regulation of the circulation during exercise...400 yl./mih i ii 200 -i 0 (excited) (8wmn) (c&m)...

Regulation of the Circulation During Exercise

Cardiac Output (Direct Fick) and Metabolic Adjustments in the Normal Dog1

A. C. BARGER, V. RICHARDS,2 J. METCALFE3 AND B. GttNTHER*

From the Department of Physiology, Harvard Medical School, Boston, Massachzcsetts

ABSTRACT

Oxygen consumption and cardiac output (direct Fick) have been measured in normal dogs at rest and during graded exercise on the treadmill up to a work intensity of 5 mph and 10~. Systemic and pulmonary artery pres- sures have also been recorded. The changes in cardiac output produced ‘at rest’ by excitement were frequently as large as those induced by moderate exercise. A short bout of exercise followed by a rest period was far more efficacious in producing lower and more uniform results during rest and subsequent exercise than a prolonged rest period alone. Under such conditions the ‘steady state’ was reached in 3 minutes or less of exercise. The linear relation between oxygen consumption and cardiac output during exercise in the dog is similar to that observed in man, and in the horse. The possible significance of this similarity is discussed and it is suggested that the data are consistent with the hypothesis that the increase in blood flow during exercise is largely the increase in muscle flow with a constant arteriovenous oxygen difference of approximately 14 vol. %.

T ECHNIQUES for the measurement of cardiac output in the dog running on the treadmill have been developed in this

laboratory (I). With the determination of cardiac output in the unanesthetized, exercising dog, data are now available in three species (man, horse and dog) for the study of the regulation of the circulation during exercise, and for a re-evaluation of the factors controlling muscle blood flow. The present report: on the dog indicates that if excitement is excluded the close relation between oxygen consumption and cardiac output in this animal is similar to that seen in man and the horse. Fur- thermore the evidence herein presented is consistent with the hypothesis that at rest only a small fraction of the total blood flow through skeletal muscle actually passes through nutrient muscle capillaries, while most of the

Received for publication July 20, 1955.

1 Aided by a grant from the Life Insurance Medical Research Fund and by the Eugene Higgins Trust through Harvard Univ.

2 Fellow of the Commonwealth Fund. 3 Fellow of the Life Insurance Medical Research

Fund. 4 Fellow of the W. K. Kellogg Foundation.

flow passes through non-nutrient: shunts. As the intensity of exercise increases, a larger and larger percentage of the total peripheral flow traverses nutrient muscle capillaries. This report, dealing with the normal dog, also pro- vides a base line for studies of cardiac output and venous gradients during exercise in dogs with mild valvular heart damage and in dogs with frank congestive failure.

METHODS

Male, mongrel dogs, weighing from 18-27 kg who ran with ease on the treadmill were selected for these studies. Since masks did not appear feasible for respiratory studies in the dog during strenuous exercise, the animals were prepared for tracheal intubation by sub- cutaneous transplantation of the trachea at a preliminary operation. Under general anesthesia a mid-line incision in the neck was made and the muscles, from the lower border of the larynx to several centimeters above the sternal notch, were displaced laterally. The trachea was dissected free, and while thus elevated, the muscles were sewed behind it. At the same time both carotid arteries were transplanted subcutaneously between the trachea and the external jugular veins.

After convalescence the animals were trained to lie quietly on the fluoroscopy table in preparation for the experiments. After local anesthetic (17~ procaine) had been injected liberally into the appropriate regions, the minor operative procedures necessary for the cardiac

by 10.220.32.247 on April 5, 2017

http://ajplegacy.physiology.org/D

ownloaded from

http://ajplegacy.physiology.org/

614 BARGER, RICHARDS, METCALFE AND GUNTHER

VENTILATION

L/MINUTE

(STPD) 25’

cl

200 MEAN ARTERIAL

6LOOD PRESSURE

MM. HC 100

cl

200 HEART RATE PER 100

MINUTE

20 ARTERIAL BLOOD

OXYGEN CONTENT

(VOLUMES IO PERCENT)

8

6 CARDIAC

OUTPUT

L./MIN. 4

RESPIRATORY QUOTIENT

El

TOTAL PERIPHERAL RESISTANCE

cm?

