Nasal heat exchange in the giraffe and other large mammals

Respiration Physiology (1979) 37, 325-333 © Elsevier/North-Holland Biomedical Press

NASAL HEAT EXCHANGE IN THE GIRAFFE AND OTHER LARGE MAMMALS*

V.A. L A N G M A N l, G . M . O . MALOIY 2, K. SCHMIDT-NIELSEN 3 and R.C. SCHROTER 4 ! The Biological Laboratories, Harvard University, Cambridge, MA 02138, U.S.A. 2 Department of Animal Physiology, University of Nairobi, Kenya, 3 Department of Zoology, Duke University, Durham, NC 27706, U.S.A., and 4 Physiological Flow Studies Unit, Imperial College, London SW7 2AZ, United Kingdom

Abstract. ]'he respiratory air of the giraffe is exhaled at temperatures substantially below body core temperature. As a consequence, the water content of the exhaled air is reduced to levels below that in pulmonary air, resulting in substantial reductions in respiratory water loss. Measurements under outdoor conditions showed that at an ambient air temperature of 24 °C, the exhaled air was 7 °C below body core temperature, and at ambient air temperature of 17°C, the exhaled air was 13°C below core temperature. The observations were extended to two additional species of wild and four species of domestic ungulates. All these animals exhaled air at temperatures below body core temperature. The average amount of water recovered due to cooling of the air during exhalation, calculated as per cent of the water loss that would occur if air were exhaled at body core temperature, amounted to between 24 and 58%, the average value for the giraffe being 56%. Airways Body temperature Expired gas temperature

Respiratory water loss Thermoregulation Ungulates

In air-breathing animals the inhaled air is heated to body temperature and saturated with water vapor before it arrives at the respiratory surfaces of the lungs. In man, air is exhaled fully saturated at near body temperature, and most of the heat and water added during inhalation is therefore lost (Walker et al., 1961). It was commonly thought that all mammals exhale air at or near body temperature and saturated with water vapor until Jackson and Schmidt-Nielsen (1964) reported Accepted for publication 19 March 1979 * Supported by grants from the National Geographic Society, the Dean's Committee, the University of Nairobi, The Welcome Trust, London, NIH Grant HL-02228, and NIH Research Career Award I-K6-GM-21,522. 325

326

V.A. LANGMAN et aL

that heat exchange in the nose of small rodents (kangaroo rat and white rat) enaLles these animals to exhale air at temperatures far below body temperature. The cooling of the exhaled air is simply explained; on inhalation the nasal surfaces are cooled by the incoming air, augmented by evaporation of water, and during exhalation warm air passes over these cool surfaces where it gives up heat and water vapor condenses. The effect on the water balance of a kangaroo rat is striking; more than 80% of the water vapor present in the saturated hmg air may be recovered, an amount of crucial importance to the water balance of a desert rodent that normally never drinks water (Schmidt-Nielsen et al., 1970). Effective heat exchange in the nasal passages depends on the area available for heat exchange, and particularly on the distance between the center of the air stream and the surfaces between which it passes. A model of the heat exchange developed by Collins et al. (1971) predicts recovery of water, resulting from heat exchange in the nasal passages, that is in accord with observations on mammals of small body size (Sehmid, 1976). However, neither measurements nor predictive models have been developed for animals of large body size. In humans nasal heat exchange and water recovery are of minor importance because of the much larger diameter of the passageways and the relatively small area available for heat exchange (Proctor et al., 1977). By implication this has led to the assumption that the situation would be similar in other large mammals. The problem has remained unexplored. Measurements of the respiratory water loss in camels (Langman et al., 1978) indicated that these animals lose less water during respiration than would be predicted from the assumption that the respiratory air is exhaled at or near body core temperature and saturated with water vapor. These measurements could be explained, however, if significant nasal heat exchange occurs in these large animals, thus'providing cooling of the exhaled air and recovery of water. We, therefore, decided to examine this problem in large mammals, to determine what mechanisms are involved, and what effects they have on water balance of animals that in nature frequently are exposed to arid conditions. After detailed information was obtained on giraffes, the investigation was extended to a number of other species of large mammals.

