Journal of Applied Physiology Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 85: 318-325, 1998;
8750-7587/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Entin, P. L.
Right arrow Articles by Rawson, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Entin, P. L.
Right arrow Articles by Rawson, R. E.
Vol. 85, Issue 1, 318-325, July 1998

Thermal drive contributes to hyperventilation during exercise in sheep

Pauline L. Entin, David Robertshaw, and Richard E. Rawson

Department of Physiology, Cornell University, Ithaca, New York 14853

    ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References

The etiology of exercise hypocapnia is unknown. The contributions of exercise intensity (ExInt), lactic acid, environmental temperature, rectal temperature (Tre), and physical conditioning to the variance in arterial CO2 tension (PaCO2) in the exercising sheep were quantified. We hypothesized that thermal drive contributes to hyperventilation. Four unshorn sheep were exercised at ~30, 50, and 70% of maximal O2 consumption for 30 min, or until exhaustion, both before and after 5 wk of physical conditioning. In addition, two of the sheep were shorn and exercised at each intensity in a cold (<15°C) environment. Tre and O2 consumption were measured continuously. Lactic acid and PaCO2 were measured at 5- to 10-min intervals. Data were analyzed by multiple regression on PaCO2. During exercise, Tre rose and PaCO2 fell, except at the lowest ExInt in the cold environment. Tre explained 77% of the variance in PaCO2, and ExInt explained 5%. All other variables were insignificant. We conclude that, in sheep, thermal drive contributes to hyperventilation during exercise.

hypocapnia; respiratory alkalosis; blood gases; hyperthermia; thermoregulation; respiratory control

    INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References

HYPOCAPNIA during exercise has been observed in species spanning three classes of vertebrates, including mammals (2, 3, 9, 11, 15, 21, 22, 24, 25, 28, 31), birds (5, 18), and reptiles (20). The hypocapnia is characteristic of not only high-intensity exercise but of moderate-intensity exercise [<= 50% of maximal oxygen consumption (VO2 max)] as well. In some species, such as dogs (9, 15, 31) and sheep (3, 25), the hypocapnia is progressive and can result in a reduction of arterial CO2 tension (PaCO2) of >10 Torr. Despite multiple reports of exercise-induced respiratory alkalosis, the physiological impetus for exercise hyperventilation is uncertain.

Several facets of exercise could contribute to the reduced PaCO2. These include the following factors.

Exercise Intensity (ExInt)

In exercising ponies, the reduction of PaCO2 is proportional to the ExInt (21). This relationship may indicate that the central respiratory drive inherently oversteps metabolic demand (6, 10). A centrally driven respiratory alkalosis would be expected to stimulate negative feedback, but progressive hypocapnia suggests that any inhibitory feedback is overridden by other factors.

Blood Lactate Concentration ([Lac])

Ventilatory suppression caused by the low PaCO2 could be counterbalanced at the peripheral chemoreceptors by an exercise-induced lactic acidosis. Traditionally, hyperventilation during high-intensity exercise in humans has been interpreted as a compensation for metabolic acidosis (32). However, lactate accumulation is not a prerequisite for hyperventilation in ponies (22), goats (28), dogs (31), or sheep (25).

Thermal Load

Thermoregulatory drive, originating from either a hot environment or internal heat production (or both), may also modify feed-forward respiratory control. An environmental heat load alone can induce resting sheep (13), dogs (12), and antelope (19) to hyperventilate, presumably to increase respiratory evaporative heat loss over the maximum level that can be achieved by isocapnic breathing. Sheep (13) and dogs (12) may increase alveolar ventilation to such an extent that PaCO2 falls to <10 Torr. Because exercise can generate a heat load as much as 10 times that of the hottest possible natural environment (30), it is logical to suppose that an exercise-induced heat load might contribute to the hypocapnia of exercise, particularly in species which rely primarily on panting for thermolysis.

Physical Conditioning

The respiratory controller may be influenced by previous exercise experiences (29). The respiratory controller may "learn" to anticipate the demands of exercise, and hyperventilation may represent a predictive component of the respiratory response to exercise (6). Regular exercise training might thereby modify the exercise respiratory response, as well as increase endurance time and reduce the lactate response to exercise. For these reasons, physical conditioning may be considered a potential modifier of exercise hypocapnia.

The goal of this study was to quantify the relative importance of five factors which may contribute to exercise hypocapnia: ExInt, [Lac], ambient temperature (Ta), body (rectal) temperature (Tre), and level of physical conditioning (Cond). Multiple-regression analysis was used because these five parameters are interrelated. For example, [Lac] is a function of both ExInt and Cond. Previous experiments on respiratory control during exercise in ponies (21, 22, 24) and goats (28) have used short durations of exercise, ranging from 5 to 15 min. A longer duration of exercise, 30 min, was selected for the present study to ensure sufficient time for significant changes in Tre, [Lac], and PaCO2, even at low-intensity exercise.

    METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Animals

Four Dorset ewes [52 ± 2 kg (±SD)] were trained to exercise on a motorized treadmill. Sheep were housed individually in pens (1.8 m2) in a room maintained at 20 ± 2°C. All sheep were fed ~600 g of hay and 250 g of grain daily (Early Market Lamb Pellets; Agway, Syracuse, NY). A minimum of 15 h separated feeding and the start of an experimental trial. Water was provided ad libitum.

Surgical Preparation

Each sheep was prepared with an exteriorized carotid artery by using the method described by Hales and Webster (13). The sheep were allowed at least 2 wk to recover from surgery before the first catheterization of the carotid artery. Two of the animals were also instrumented with a catheter (Tygon microbore tubing, 1.78 mm OD) implanted in the right circumflex iliac artery. The catheter was flushed daily with heparinized physiological saline and was kept filled with heparin sodium (1,000 U/ml).

On the morning of an experiment, unless the sheep had a patent circumflex iliac catheter, the carotid artery was acutely catheterized with a 16-gauge 6.35-cm plastic catheter (IV Cath; Becton-Dickenson, Rutherford, NJ). This catheter was removed at the end of the experimental trial.

Experimental Protocol

Experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals [DHEW Publication No. (NIH) 86-23, Revised 1985, Office of Science and Health Reports, DRR/NIH, Bethesda, MD 20892]. The research protocol was approved by the Institutional Animal Care and Use Committee of Cornell University.

ExInt and Physical Conditioning

Three exercise intensities were selected to represent ~30, 50, and 70% of VO2 max, assuming the average VO2 max of sheep to be 57 ml · kg-1 · min-1 (17). For each ExInt, the treadmill speed and incline were set according to the linear relationship between these factors and the O2 consumption (VO2) of sheep (17). Treadmill settings for each of the three intensities (hereafter referred to as low-, medium-, and high-intensity exercise) are shown in Table 1.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Treadmill inclines and speeds used to produce low-, medium-, and high-intensity exercise

The experiment was divided into two parts on the basis of environmental condition.

Part 1. The warm condition. A moderate environmental heat load was imposed. The average room temperature was 22.4 ± 2.5°C, and each sheep bore a fleece coat at least 2 cm thick. The sheep were habituated to the experimental procedures but were not physically conditioned before the start of experimental trials. Each sheep exercised for 30 min at low and medium intensity and until exhaustion (average 20.6 ± 3.7 min) at high intensity. The trials were ordered randomly with respect to exercise intensity. After these trials, each sheep was physically conditioned for 5 wk. Conditioning consisted of five 30-min exercise bouts/wk and was progressive. For the first week of conditioning, the sheep were exercised for 30 min at low intensity. Over the next 4 wk, the exercise intensity was increased gradually, so that in the final week, the sheep were exercised for 20 min at high intensity plus 10 min at medium intensity. After conditioning, experimental trials were conducted in which each animal again exercised for 30 min at low and medium intensity and to exhaustion at high intensity (average 26.7 ± 2.2 min).

Part 2. The cold condition. The environmental heat load was reduced by lowering the Ta to 12.8 ± 0.5°C and shearing the wool to a thickness of <0.5 cm. Two of the four sheep used in part 1 were randomly selected for participation in part 2. The two sheep completed the experimental trials only in the physically conditioned state. Technical problems with reducing the laboratory temperature prevented completion of the trials with the other two sheep.

During all trials, a window fan was placed behind the sheep. A minimum of 24 h separated any two experimental trials on an individual ewe.

Data Collection

Tre (an index of body core temperature) and temperature of the skin of the ear (Tsk; an indicator of peripheral vasoconstriction or vasodilation) were measured continuously from 5 min preexercise through the end of exercise. Tre was monitored with a copper-constantan thermocouple encased in a sealed plastic tube and inserted ~12 cm into the rectum. Tsk was monitored by using a copper-constantan thermocouple glued to the sheep's ear on the morning of an experimental trial. VO2 was also measured continuously by using an open-flow system described previously (17). Data were stored on computer discs.

Arterial blood samples were drawn between 1 and 5 min preexercise and at 2, 5, 10, 20, and 30 min during exercise. During high-intensity trials, blood was also collected at 15 and 25 min of the duration of the exercise. The blood samples were collected anaerobically into 1-ml syringes, which were capped and immediately placed on ice. Analyses for pH, arterial oxygen tension (PaO2), and PaCO2 were performed, generally within 2 h, with the use of a self-calibrating acid-base analyzer (Radiometer ABL-30; Copenhagen, Denmark). The analysis temperature was adjusted to the Tre of the sheep at the time the sample was taken. The accuracy of the acid-base analyzer was occasionally checked against tonometered sheep or horse blood.

At all blood-gas sampling points, except at 2 min of exercise, an additional 1 ml of arterial blood was collected for analysis of blood [Lac]. Blood was drawn into heparinized plastic syringes and immediately dispensed into microcentrifuge tubes and placed on ice. [Lac] was measured typically within 15 min by using a lactate-glucose analyzer (model 2300 Stat; Yellow Springs Instruments, Yellow Springs, OH). The accuracy of the lactate analyzer was checked against known standards on the day of each experimental trial. [Lac] was not measured in two low-intensity trials.

