The cause of this difference has not yet been elucidated. The occurrence of a slight elevation of the resting metabolic rate of the present lowlanders upon arrival at altitude is consistent with those previous observations. Also in agreement with previous findings (26), this increase was reversed during the succeeding 6 wk at altitude. Indeed, irO, fell under these conditions to a value below that at sea level originally and, moreover, remained at this lowered level upon return to 150 m. Although the basis for these changes was not investigated specifically in this study, it might be speculated that the subjective discomfort and agitation experienced by these subjects on arrival at altitude and their habituation to the experimental conditions subsequently could have accounted for these differences, as already suggested by others (15). The observation that the Tsk)s of both the newly arrived lowlanders and the highlanders stabilized in the cold at higher levels at each successive altitude than at sea level has been made previously (1, 14, 19, 23, 24, 30, 42-46, 49). However, a novel finding in this study was the increase in 26°C of the Tsk’s of the lowlanders during their 6-wk stay at 4,360 m. A consequence of this adjustment was that their warmer Tsk)S in 10°C were due, on arrival at altitude, to a smaller fall from their level in 26°C and after 6 wk to as great a fall as previously at sea level, but from higher initial values. Furthermore, the Tsk)S of both the lowlanders and the highlanders continued elevated in 26OC upon their descent to sea level; but they decreased in 10°C to values lower than previously at altitude. The mechanisms underlying these effects cannot be ascertained from the present data. But it might be speculated that the persistence of a high Tsk in 26°C at sea level after 6 wk at altitude reflected a physiological adaptation of the cutaneous circulation of the lowlanders to chronic hypoxia. Thus, it is possible that the cutaneous vasoconstriction which occurs on acute exposure to hypoxia (22), and which may have been reflected in the slightly lower Tsk’s of these subjects on arrival at altitude, was superceded upon their continued residence there by vasodilatation. The reservoir of blood stored in this circulation, therefore, would be available for redistribution as needed, That cold-induced vasocone-g* 7 upon cold exposure. striction proceeded normally in these altitude-acclimatized lowlanders at 4,360 m is suggested by the fact that the degree of fall of their Tsk was as great after 6 wk at this site as at sea level originally. Hence, the finding that the Tsh)S of both the lowlanders and the highlanders stabilized at altitude at higher values than at sea level (originally and finally) might be attributed not to a physiological but to a~ physical difference between the two environments. This difference may have been the effect of the lowered barometric pressure at altitude on convective heat exchange (47). Indeed, total heat loss may have been reduced by this effect, despite the greater skin-to-environment thermal gradient at equilibrium in the cold at altitude than at sea level, thereby compensating for the diminished heat production and

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

856 helping to maintain internal body temperature. The tendency of the T,, of all the present subjects to decrease, similarly under all altitudinal conditions, probably was due to their apparent leanness and small body size (11). It is noteworthy, in this context, that the fall in T,, began when both Tsk and Vo, became stabilized. This would imply that the drive in these subjects for increased heat production in the cold may have been related to the lowering of their skin tern .perature rather than to that of their core temperature. It is unlike ly that the reduced Vo, of these subjects was due, on the other hand, to a smaller feedback from their now warmer skin, since shivering intensity continued unchanged or increased under these conditions. Also, the breathing of an hypoxic mixture at sea level causes a decrease in cold-induced calorigenesis without raising the stabilization level of skin temperature (3-5). The respiratory characteristics at 26OC of both the lowlanders exposed to the two altitudes and of the highlanders from 4,360 m at their resident altitude and upon descent to sea level were in general conformity with previous observations under similar conditions (13). By contrast, the ventilatory responses of the highlanders from 3,330 m appeared exaggerated both at their habitual altitude and at sea level. No explanation is apparent for these observed discrepancies. It is possible that the hypoxic sensitivity of this group was untypically high (53, 54), as reflected by the large reduction of their pulmonary ventilation on exposure to sea level. The increased discomfort n the cold reported by these lo wlanders at altitude ha been reported previ .ously (19). One might speculate, in this connection, that cold could be an added, precipitating factor in the development of mountain sickness since, in this study, the symptoms of acute mountain sickness (“soroche agudo”) appeared to be related to the suddenly added stress of cold exposure. It is interesting that all the subjects in this study reported that their subjective impression of 10°C consistently was more intense at altitude than at sea level despite their warmer skin temperature at altitude. This could imply that thermal sensations might be influenced by psychological factors in addition to skin and core temperatures such as, possibly, visual reinforcement (“expectation”) from the “cold,” bleak, and inhospitable surroundings of these field, as contrasted with laboratory, conditions (30). In conclusion, these results demonstrate that altitude exposure reduces the calorigenic response of man to cold, apparently as in other species, by selectively depressing NST. This effect is not reversed by acclimatization to altitude. However, cold-induced vo, reverts to its higher rate immediately upon descent to sea level. ADDENDUM The respiratory volume meter used in these studies was a turbinetype flowmeter. The body of the meter was a thick-walled, Plexiglas cylinder through which a series of holes had been drilled tangentially to the inner surface. During inspiration, air entered the meter