100 STROKE

75 VOLUME

so ML&EAT

25 ml

0

I

STANDING 3MPH-0’ SITTING 3MPH-C

20 MIXED VENOUS

BLOOD 10 OXYGEN CONTENT

(VOLUMES PERCENT)

3MPH -IO’

1 800

I 600

OXYGEN

CONSUMPTION

400 YL./MIH

I II 200

-I 0

(EXCITED) (8WmN) (C&M) (WMfiIN.) (8Q5M /MIN)

FIG. I. Effect of excitement on the circulation and respiration of a dog at rest and during exercise.

output determinations were performed. For the sampling of the mixed venous blood a radiopaque catheter was inserted into the pulmonary artery through a metal cannula introduced into a jugular vein. The position of the catheter tip in the pulmonary artery was checked by fluoroscopy during insertion. During the course of the experiment the maintenance of this position was ascertained by the character of the pressure tracings. A polyethylene catheter was next threaded into an exposed carotid artery. Finally, the trachea, lying just beneath the anesthetized skin, was exposed and incised through several rings. A curved polyethylene tube with an in- flatable rubber cuff was then inserted. This tube had an internal diameter of 1.5 cm and fitted snugly into the trachea. At the completion of the experiment the catheters were removed and the tissue edges approximated

under pressure bandages. The trachea itself did not require sutures for successful healing, and up to six experiments have been performed on the same dog over a period of several months.

Heart rate was recorded by a cardiotachometer. Respiratory rate was registered with a pneumograph connected to an ink-writing oscillograph. The zero reference point was the level of the catheter as it passed through the right atrium. This position was determined fluoroscopically while the dog stood upright, and the site was marked on the chest wall.

Prior to an experiment water was given ad libitum but food was withheld for 24 hours. Then, with the tracheal cannula and catheters firmly fixed in place under bandages the dogs were exercised on a motor driven treadmill at graded intensities of work, ranging from

by 10.220.32.247 on April 5, 2017


ownloaded from


CIRCULATION DURING EXERCISE 615,

TABLE I. SUMMARY OF CIRCULATORY AND BETABOLIC CHANGES IN THE DOG DURING GRADED EXERCISE

Dog I, 21.5 kg Standing

3 mph-o”

3 mPh-$O

3 mph--IO0

Standing

3 mph-o”

3 mph--IO’

5 mph----IO’

Dog 2, I7*7 ks Standing

3 mPh-5O

4 mph---IO0

Dog 3, 23 kg Sitting

3 mph-o’

3 mph---IO’

Dog 4, 20.2 kg Standing

3 mph-o”

4 mph-IO0

Dog 5, 265 ki Sitting

3 mph-o”

3 mph--IO0

Dog 6, r9.8 kj 3 mph-o”

Cardiac output

l/min.

3.6 4.0 50 s-7 $*9 7-6 8.0 4.1 3.9 54 5** 7.6 8.7

12.1

12.2

3.4 3.6 ii** 4.9 7.4 8.4

2.4 4.1 4.3 6.1

58

1.8 2.1

3*7

36.: .

4.3 7*8 7.8 9*$

IO.$

56

57 52

=

02

Consump. A-V02

Diff.

Mixed Venous OgiJ-;

.

Stroke Vent. Volume (STPD)

ml/min. ml/l. ml/l. min. ml/beat I/??&.

200 SW 118.9

238 47.0 13$*1 393 78.3 104.1

$98 105.7 74.8 620 105 94 75-I 785 103.4 71-7 855 106.4 73-I 207 $04 85.7 I93 so*5 857 442 81.0 61.8