Statistical Analysis

Multiple-regression analysis was used to form a model of PaCO2 based on ExInt, Tre, Ta, [Lac], Cond, and individual ewe (Sheep). For the multiple regression, Ta and Cond were coded as zero (warm or unconditioned) or one (cold or conditioned). ExInt was represented by the mean VO2 during each exercise bout. The variables measured repeatedly (Tre, [Lac], and PaCO2) were summarized within each experimental trial to eliminate autocorrelation.

Summarization process for Tre, [Lac], and PaCO2. For each exercise bout completed by each sheep, Tre, [Lac], and PaCO2 were regressed against time during exercise. Each variable was modeled separately. A regression equation of order 0, 1, 2, or 3 was applied to the data within each plot. If there was no change in the variable over time, the mean value was used (0 order); otherwise, a first-order equation was tested first. The regression order was then increased until the resulting improvement in the correlation coefficient was <0.05. As a result, >90% of the final correlation coefficients were at least 0.90, excepting the cases in which a zero-order equation was appropriate. The regression equations were then used to calculate the theoretical Tre, [Lac], and PaCO2 at minute 25 (time25) of the exercise period of each experimental trial. The statistical analysis was performed on these index numbers (Tre 25, [Lac]25, and PaCO2 25 ), not on the original data points.

The variables were entered into the multiple regression shown in Eq. 1.
Pa<SUB>CO<SUB>2 25</SUB></SUB> = &mgr; + T<SUB>re 25</SUB> + T<SUB>a</SUB> + ExInt + [Lac]<SUB>25</SUB> + Cond
+ Sheep + all 2-way interactions + Error (1)
where µ signifies the grand mean and Error refers to the residual error.

Because the primary goal was to rank order the main effects, interactions of third, fourth, fifth, and sixth order were excluded from the model.

The variables were entered into Eq. 1 by using the stepwise form of multiple regression (Minitab 11 for Windows); thus the order of the variables shown in Eq. 1 is unimportant. In addition, the best subsets function was used to compare the model produced by stepwise regression with other possible models having the same or a smaller number of regressors. The interrelationships of all variables except Sheep (which is a nominal variable) were quantified via a correlation matrix.

Data are reported as means ± SD except where noted. Because a low alpha level is desirable when comparing correlations (because statistical significance implies only that a given regression coefficient is significantly >0), the standard for statistical significance was set at P < 0.025. Differences between significant correlation coefficients were tested (8) when appropriate.

    RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References

Neither Ta nor Cond had a significant effect on VO2 during exercise. The VO2 values for low-, medium-, and high-intensity exercise were 18.0 ± 1.3, 30.8 ± 0.3, and 44.9 ± 0.9 (±SE) ml · kg-1 · min-1, respectively.

Physical conditioning reduced the [Lac] response to medium- and high- intensity exercise (P < 0.01 and P < 0.02, respectively; Table 2). During low-intensity exercise, [Lac] did not exceed 1 mM in either the unconditioned or the conditioned state.

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Whole blood lactate concentration in response to low-, medium-, and high-intensity exercise before and after physical conditioning

Ta affected Tre and Tsk. Before exercise in the warm environment (part 1), the mean Tre was 38.9 ± 0.3°C (n = 24); in the cold environment (part 2), the mean Tre was 38.2 ± 0.3°C (n = 6). Tre rose during all exercise trials, although in the cold environment, low-intensity exercise failed to raise Tre above 39.1°C, which is within the control range (Fig. 1). The average rate of rise of Tre was proportional to the ExInt (Fig. 1). The highest Tre reached was 41.7°C after 25 min of high-intensity exercise.


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1.   Rectal temperature of sheep during exercise at three intensities in a warm environment (left, open symbols) and a cold environment (right, closed symbols). Exercise began at time 0. Values are means ± SD of 4 sheep in warm environment and means of 2 sheep in cold environment. Circles, low-intensity exercise; squares, medium-intensity exercise; triangles, high-intensity exercise.

During warm trials (part 1), Tsk invariably indicated that the sheep were vasodilated within 5 min of the onset of exercise, and the sheep remained vasodilated throughout exercise. Before exercise in the cold (part 2), Tsk was in all cases <25°C; in five of six cases, Tsk was <20°C, indicating that the sheep were vasoconstricted before exercise. Tsk remained <25°C for the duration of low-intensity exercise. In contrast, at medium- and high-intensity exercise in the cold, Tsk rose by at least 10°C to 25-33°C within 10 min of the onset of exercise, indicating vasodilation.

Before exercise, the mean PaO2 was higher in the warm environment than in the cold environment (115.1 ± 7.7 vs. 102.3 ± 6.4 Torr). In both environments, PaO2 tended to rise during exercise (Fig. 2). At the end of medium- and high-intensity exercise in the warm environment, the mean PaO2 was >135 Torr (Fig.2). At each intensity, the mean PaO2 remained higher in the warm environment than in the cold environment (Fig. 2).