C. M.

BLATTEIS

AND

L. 0.

LUTHERER

through these intake ports and imparted a force perpendicular to the impeller blades, resulting in an angular displacement of the impeller shaft. The rotations of the shaft were reduced 3,600:1 by a gear train and displayed on a counter. Although the meter was unidirectional, its exhaust ports had been provided with a neoprene valve in order to prevent the accumulation of condensate, which would occur if the expired air were allowed to reenter the impeller chamber. The sensitivity of the meter was primarily a function of the design of the intake ports, since the momentum of the inspired air was directly proportional to its instantaneous velocity and had to exceed the starting torque of the impeller. It was determined experimentally that 20 tangential jets, l/8 in. in diameter, would be required to limit the back pressure to 1 in. of water at 85 l/min of gas flow, thus forcing a compromise which rendered the meter insensitive to very low velocity air movement. At air movements above the stall point, however, the meter responded in a linear fashion. A provision was included in the design of the flange to reduce the total cross-sectional area of the intake ports by one-half, to accommodate the respiratory minute volume range of sedentary subjects. Calibration of each respiratory volume meter was accomplished dynamically, by withdrawing air from a Tissot respirometer through the meter by means of a piston pump which simulated normal breathing airflow patterns, then correlating the meter readings with calibrated Tissot volumes reduced to standard conditions. In addition to this dynamic calibration, a static airflow test was performed to permit a measure of back pressure and to determine the linearity of the response above the stall point. It was noted that the calibration was independent of the cyclic rate of airflow, such as occurs with each breath, within a 2% limit of experimental error. These respiratory volume meters were calibrated at ambient temperatures ranging from - 12 to +24”C. No trend of variation was evident, and all random variations were within the 2% limit of experimental error. The calibration was repeated in an altitude chamber at ambient pressures ranging from 760 to 420 Torr. Although calibration was maintained over a considerable range of ambient temperatures, which in fact represented a small change in gas density, the turbine efficiency of the meter was low enough to require compensation for large changes in gas density, as at altitude. Conversion factors for reduction of altitude observations to standard conditions were therefore derived from the resulting data. An additional correction was required for the reduction of experimental values at altitude to accommodate the mass of water vapor in the inspired air.

The skilled technical assistance of Dr. H. R. Scholnick, Mr. F. Mastroianni, and Mr. L. Sirvio is gratefully acknowledged. The specialized contributions of Mr. Frank Botsch (mass flowmeter) and Walter Lafferty (statistics) also are very much appreciated. We also are greatly indebted to the Peruvian Army Medical Service, which very generously made available to us the sea-level subjects, various support personnel, and facilities and equipment at the Hospital Militar Central in Lima; and to the Cerro Corporation, which graciously allowed us the use of its, then, space in Rio Blanco, Casapalca, Pachacayo, and Cochas. A portion of this work was supported by Contract DADA 17-68-C8136 from the US Army Medical Research and Development Command to the University of Tennessee Center for the Health Sciences Biometric Computer Center. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. This manuscript has been cleared per provisions of AR 3605, OTSG Circ. 360-3, and USARIEM Memo 360-l. Present addresses: C. M. Blatteis, Dept. of Physiology and Biophysics, University of Tennessee Center for the Health Sciences, Memphis, Tenn. 38163; L. 0. Lutherer, Dept. of Physiology, Texas Tech University School of Medicine, Lubbock, Texas 79409. Received

for publication

24 March

1975.