Materials and methods

Our study concentrated on the giraffe (Giraffa camelopardalis), a species native to the semi-arid and savannah country of East Africa. Two female and one male giraffe with an average body weight of 600 kg were held at a field station near Athi River, Kenya, by the Department of Animal Physiology, University of Nairobi. We also examined two additional species of large wild and four species of large domestic ungulates. These were two wildebeest (Connochaetes taurinus), three waterbuck (Kobus ellipsiprymus), two domestic sheep, one donkey, two goats,

NASAL HEAT EXCHANGE IN THE G I R A F F E

327

and two cows, which were located at the East African Veterinary Research Organization near Kikuyu, Kenya. All animals appeared to be healthy and in good physical condition and had been in captivity for at least two months. In order to determine respiratory heat exchange and the degree of cooling of exhaled air we measured ambient air temperature (TI), relative humidity (RH), rectal temperature (TB), and exhaled air temperature (TE). Exhaled air temperature was measured with rapidly responding, welded copperconstantan thermocouples made from wire of 0.125 mm diameter. All exhaled air temperatures were measured under natural outdoor conditions on moderately restrained unanesthetized animals. Measurements were not begun until the animal appeared calm and had a steady respiratory pattern. A short piece (about 6 cm) of plastic tube (12.5 mm diameter) was placed in one nostril, arranged in and parallel to the air flow and wedged against the dorsal opening of the naris. Air could pass freely through this tube, as well as through the other naris which was left undisturbed. The tube was provided with a sidetube to form a T, and the thermocouple was inserted through this sidetube and fixed in place with the tip extending into the center of the air stream. All respiratory air passed through the nose, and the mouth always remained closed. The thermocouple was connected to a Wescor TH-50 DC thermocouple thermometer and recorded on a strip chart recorder. Ambient air temperature and relative humidity were measured with a sling psychrometer, read to within + 0.1 °C. Rectal temperature was measured with a clinical thermometer to the nearest 0.2 °C. Calibration against standardized mercury thermometers indicated that errors did not exceed + 0.2 °C. In our calculations of water loss and water recovery, we assumed that air in the lungs of mammals is saturated with water vapor at body core temperature, and that the exhaled air is saturated with water vapor at the temperature at which it is exhaled. The first assumption has been amply documented in the literature and is generally accepted. That air on exhalation is saturated, rather than supersaturated or less than fully saturated, is theoretically valid, but to test the assumption we compared the water loss calculated from the measured exhaled air temperatures with a dew-point hygrometer and with directly measured water losses. The latter were measured gravimetrically by placing an air-tight mask over the nose and mouth of the animal with the inlet connected to a three-way Otis-McKerrow valve and the outlet to a Collins 'J' valve, which in turn was connected to a canister containing Drierite for water absorption. The amount of water absorbed during 25 breaths (2 to 5 min) was determined by weighing. A second absorbant canister placed at the outlet of the first was used to ensure that water absorption had been complete. The volume of expired air in a given time period was determined by collection in Douglas bags and measured on a Parkinson and Cowan volumeter. The effectiveness of nasal heat exchange was expressed in terms of the amount of'water recovered on exhalation. Water recovery was calculated as the percentage of the water that would be lost if the exhaled air were to leave at body temperature and saturated with water. It was calculated as the difference between the water

328

v.A. LANGMANet aL

content of the lung air (Ws) and the water content of the exhaled air (WE) divided by the water content of the lung air (WB) minus the water content of the ~ l e d air (Wl), multiplied by 100 (Langman et al., 1978). Percent recovery of respiratory water

WB

-- WE

We

- Wl

=

x 100

Effective heat exchange in the nasal passageways requires close contact between air and the surfaces over which it flows; important parameters are the area available for heat exchange, the length of the passages, and the rate of shear (effectively the linear velocity) over the exchange surface (Collins et al., 1971). The anatomical features of the nasal passages are therefore of importance, and appropriate dimensions were measured on heads sectioned sagittally of giraffes and compared with cattle. The left half of each specimen was cut in serial transverse sections approximately 25 mm thick, cranially from the entrance of the nares. Longitudinal and transverse dimensions were available from these sections.