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2.   Arterial oxygen tension (PaO2) of sheep during exercise at 3 intensities in a warm environment (left, open symbols) and a cold environment (right, closed symbols). Exercise began at time 0. Values are means ± SD of 4 sheep in warm environment and means of 2 sheep in cold environment. Symbols are the same as in Fig. 1.

The mean preexercise PaCO2 was slightly lower in the warm environment than in the cold environment (33.8 ± 1.7 vs. 35.5 ± 1.2 Torr). In the warm environment, PaCO2 decreased progressively during exercise at all three intensities (Fig. 3). The rate at which PaCO2 fell was proportional to the exercise intensity (Fig. 3). The lowest PaCO2 recorded was 13.0 Torr, measured in one sheep after 30 min of medium-intensity exercise. In the cold environment, low-intensity exercise failed to depress PaCO2 below the control range (Figs. 3 and 4). During medium- and high-intensity exercise in the cold, PaCO2 fell, but not to as great an extent as in the warm environment (Fig. 3).


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3.   Arterial carbon dioxide tension (PaCO2) of sheep during exercise at three intensities in a warm environment (left, open symbols) and a cold environment (right, closed symbols). Exercise began at time 0. Values are means ± SD of 4 sheep in warm environment and means of 2 sheep in cold environment. Symbols are the same as in Fig. 1.


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 4.   PaCO2 and rectal temperature of 2 sheep during low-intensity exercise in a cold environment. Exercise began at time 0. Open symbols, rectal temperature; closed symbols, PaCO2. Note that both rectal temperature and PaCO2 remained within control range throughout exercise.

Arterial pH tended to rise during exercise, except when the unconditioned sheep exercised at high intensity (Fig. 5). When the sheep were physically conditioned, pH rose more during medium- and high-intensity exercise in the warm environment than it did at the same intensities in the cold environment (Fig. 5). The highest pH observed was 7.72, measured in one unconditioned sheep after 30 min of medium-intensity exercise in the warm environment. The lowest pH recorded was 7.32, measured in one unconditioned sheep after 10 min of high-intensity exercise in the warm environment.


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 5.   Arterial pH of sheep during exercise at 3 intensities in a warm environment (open symbols) before physical conditioning (left), after conditioning (middle), and in a cold environment after conditioning (right, closed symbols). Exercise began at time 0. Values represent means ± SD of 4 sheep in warm environment and means of 2 sheep in cold environment. Symbols are the same as in Fig. 1.

The regression analysis produced two two-regressor equations of equal predictive value
Pa<SUB>CO<SUB>2</SUB></SUB> = −5.93T<SUB>re</SUB> − 0.22ExInt + 269
<IT>R</IT><SUP>2</SUP> = 0.82 (2)
Pa<SUB>CO<SUB>2</SUB></SUB> = −5.74T<SUB>re</SUB> − 0.01(T<SUB>re</SUB> · ExInt) + 261
<IT>R</IT><SUP>2</SUP> = 0.82 (3)
Because the purpose of the multiple regression was to rank order the main effects for influence on PaCO2, Eq. 2 was preferred over Eq. 3 and will be used for purposes of discussion. In either case, Tre alone accounted for 77% of the variance in PaCO2 during exercise (P < 0.0002; Table 3). ExInt (or the interaction between ExInt and Tre) explained an additional 5% of the variance in PaCO2 (P < 0.02). All other variables, i.e., [Lac], Cond, Ta, and individual ewe (Sheep), did not contribute significantly (P > 0.025) to the variance in PaCO2 (Table 3).

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Contribution of ExInt, [Lac], Tre, Ta, Cond, and Sheep to the variance of PaCO2 during exercise

The negative correlation between PaCO2 and Tre during exercise is evident in the primary data (Fig. 6). Each data point represents a single sampling point within a particular trial on an individual ewe; 124 data points originating from 30 trials are shown.


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 6.   PaCO2 plotted as function of rectal temperature (Tre) during exercise. Each data point represents a single sampling point within an exercise trial; 124 data points from 30 trials are shown. Equation for the regression line is PaCO2 = -8.18 Tre + 352.6; R2 = 0.77. Symbols are the same as in Fig. 1.

The correlation matrix for the five independent variables and PaCO2 during exercise is given in Table 4. Correlations between indicator variables (i.e., Ta and Cond) are meaningless and therefore are omitted. Tre, ExInt, and [Lac] were significantly correlated with PaCO2. Both Tre and [Lac] were positively correlated with ExInt; higher exercise intensity produced greater hyperthermia and [Lac]. [Lac] was positively correlated with Tre, indicating that [Lac] and hyperthermia progressed with a similar time course. In addition, Tre was correlated with Ta, indicating that Tre tended to be lower during cold trials.