REFERENCES 1. BAKER, P. T., E. R. BUSKIRK, J. KOLLIAS, AND R. B. MAZESS. Temperature regulation at high altitudes: Quechua Indians and U.S. Whites during total body cold exposure. Human BioZ. 39:

155-169, 2. BLATTEIS, newborn

1967. C. M. Hypoxia and the metabolic response to cold rabbits. J. Physiol., London 172: 358-368, 1964.

in

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

MAN,

ALTITUDE,

AND

COLD

3. BLATTEIS, C. M. Shivering and nonshivering thermogenesis during hypoxia. In: Proc. Intern. Symp. Environ. Physiol., Bioenergetics, edited by R. Em. Smith. Bethesda, Md.: Fed. Am. Sot. Exptl. Biol., 1972, p. 151-160. 4. BLATTEIS, C. M., AND T. M. GILBERT. Hypoxia and shivering thermogenesis in cold-acclimatized miniature pigs. J. Appl. PhysioZ. 36: 453-456, 1974. 5. BLATTEIS, C. M., AND L. 0. LUTHERER. Cold-induced thermogenesis in dogs: its reduction by moderate hypoxia. J. Appl. Physiol. 35: 608-612, 1973. 6. BLATTEIS, C. M., AND L. 0. LUTHERER. Reduction by moderate hypoxia of the calorigenic action of catecholamines in dogs. J. Appl. PhysioZ. 36: 337-339, 1974. 7. BROWN, A. L. J., G. F. VAWTER, AND J. P. MARBARGER. Temperature changes in human subjects during exposure to lowered oxygen tension in a cool environment. J. Aviation Med. 23: 456463, 1962. 8. BUDD, G. M., AND N. WARHAFT. Cardiovascular and metabolic responses to noradrenaline in man, before and after acclimatization to cold in Antarctica. J. PhysioZ., London 186: 233-242,1966. 9. BULLARD, R. W. Effect of hypoxia on shivering in man. J. Aerospace Med. 32: 1143-1147, 1961. 10. BURRUS, S. K., D. B. Dill, D. L. BURK, D. V. FREELAND, AND W. C. ADAMS. Observations at sea level and altitude on basal metabolic rate and related cardiopulmonary functions. Human Biol. 46: 677-692, 1974. 11. BUSKIRK, E. R., AND J. KOLLIAS. Total body metabolism in the cold. In: Influence of Cold on Metabolic Regulation, edited by E. M. Zipf. New Brunswick, N. J.: N. J. Acad. Sci., 1969, Special Symp. Issue, p. 17-25. 12. CANNON, W. B., A. QUERIDO, S. W. BRITTON, AND E. M. BRIGHT. Studies on the conditions of activity in endocrine glands. XXI. The role of adrenal secretion in the chemical control of body temperature. Am. J. PhysioZ. 79: 466-517, 1927. 13. CHIODI, H. Respiratory adaptation to high altitude. In: The Regulation of Human Respiration, edited by D. J. C. Cunningham and B. B. Lloyd. Oxford: Blackwell, 1963, p. 363-378. 14. CIPRIANO, L. F., AND R. F. GOLDMAN. Thermal responses of unclothed men exposed to both cold temperatures and high altitudes. J. Appl. PhysioZ. 39: 796-800, 1975. 15. CONSOLAZIO, C. F., R. A. NELSON, L. 0. MATOUSH, AND J. E. HANSEN. Energy metabolism at high altitude (3,475 m). J. AppZ. PhysioZ. 21: 1732-1740, 1966. 16. DAVIS, T. R. A. Chamber cold-acclimatization in man. J. AppZ. PhysioZ. 16: 1011-1015, 1961. 17. DAVIS, T. R. A. Nonshivering thermogenesis. Federation Proc. 22: 777-782, 1963. 18. DAVIS, T. R. A., AND D. R. JOHNSON. Seasonal acclimatization to cold in man. J. AppZ. PhysioZ. 16: 231-234, 1961. 19. DAVIS, T. R. A., H. S. NAYAR, K. C. SINHA, S. D. NISBITH, AND R. M. RAI. The effect of altitude on the cold responses of low altitude acclimatized Jats, high altitude acclimatized Jats and Tibetans. In: Biometerology II, edited by S. W. Tromp and W. H. Weihe. Oxford: Pergamon, 1967, p. 191-198. 20. DIKSHIT, P. K., B. K. ANAND, G. S. CHHINA, B. SINGH, AND J. S. BAJAJ. Metabolic correlates of cold acclimatization at sea level and high altitude (Abstract). Proc. Intern. Union Physiol. Sci. 11: 401, 1974. 21. DONHOFFER, S. Nonshivering thermogenesis: past and present. In: Nonshivering Thermogenesis, edited by L. Jansky. Amsterdam: Swets & Zeitlinger, 1971, p. 11-26. 22. DURAND, J., AND J. P. MARTINEAUD. Resistance and capacitance vessels of the skin in permanent and temporary residents at high altitude. In: High AZtitude Physiology: Cardiac and Respiratory Aspects, edited by R. Porter and J. Knight. London: Churchill Livingstone, 1971, p. 159-167. 23. EL~NER, R. W. Comparison of Australian Aborigines, Alacuf Indians, and Andean Indians. Federation Proc. 22: 840-842, 1963. 24. ELSNER, R. W., AND A. BOUTAD. ThermaL and MetaboZic Responses to Cold of Peruvian Indians Native to High Altitude. Ft. Wainwright, Alaska: Arctic Aeromed. Lab., 1963, Rept. AALTDR-62-64, p. l-25. 25. FRISANCHO, A. R. Functional adaptation to high altitude hypoxia. Science 187: 313-319, 1975. 26. GILL, M. B., AND L. G. C. E. PUGH. Basal metabolism and respiration in men living at 5,800 m (19,000 ft.). J. Appl. Physiol.