Results The temperature of the exhaled air of the giraffes was measured under natural outdoor conditions with ambient air temperatures ranging from 16 to 24 °C (fig. 1). The exhaled air temperature was consistently below body core temperature. At an ambient air temperature of 24 °C the exhaled air was about 7 °C below core temperature and at ambient air temperature of 17 °C it was up to 13 °C below core temperature. Thus, as the ambient temperature increased, the exhaled air temperature also increased. There were no measurements in which the exhaled

.a.

3S. Tie*

*%

30.

'dl~" '=dll ~

•

2si. I. •

/

/

.

/~-

'~ 1 6 ~ /

~mbientAir Temperature"C Fig. I. Rectal temperature and temperature of the exhaled air of giraffe over a range of ambient air temperatures. Rectaltemperature,TB(0); expiredair temperature, TE ~ ) ; ambient air temperature, TI.

NASAL HEAT EXCHANGE IN THE G I R A F F E

329

w" s.-

•~

40

s._

.,,m,

e

"E

e

~

ee

W!

20 10

I

I

I0

I

I

I

20

I

30

Ambient Air Temperature °C

Fig. 2. Water content of air in the lungs and measured water content of exhaled air of g~ra~Te plotted against ambient air temperature at the time of measurement. Top line: linear regression of the water content of air in the lungs, WB a ) calculated from rectal temperature (Y - 0.28 Ti + 38.4; n = 35; r - 0.4361 ; P < 0.01). Lower line: linear regression of measured water content of exhaled air, WE (O), ( Y - 0 . 3 4 T l + 16.8; n - 3 5 ; r - 0 . 1 4 4 5 ; P > 0 . 1 ) . (Values indicated by open circles were obtained during measurements of exhaled air temperature, the upper points being calculated from Te and the lower points from TE. These points are not included in the calculated regression lines.)

air temperature was lower than the ambient (inhaled) air temperature, as has been observed in small rodents (Schmidt-Nielsen et al., 1970). The calculated water content of the exhaled air, assuming complete saturation, is compared with measured water losses in fig. 2. The calculated water content of exhaled air falls within the range of the directly measured values, although apparently with a bias towards the upper part of the range. However, statistical comparison of regression lines fitted to the data for water loss calculated from exhaled air temperatures and for directly measured water loss over the ambient air temperature range of 15 to 25 °C showed that the two regression lines were not significantly different. The exact exhaled air temperature could not be measured during the water collection periods, and the data thus represent different times of measurement. In conclusion, the estimated water content of the exhaled air agrees with direct determinations, but apparently with a bias that may slightly underestimate the average recovery of water in the nasal passageways. The dimensions of the passageways in which heat and water exchange and recovery take place can be estimated from fig. 3, which shows sagittal and transverse sections of the nasal turbinates of giraffe and cattle, two animals of approximately the same body mass (600 kg). This figure permits a comparison of the length and the transverse dimensions of the airways, and thus a comparison of the surface areas available for heat exchange. The distance from the apex of the premaxilla to the cribriform plate (marked B in fig. 3) was about 40 cm in the giraffe and 35 cm in the ox. The cross-sections

330

V.A. LANGMAN et aL I

/

!

I

1Ocm

/

O×

I

I

|

2 em

Fig. 3. Sagit~l sections (top) of heads of giraffe and ox. Cross sections (bottom) cut at the region of the third premolar (A). The crosshatched areas represent bone and cartilage. The dark areas on the cross sections represent airways. The distances from premaxilla to cribriform plate (B) and from cribriform plate to foramen occipitale magnum (C) provide useful indices for interspecies comparisons of the length of the heat exchanger and the braincase.