                              
View this table:
[in this window]
[in a new window]
  		Table 4.   Correlation matrix showing correlations between independent variables and between each independent variable and PaCO2 during exercise

    DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References

Of the variance in PaCO2 during exercise in the sheep, 82% was accounted for by a linear model dependent on only Tre and ExInt (Eq. 2). Tre alone accounted for 77% of the variance in PaCO2 during exercise. These results suggest that the need to increase respiratory evaporation and dissipate metabolic heat during exercise contributes to the development of alveolar hyperventilation and hypocapnia. Thermal drive appears to arise principally from core receptors, because Ta exerted no statistically significant effect on the magnitude of the hypocapnia.

A similar relationship between Tre and PaCO2 during exercise was found previously in the dog (31). Wagner et al. (31) exercised dogs for 30 min at 32 and 50% of VO2 max and observed progressive hypocapnia concomitant with a gradual rise in Tre. They reported a correlation between Tre and PaCO2 in the dog (Eq. 4), which was remarkably similar to that found in the present experiment on sheep (Eq. 5)
Pa<SUB>CO<SUB>2</SUB></SUB> = −7.26T<SUB>re</SUB> + 318.6
<IT>R</IT><SUP>2</SUP> = 0.79 (dog, preexercise and exercise
data pooled; Ref. B31) (4)
Pa<SUB>CO<SUB>2</SUB></SUB> = −8.18T<SUB>re</SUB> + 352.6
<IT>R</IT><SUP>2</SUP> = 0.77 (sheep, present experiment) (5)
Thermal drive to increase respiratory evaporative heat loss appeared to be the primary ventilatory stimulant during exercise in the dog (31).

A negative correlation between Tre and PaCO2 also appears to be present in the results of previous experiments on goats (28) and sheep (3, 25). In addition, although exercise in humans is generally considered to be isocapnic, Hanson et al. (14) found that, during endurance exercise (at least 60 min), runners experienced progressive hypocapnia in association with progressive hyperthermia. Similarly, White and Cabanac (33) found that, once a threshold core temperature was reached, minute ventilation in exercising humans increased in proportion to increases in body core temperature. The latter authors attributed the increased ventilation to thermoregulatory drive that particularly served to cool the brain.

A relationship between Tre and PaCO2 during exercise has not been documented in all species studied, however. Fregosi and Dempsey (11) found no relationship between Tre and the magnitude of hypocapnia in the exercising rat. Although ponies do exhibit both mild hypocapnia (7-11 Torr below rest) and increasing body temperature during exercise (24), a relationship between Tre and PaCO2 has not been definitively shown in ponies. Instead, Pan et al. (21) suggested that PaCO2 may be inversely related to ExInt in the pony. The case for a causal relationship between ExInt and PaCO2 is supported by the observation that the PaCO2 of ponies decreased during a step-up in intensity and increased during a step-down in intensity (21). The rapidity of these changes in PaCO2, corresponding to work-to-work transitions, excludes the possibility that the PaCO2 of the ponies was altered in response to ExInt-related changes in core temperature per se. A negative correlation between ExInt and PaCO2 has also been reported previously in the dog (31), goat (28), and sheep (25).

Because ExInt has a direct effect on the rate of rise of body temperature (Fig. 1), it can be difficult to dissect the effects of body temperature on PaCO2 (in part an indirect effect of ExInt) from the direct effects of ExInt per se on PaCO2. Few studies have been designed to make this distinction. Feistkorn et al. (7) exercised goats instrumented with intravascular heat-exchange devices at the same ExInt at three different core temperatures. They found that PaCO2 during exercise varied inversely with body temperature. Similarly, Bell et al. (3) exercised sheep for 30 min at the same ExInt in a warm environment and in a cool environment. They found a significantly greater decline in PaCO2 in the warm environment, in concert with a significantly greater rise in Tre. In the present experiment, hypocapnia was not observed during low-intensity exercise in the cold environment, the only condition in which the sheep remained vasoconstricted and Tre remained in the normothermic range (Fig. 4).

The differences among species in the apparent influence of body temperature and ExInt on PaCO2 may stem from differences in thermoregulatory strategy and experimental conditions. Sheep, goats, and dogs are panting species and thus rely on the respiratory system for evaporative heat loss, whereas horses depend primarily on cutaneous evaporative heat loss during heat stress (16). Rats neither pant nor sweat but instead exhibit behavioral thermoregulation. Regardless of the thermoregulatory strategy, respiration does provide a potential avenue of heat loss; nonetheless, a panting species might logically demonstrate a stronger thermal drive to respiration than a nonpanting species.