857 19: 949-954, 1964. 27. GLASER, E. M. Central nervous mechanisms of adaptation. In: Biometerology II, edited by S. W. Tromp and W. H. Weihe. Oxford: Pergamon, 1967, p. 251-257. 28. GLASER, E. M., M. S. HALL, AND G. C. ALBRITTON. Habituation to heating and cooling of the same hand. J. PhysioZ., London 146: 152-164, 1959. 29. GROVER, R. F. Basal oxygen uptake of man at high altitude. J. Appl. PhysioZ. 18: 909-912, 1963. 30. HANNA, J. M. A comparison of laboratory and field studies of cold responses. Am. J. Phys. Anthropol. 32: 227-232, 1970. 3 1. HELDMAIER, G. Relationship between nonshivering thermogenesis and body size. In: Nonshivering Thermogenesis, edited by L. Jansky. Amsterdam: Swets & Zeitlinger, 1971, p. 73-80. 32. HEMINGWAY, A. Shivering. Physiol. Rev. 43: 397-422, 1963. 33. H~LNHAGEN, 0. Uber Stiirungen der Warmerregulation im Akuten O,-Mange1 bei . Kaltebelastung. Luftfuhrtmed. 9: 16-25, 1944. 34. IAMPIETRO, P. F., J. A. VAUGHAN, R. F. GOLDMAN, M. B. KREIDER, F. MASUCCI, AND D. E. BASS. Heat production from shivering. J. Appl. PhysioZ. 15: 632-633, 1960. 35. ITQH, S. PhysioZogy of Cold-Adapted Man. Sapporo: Buneido, 1974, p. 55-63. 36. JANSK~, L. Nonshivering thermogenesis and its thermoregulatory significance. BioZ. Rev. 48: 85-132, 1973. 37. JOHANSSON, J. E. fiber den Einfluss der Temperatur in der Umgebung auf die Kohlensaure Abgabe des menschlichen Korpers. Skand. Arch. Physiol. 7: 122-177, 1897. 38. JOY, R. J. T. Responses of cold-acclimatized men to infused norepinephrine. J. AppZ. PhysioZ. 18: 1209-1212, 1963. 39. KOTTKE, F. J., J. S. PHALEN, C. B. TAYLOR, M. B. VISSCHER, AND G. T. EVANS. Effect of hypoxia upon temperature regulation of mice, dogs, and man. Am. J. PhysioZ. 153: 10-15, 1948. 40. LIM, T. P. K., AND U. C. LUFT. InfZuence of Hypoxia on Therm& Homeostasis in Man. Ft. Wainwright, Alaska: Arctic Aeromed. Lab., 1962, Rept. TDR-61-55. 41. LINTZEL, W. Uber die Wirkung der Luftverdunnung auf Tiere. Pfluegers Arch. 227: 693-708, 1931. 42. LITTLE, M. A. Temperature regulation at high altitude: Quechua Indians and U.S. Whites during foot exposure to cold water and cold air. Human BioZ. 41: 519-535, 1969. 43. LITTLE, M. A., R. B. THOMAS, AND J. W. LARRICK. Skin temperature and cold pressor responses of Andean Indians during hand immersion in water at 4°C. Human BioZ. 45: 643-662, 1973. 44. LITTLE, M. A., R. B. THOMAS, R. B. MAZESS, AND P. T. BAKER. Population differences and developmental changes in extremity temperature responses to cold among Andean Indians. Human BioZ. 43: 70-91, 1971. 