showed a much larger perimeter, and thus area available for heat exchange, in the giraffe than in the ox. The total perimeter (both sides combined) at section A in the giraffe was about 116 cm, while in the ox the total perimeter measured at the same place (the third premolar) was about 58 cm. Because no information is available on the rate of air flow over the various parts of this surface, the linear flow rate at any given point cannot be estimated. What is of importance, however, is that the distance from the center of the air stream to the adjoining wall in the giraffe is everywhere in the magnitude of about 2 ram. According to Collins et al. (1971) effective nasal heat and water exchange depends on small transverse and large axial dimensions of the airways. The giraffe, therefore, appears better equipped anatomically for nasal heat exchange and water recovery than cattle. This may be of importance under more extreme conditions than prevailed during our measurements. Measurements on exhaled air temperatures similar to those in the giraffe were carried out on two additional wild animals, waterbuck and wildebeest, and four domestic anim~ls, goat, cow, donkey, and sheep (table 1). In the giraffe and waterbuck as well as the goat and cow the mean exhaled air temperatures ranged from 8.3 to 9.9 °C below body core temperature. The degree of cooling of the exhaled air was less in donkey, sheep, and wildebeest, ranging from 3.7 to 5.3 °C below body temperature. The reason for these differences are, at the moment, unclear. The calculated recovery of water from the exhaled air (table 1, last column) was substantial, ranging from a low of 17% in the wildebeest to over 40% in the water-

NASAL HEAT EXCHANGE IN THE GIRAFFE

331

TABLE 1

Water recovery from exhaled air of seven species of ungulates. Ti = ambient air temperature, Ta = rectal temperature, TE = exhaled air temperature, Wt = water content of air saturated at Ta, WE = water content of air saturated at TE. All values are averages or calculated from averages Animal

Tl (°C)

Ta (°C)

TE (°C)

TB - TE (°C)

Wl (mg/l)

WB (mg/l)

WE (mg/l)

Recover

of water

(%) Goat 20 kg, n - 7

23.4

39.4

29.5

9.9

13.2

49.5

29.5

55

Waterbuck 100 kg, n - 3

22.4

38.4

28.6

9.8

13.7

47.2

27.9

58

21.1

37.3

28.0

9.3

13.3

44.6

27.2

56

21.1

38.8

30.5

8.3

12.1

48.1

31.2

47

23.6

37.6

32.3

5.3

13.6

45.2

34.4

34

24.1

39.0

35.3

3.7

13.8

48.6

40.2

24

23.9

39.1

35.4

3.7

13.9

48.8

40.5

24

Giraffe 600 kg, n = 15

Cow 200 kg, n = 10

Donkey 170 kg, n = 4

Sheep 23 kg, n - 8

Wildebeest 96 kg, n - 9

buck. Recovery is directly correlated with the degree of cooling of the exhaled air, but not strictly proportional because the body core temperature is not identical in all of the animals. It is possible that the mechanism for water recovery, if measured in completely undisturbed animals, would be at least as effective in the wildebeest, a dry steppe and grassland arrimal, as in the waterbuck, which is found in moist habitats near open water.

Discussion The measurements in the giraffe show that the respiratory air during exhalation loses considerable amounts of heat on its passage over the respiratory surfaces (fig. 1). In the process its water content is greatly reduced (fig. 2), resulting in substantial reductions in respiratory water loss. These results extend the importance of nasal heat and water recovery to animals of large body size, previously shown to be essential in the water balance of e.g. desert rodents (Schmidt-Nielsen et al., 1970). All the other large mammals studied also exhaled air at temperatures substantially below body core temperature, although there were considerable differences between the different species. The high degree of effectiveness of heat exchange and water recovery in the giraffe appears to result from the long axial dimension, the narrow

332

V.A. L A N G M A N et al.

transverse dimension, and the large surface available for heat exchange in the nasal region. The high degree of heat exchange in goat and waterbuck indicates that these animals also possess nasal passages that meet the anatomical requirements for effective heat transfer between airstream and nasal mucosa. However, the reason that some of the animals studied showed a much lower degree of cooling and water recovery (table 1) cannot be construed as solely due to anatomical differences; one reason may be that the experimental conditions (e.g. excitement) cause an increased blood flow to the nasal mucosa, thereby overcoming the heat recovery mechanism. For example, the results obtained by Murrish (1973) suggest that changes in the blood flow to the mucosal lining of the nasal passages in penguins can drastically change the temperature of the exhaled air and thus water and heat losses. Is the nasal heat exchange in the giraffe important to the water economy of the animal? To answer this question we estimated the total respiratory water loss, using the respiratory minute volume and its measured water content. By subtracting the water content of the inspired air, the net water loss via respiration was calculated, giving the results shown i~ fig. 4. In fig. 4 the measured rate of net respiratory water loss (WE -- Wx) is compared to the water loss that would occur if the air were exhaled saturated at body temperature ( W B - WI). There was a direct and significant relationship between •

|1

i

|

e

I

•

u

!