Increased respiratory evaporative heat loss caused by hyperventilation provides a rationale for the inverse relationship between body temperature and PaCO2 observed in some animals during exercise. In contrast, the significance of the relationship between ExInt per se and PaCO2 is unclear. On the basis of data collected in the exercising pony, Forster and Pan (10) speculated that nonhuman mammals might exhibit fewer alveolar-capillary adjustments to exercise than do humans (who generally maintain isocapnia during low- and moderate-intensity exercise), thus rendering these nonhuman species dependent on an increased alveolar PO2 to maintain PaO2 homeostasis. Actual measurements of ventilation-perfusion inequality do not support the premise of this hypothesis. Galloping horses have better matching of ventilation to perfusion than do running humans (27). Nonetheless, hyperventilation may benefit PaO2 at severe exercise intensities (80-100% of VO2 max), when the pulmonary transit time decreases toward the minimum time required for alveolar-arterial PO2 equilibration. An increase in alveolar PO2 could serve to prevent arterial oxygen desaturation (10, 22).

At moderate exercise intensities, however, arterial oxyhemoglobin desaturation is not likely. Bayly et al. (2) found that PaO2 homeostasis was maintained by horses exercising at 40% of VO2 max, both before and after the horses began to hyperventilate. These authors concluded that thermal drive was the most likely impetus for the progressive hypocapnia exhibited by the horses after 25 min of exercise (2). In the exercising sheep, hyperventilation raises PaO2 (Fig. 2; see also Ref. 3), suggesting that sheep are not dependent on hyperventilation to maintain PaO2 at its resting level. Even if PaO2 were to fall in the absence of hyperventilation, at low altitudes, a very large fall in PaO2 (30-40 Torr) would be required for a significant decrease in arterial oxyhemoglobin saturation. In contrast, the thermoregulatory system is sensitive to small changes in body temperature, and thermal drive is a more tenable explanation for the hyperventilation observed in the exercising sheep. The inverse correlation observed between ExInt per se and PaCO2 may reflect an inherently excessive feed-forward respiratory drive during exercise (6), rather than a causative relationship.

The lack of a significant independent effect of [Lac] on exercise PaCO2 in the present study agrees with a previous report in sheep (25) and with a similar result in the exercising pony (22). Furthermore, the traditional hypothesis of H+ stimulation of hyperventilation during high-intensity exercise in humans has been questioned (6). The lack of an independent effect of Ta on PaCO2 suggests that the influence of Tsk can be overridden or overwhelmed by core temperature. Physical conditioning does not appear to modify the response of the respiratory controller to input from the other variables.

Limitations

In exercising ponies (23) and horses (16), the use of Tre instead of arterial temperature to correct arterial gas tensions has been shown to cause analytical error, because Tre increased more slowly than arterial temperature during exercise. The severity of the error in correction of PaCO2 was found to depend on the ExInt and the duration of exercise, increasing with either factor (23). However, in a sheep exercising at ~50% of VO2 max, Tre lagged only transiently behind blood temperature in the right ventricle. By 20 min of exercise Tre surpassed right ventricular temperature (26). The maximum difference between Tre and right ventricular temperature during 30 min of exercise was 0.2°C (26). Bradley et al. (4) reported that a 1°C fall in blood temperature lowers the PCO2 by 4.4%. Based on these findings, the error in PaCO2 during exercise caused by correction by Tre should be <1 Torr. Thus, it is unlikely that error in estimation of PaCO2 caused by using Tre for temperature adjustment has significantly biased the conclusions of this study.

The significant correlations among several of the independent variables, particularly among Tre, ExInt, and [Lac], obscure determination of which variable contributes most to the prediction of PaCO2. Although physical conditioning achieved some dissociation between ExInt and [Lac], unfortunately removal of the fleece and cold Ta were not sufficient to uncouple ExInt and Tre. A univariable linear model for exercise PaCO2, using only Tre, would have an R2 of 0.77 (Eq. 5). A similar model, using only ExInt, would have an R2 of 0.64 (square of the Pearson correlation coefficient given in Table 4). On post hoc statistical examination, the correlation coefficient between Tre and PaCO2 was not significantly larger than that between ExInt and PaCO2 (P > 0.025). Nonetheless, the interchangability of ExInt with the ExInt · Tre interaction (Eqs. 2 and 3) reduces the status of ExInt per se as a predictor of exercise PaCO2. In contrast, Tre appears in both equations. Based on the relationship between Tre and PaCO2 at each ExInt (Fig. 6), the significance of the ExInt · Tre interaction may indicate that Tre is particularly influential at medium and high intensities. The omission of [Lac] from the multiple-regression model is expected, given the relatively low independent correlation of [Lac] and PaCO2 (0.51) and the fact that the correlations between [Lac] and both Tre and ExInt are greater than that between [Lac] and PaCO2.

Implications

In the sheep, a panting species, thermal drive has a greater role in determining PaCO2 during exercise than does either ExInt per se or [Lac]. Thus, during exercise, temperature regulation appears to be a higher homeostatic priority than acid-base regulation. Experiments addressing respiratory control in panting species, especially those involving exercise, should be designed to account for thermal effects. Moreover, continued study of the mechanism by which hyperthermia affects hyperventilation is merited. Hyperthermia may increase chemoreceptor gain, leading to an enhanced respiratory response to a given CO2 load. Alternatively, hyperthermia may lower the set point at which PaCO2 is regulated. In the case of either increased gain or decreased threshold, a direct effect of temperature on the hypothalamus and respiratory centers is a possible mechanism. However, panting animals can maintain brain temperature below that of the body (1). Consequently, any direct effect of temperature on the brain must be exerted by a smaller change in temperature than that experienced peripherally.