45. MAZESS, R. B., AND R. LARSEN. Responses of Andean Highlanders to night cold. Intern. J. Biometeorol. 16: 181-192, 1972. 46. MAZESS, R. B., E. PI&N-REATEGUI, R. B. THOMAS, AND M. A. LITTLE. Oxygen intake and body temperature of basal and sleeping Andean natives at high altitude. Aerospace Med. 40: 6-9, 1969. 47. MCADAMS, W. H. Heat Transmission (3rd ed.). New York: McGraw, 1954, p. 319-324. 48. MITCHELL, D., AND C. H. WYNDHAM. Comparison of weighting formulas for calculating mean skin temperature. J. AppZ. PhysioZ. 26: 616-622, 1969. 49. NAIR, C. S., AND S. GEORGE. The effect of altitude and cold on body temperature during acclimatization of man to 3,300 m. Intern. J. Biometeorol. 16: 79-84, 1972. 50. PI&N-REP~TEGUI, E. Basal metabolic rate and body composition at high altitudes. J. Appl. PhysioZ. 16: 431-434, 1961. 51. RAHN, H., AND A. B. OTIS. Continuous analysis of alveolar gas composition during work, hyperpnea, hypercapnia, and anoxia. J. Appl. Physiol. 1: 717-724, 1949. 52. SIEGEL, S. Non-Parametric Statistics for the Behavioral Sciences. New York: McGraw, 1956, p. 166-172. 53. ~RENSEN, S. C., AND J. W. SEVERINGHAUS. Respiratory sensitivity to acute hypoxia in man born at sea level living at high altitude. J. Appl. Physiol. 25: 211-216, 1968. 54. SORENSEN, S. C., AND J. W. SEVERINGHAUS. Irreversible respiratory insensitivity to acute hypoxia in man born at high altitude. J. Appl. Physiol. 25: 217-220, 1968. 55. SWIFT, R. W. The effects of low environmental temperature upon

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.

858 metabolism. II. The influence of shivering, subcutaneous fat, and skin temperature upon heat production. J. Nub. 5: 227-249, 1932. 56. SZEGVARI, G., J. VARNAI, AND S. DONHOFFER. The effect of hypoxia on colonic and muscle temperature in the rat: the localization of thermoregulatory heat production. Acta Physid. Acad. Sci. Hung. 19: 287-294, 1961. 57. SZEGVARI, G., J. VARNAI, AND S. DONHOFFER. The effect of

C. M.

58.

BLATTEIS

AND

L. 0.

LUTHERER

environmental temperature, hypoxia and hypercapnia on total heat production and the electrical activity of muscle in the rat: shivering and nonshivering thermogenesis and the site of nonshivering thermogenesis. Acta Physd. Acad Sci. Hung. 23: 4962, 1963. WEZLER, K., AND E. FRANK. Chemische Warmerregulation gegen Kalte und Hitze in Sauerstoffmangel. PfZuegers Arch. 250: 439-464, 1948.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (134.129.182.074) on November 21, 2018. Copyright © 1976 the American Physiological Society. All rights reserved.