II B

" ,;2,

25

m

• aa / a

2.0

II •~ 1.5

• • . _ / a q.aan m" •a

•

:.

o.s

"~,-'-: •

~"~.

--

t '

• .

I

I

I

10 15 20 2s AmbientlirYemperature "C

I

30

~lS

Fig. 4. Rate of net respiratory water loss (WE - "~Vn) compared with the net loss that would occur if air were exhaled at body temperature ( W n - Wx). Both lines are linear regression lines, calculated using the least squares method. Net respiratory water loss: W~- - Wx -- 0.047 T! - 0 . 3 2 ; n - 45; r - 0.4078; P < 0.01. Net respiratory water loss estimated at body temperature: "~s - "~x ffi 0.077 TI + 0.041; n ffi 45; r ffi 0.4180; P < 0.01.

NASAL HEAT EXCHANGE IN THE GIRAFFE

333

the ambient air temperature and the net water loss per unit time (P < 0.01). The increase in respiratory water loss with increasing ambient air temperature reflects the increase in respiratory rate, which between 13 °C and 29 °C increased from 6.5 to 13.4 breaths per minute. The savings of water by the giraffe due to heat exchange in the nasal region increased i.6-fold over this range of ambient air temperatures. The amount of water recovered in the n~sal heat exchanger of the giraffe can reduce water loss by 1.5 liter to 3 liter of water per day, a saving of 11% to 21% of the total daily water loss in this animal (Langman, unpublished). Therefore, the recovery of respiratory water is a major and important consequence of the nasal heat exchange in the giraffe. The cooling of the exhaled air, and especially the recovery of some of the water contained in the lung air, also entails the recovery of a certain fraction of the heat added to the inhaled air (Schmidt-Nielsen et al., 1970). The heat recovery on exhalation in the giraffe averaged 8% of the metabolic heat production, and under cool conditions this may be of importance to the animal. Under hot conditions the recovery system can be bypassed by increasing the circulation to the nasal mucosa, and under such conditions the added water vapor content of the exhaled air contributes to heat dissipation. It is apparent from the data presented in this paper that nasal heat exchange is found, not only in small animals as reported earlier, but also in a range of large mammals. It represents an important factor in water and heat conservation. The operation of nasal heat exchange requires no metabolic energy, yet provides significant benefits, a factor which would seem to favor its evolutionary development.

References Collins, J.C., T.C. Pilkington and K. Schmidt-Nielsen (1971). A model of the respiratory heat transfer in a small mammal. Biophys. J. 11 : 886-9!4. Jackson, D.C. and K. Schmidt-Nielsen (1964). Counter-current heat exchange in the respiratory passages. Proc. Natl. dcad. Sci. 51:1192-1197. Langman, V.A., G. M.O. Maloiy, K. Schmidt-Nielsen and R.C. Schroter (1978). Respiratory water and heat loss in camels subjected to dehydration. J. Physiol. (London) 278: 35P. Murrish, D. E. (1973). Respiratory heat and water exchange in penguins. Respir. Physiol. 19: 262-270. Proctor, D.F., I. Andersen and G.R. Lundqvist (1977). Human nasal mucosai function at controlled temperatures. Respir. Physiol. 30: 109-124. Schmid, W.D. (1976). Temperature gradients in the nasal passage of some small mammals. Comp. Biochem. Physiol. 54A: 305-308. Schmidt-Nielsen, K., F.R. Hainsworth and D.E. Murrish (1970). Counter-current heat exchange in the respiratory passages: effect on water and heat balance. Respir. Physiol. 9: 263--276. Walker, J. E. C., R.E. Wells, Jr. and E.W. Merrill (1961). Heat and water exchange in the respiratory tract. Am. J. Med. 30: 259-267.