    ACKNOWLEDGEMENTS

The authors thank Victoria Wildman and Dr. Susan Kimmel for assistance with this research.

    FOOTNOTES

Address for reprint requests: P. L. Entin, Dept. of Medicine, Box 0623A, 9500 Gilman Drive, Univ. of California, San Diego, La Jolla, CA 92093-0623.

Received 30 December 1996; accepted in final form 4 March 1998.

    REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References

1.   Baker, M. A. Brain cooling in endotherms in heat and exercise. Annu. Rev. Physiol. 44: 85-96, 1982[Medline].

2.   Bayly, W., H. Schott II, and R. Slocombe. Ventilatory responses of horses to prolonged submaximal exercise. Equine Vet. J. 18: 23-28, 1995.

3.   Bell, A. W., J. R. S. Hales, R. B. King, and A. A. Fawcett. Influence of heat stress on exercise-induced changes in regional blood flow in sheep. J. Appl. Physiol. 55: 1916-1923, 1983[Abstract/Free Full Text].

4.   Bradley, A. F., M. Stupfel, and J. W. Severinghaus. Effect of temperature on PCO2 and PO2 of blood in vitro. J. Appl. Physiol. 9: 201-204, 1956[Abstract/Free Full Text].

5.   Butler, P. J., N. H. West, and D. R. Jones. Respiratory and cardiovascular responses of the pigeon to sustained, level flight in a wind-tunnel. J. Exp. Biol. 71: 7-26, 1977[Abstract/Free Full Text].

6.   Dempsey, J. A., H. V. Forster, and D. M. Ainsworth. Regulation of hyperpnea, hyperventilation, and respiratory muscle recruitment during exercise. In: Regulation of Breathing, edited by J. A. Dempsey, and A. I. Pack. New York: Dekker, 1995, p. 1065-1117.

7.   Feistkorn, G., A. Nagel, and C. Jessen. Circulation and acid-base balance in exercising goats at different body temperatures. J. Appl. Physiol. 57: 1655-1661, 1984[Abstract/Free Full Text].

8.   Ferguson, G. A. Statistical Analysis in Psychology, and Education (3rd ed.). New York: McGraw-Hill, 1971, p. 170-171.

9.   Flandrois, R., J. R. Lacour, and H. Osman. Control of breathing in the exercising dog. Respir. Physiol. 13: 361-371, 1971[Medline].

10.   Forster, H. V., and L. G. Pan. Breathing during exercise: demands, regulation, limitations. In: Oxygen Transfer from Atmosphere to Tissues, edited by N. C. Gonzalez, and M. R. Fedde. New York: Plenum, 1988, p. 257-276.

11.   Fregosi, R. F., and J. A. Dempsey. Arterial blood acid-base regulation during exercise in rats. J. Appl. Physiol. 57: 396-402, 1984[Abstract/Free Full Text].

12.   Hales, J. R. S., and J. Bligh. Respiratory responses of the conscious dog to severe heat stress. Experientia 25: 818-819, 1969[Medline].

13.   Hales, J. R. S., and M. E. D. Webster. Respiratory function during thermal tachypnoea in sheep. J. Physiol. (Lond.) 190: 241-260, 1967[Abstract/Free Full Text].

14.   Hanson, P., A. Claremont, J. Dempsey, and W. Reddan. Determinants and consequences of ventilatory responses to competitive endurance running. J. Appl. Physiol. 52: 615-623, 1982[Abstract/Free Full Text].

15.   Hastings, A. B., F. C. White, T. M. Sanders, and C. M. Bloor. Comparative physiological responses to exercise stress. J. Appl. Physiol. 52: 1077-1083, 1982[Abstract/Free Full Text].

16.   Hodgson, D. R., L. J. McCutcheon, S. K. Byrd, W. S. Brown, W. M. Bayly, G. L. Brengelmann, and P. D. Gollnick. Dissipation of metabolic heat in the horse during exercise. J. Appl. Physiol. 74: 1161-1170, 1993[Abstract/Free Full Text].

17.   Jones, M. T., R. E. Rawson, and D. Robertshaw. Determination of maximal oxygen consumption in exercising pregnant sheep. J. Appl. Physiol. 73: 234-239, 1992[Abstract/Free Full Text].

18.   Kiley, J. P., W. D. Kuhlman, and M. R. Fedde. Respiratory and cardiovascular responses to exercise in the duck. J. Appl. Physiol. 47: 827-833, 1979[Abstract/Free Full Text].

19.   Maskrey, M., P. P. Hoppe, and O. S. Bamford. Blood gas changes during panting in a small East African antelope, the dik-dik. J. Appl. Physiol. 44: 534-537, 1978[Abstract/Free Full Text].

20.   Mitchell, G. S., T. T. Gleeson, and A. F. Bennett. Ventilation and acid-base balance during graded activity in lizards. Am. J. Physiol. 240 (Regulatory Integrative Comp. Physiol. 9): R29-R37, 1981.

21.   Pan, L. G., H. V. Forster, G. E. Bisgard, R. P. Kaminski, S. M. Dorsey, and M. A. Busch. Hyperventilation in ponies at the onset of and during steady-state exercise. J. Appl. Physiol. 54: 1394-1402, 1983[Abstract/Free Full Text].

22.   Pan, L. G., H. V. Forster, G. E. Bisgard, C. L. Murphy, and T. F. Lowry. Independence of exercise hyperpnea and acidosis during high-intensity exercise in ponies. J. Appl. Physiol. 60: 1016-1024, 1986[Abstract/Free Full Text].

23.   Pan, L. G., H. V. Forster, and R. P. Kaminski. Arterial vs. rectal temperature in ponies: rest, exercise, CO2 inhalation, and thermal stresses. J. Appl. Physiol. 61: 1577-1581, 1986[Abstract/Free Full Text].

24.   Parks, C. M., and M. Manohar. Blood-gas tensions and acid-base status in ponies during treadmill exercise. Am. J. Vet. Res. 45: 15-19, 1984[Medline].

25.   Pethick, D. W., C. B. Miller, and N. G. Harman. Exercise in merino sheep: the relationships between work intensity, endurance, anaerobic threshold and glucose metabolism. Aust. J. Agric. Res. 42: 599-620, 1991.

26.   Rawson, R. E., D. Robertshaw, and P. L. Entin. Control of respiratory evaporation during exercise in sheep (Abstract). Physiologist 35: 74, 1992.

27.   Seaman, J., B. K. Erickson, K. Kubo, A. Hiraga, M. Kai, Y. Yamaya, and P. D. Wagner. Exercise induced ventilation/perfusion inequality in the horse. Equine Vet. J. 27: 104-109, 1995[Medline].

28.   Smith, C. A., G. S. Mitchell, L. C. Jameson, T. I. Musch, and J. A. Dempsey. Ventilatory response of goats to treadmill exercise: grade effects. Respir. Physiol. 54: 331-341, 1983[Medline].

29.   Somjen, G. G. The missing error signal: regulation beyond negative feedback. News Physiol. Sci. 7: 184-185, 1992.[Free Full Text]

30.   Taylor, C. R. Exercise and environmental heat loads: different mechanisms for solving different problems? In: Environmental Physiology (2nd ed.), edited by D. Robertshaw. Baltimore, MD: University Park, 1977, p. 119-145.

31.   Wagner, J. A., S. M. Horvath, and T. E. Dahms. Cardiovascular, respiratory, and metabolic adjustments to exercise in dogs. J. Appl. Physiol. 42: 403-407, 1977[Abstract/Free Full Text].

32.   Wasserman, K. Breathing during exercise. N. Engl. J. Med. 298: 780-785, 1978[Medline].

33.   White, M. D., and M. Cabanac. Exercise hyperpnea and hyperthermia in humans. J. Appl. Physiol. 81: 1249-1254, 1996[Abstract/Free Full Text].

J APPL PHYSIOL 85(1):318-325
8570-7587/98 $5.00 Copyright © 1998 the American Physiological Society


This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
N. Fujii, Y. Honda, K. Hayashi, H. Soya, N. Kondo, and T. Nishiyasu
Comparison of hyperthermic hyperpnea elicited during rest and submaximal, moderate-intensity exercise
J Appl Physiol, April 1, 2008; 104(4): 998 - 1005.
[Abstract] [Full Text] [PDF]

Home page
 J ANIM SCIHome page
G. S. Archer, T. H. Friend, D. Caldwell, K. Ameiss, and P. D. Krawczel
Effect of the seaweed Ascophyllum nodosum on lambs during forced walking and transport
J Anim Sci, January 1, 2007; 85(1): 225 - 232.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
M. D. White
Components and mechanisms of thermal hyperpnea
J Appl Physiol, August 1, 2006; 101(2): 655 - 663.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
D. Robertshaw
Mechanisms for the control of respiratory evaporative heat loss in panting animals
J Appl Physiol, August 1, 2006; 101(2): 664 - 668.
[Abstract] [Full Text] [PDF]

Home page
 J. Exp. Biol.Home page
S. M. Deban and D. R. Carrier
Hypaxial muscle activity during running and breathing in dogs
J. Exp. Biol., July 1, 2002; 205(13): 1953 - 1967.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
P. L. Entin, D. Robertshaw, and R. E. Rawson
Effect of locomotor respiratory coupling on respiratory evaporative heat loss in the sheep
J Appl Physiol, November 1, 1999; 87(5): 1887 - 1893.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Entin, P. L.
Right arrow Articles by Rawson, R. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Entin, P. L.
Right arrow Articles by Rawson, R. E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online