431 84.9 $0 820 108.3 38*9 870 99.8 49.6

1420 117.2 30.4 1383 112.4 37.2

8s 42 100 40 III 45 126 45 130 45 160 47 160 50

90 46

*so 36 130 39 148 51 180 49 210 $8 204 60

7.6 0.80 7-6 0*79

12.1 0.76

39-o 0.93 29*4 0.83 50.8 0.91

65.1 0.85

33.6 o-94 27.1 1.09

33.2 0.92

3$*6 0.92 49.6 1.03

62.8 0.99 78.9 1.04

83.5 0.93

255 74*7 125.8 III

231 65.0 133.5 96 559 108.8 95-o *so 521 106.8 91.7 144 925 125.0 774 207

1016 120.3 82.2 201

7.3 o-77 7-o o*79

20.4 0.84 18.9 0.85

38*7 0.92 60.3 0.8~

185 410 463 786 696

138 163 406 425 945

238 644 $89

1008

111s

420 490 516

76.3 132.4 100.8 102.4

106.6 97*7 129.9 77.0 119.6 84.6

78

=zP 204

76.6 147-o 77-o 145.8

110.0 113.0

114*3 114.8

I$$*3 77-9

31 37 34 34 39 42

31

29 30

25

30

43

47

49

33 33 32

7.3 0.75 20.0 0.90 24.8 0.78 42.7 0.90

39-o o*94

72

123

140

13.0 1.24

10.8 1.03

25.4 0.96

3497 1.02

43.6 0.93

559 138.1

82.1 120.0

75.9 120.1

106.5 91.6

I%*7 92.7

90

195

7-4 0.74 30.2 o-79 30.1 0.78 44.6 0.86 49.2 0.85

75-s 92.5 168

86.0 83.6 =7= 10493 69.7 165

4o*5 I.O$

43-I 0.88 44.8 0.83

3 mph on the level to 5 mph at IO'. The first measurement of cardiac output during each period of exercise was begun after the dog had been running for 3 minutes. During each work period two consecutive determinations of cardiac output were usually made. Between the graded bouts of exercise the dogs were allowed to rest for 15-30 minutes.

Expired air was collected in a Tissot spirometer over

=

-

Mean Blood Pressure

Puhnon artery

-- mm Hg

II

16

20

12

16

16

16

8

12

IO

14

Carotid artery

mm Hg Zyne-sec/cmb

80 IS95 128 2050

120 1680

112

160

160

120

184

1180

3w

2320

1260

I2I$

128

128

112

96

130

I36

144

3005

2OIO

1065

1730

= 790

s70

3130

=

Total Periph.

Resistance

a period of 3-4 minutes for the lighter work loads, and 1.5-2 minutes for the higher work intensities. Low resistance molded flap valves were used to direct fIow, and the total dead space probably did not exceed the normal dead space of the animal, since the tracheal tube was inserted below the larynx. The expired air was analyzed for CO2 in the Haldane apparatus, and for O2 in a Pauling oxygen meter.

by 10.220.32.247 on April 5, 2017


ownloaded from



TABLE 2. COMPARISON OF CONSECUTWE DETERMINATIONS OF CARDIAC OUTPUT (ML/KG/MIN.) AND RESPIRATORY

QUOTIENT DURING EXERCISE*

Dog I Dog I Dog 2 Dog 3 Dog 4 Dog 6 Work Load

Standing

3 mph-o”

3 mph--s0

3 mph----IO0

4 mph--IO0

5 mph----IO’

- I R.Q. C.O. R.Q. C.O.

168

186

232

R.Q. R.Q. C.O.

104

I77 186

264

2s

C.O.

89 104

183

183

301

R.Q.

o-77 o*79

0.84 0.85

0.92

0.81

R.Q.

I. 24

1.03

0.96 1.02

o-93

R.Q. ___--

283 I.05

288 0.88

263 0.83

0.80

o-79 0.76

o*93 0.83 0.91

0.85

161

294

294

359

396

o-75

0.90

0.78

0.90

o-94

191 182

256

237

353 405

562

566

0.94 1.09 0.92

0.92

288

0974

o-79 0.78

0.86

0.85

276 1.03

o-99 418

475 1.04

o-93 I

1 1 l * All dogs incluaea m this table were given a preliminary bout of mild exercise and then allowed a rest period of 30 min. before these determinations were started.

Mixed venous and arterial blood samples were drawn simultaneously in heparinized, oiled syringes continu- ously over nearly the entire period of collection of expired air. The blood specimens were stored in an ice bath, and then analyzed as quickly as possible for 02 content by the method of Van Slyke and Neil (2).

In hot weather a fan was placed at the side of the treadmill to blow a stream of air through the dog’s mouth in order to increase the evaporation of saliva and to avoid any complicating hyperthermia in the dogs whose respired air passed through the tracheal cannula rather than over the tongue.

of 36.8 l/min. with a respiratory quotient of 1.07).

With the onset of very mild exercise (3

mph on the level) while the animal was still excited, cardiac output rose to 7.1 l/min. with an oxygen consumption of 468 ml/min. Heart rate rose abnormally high for such a light work load (180 beats/min.) and ventilation increased to 48 l/min. Following duplicate determinations of cardiac output at 3 mph the

OBSERVATIONS animal was allowed to rest for 30 minutes.

After this rest period, and when the dog was Effect of Excitement on Cardiac Output at more relaxed, cardiac output was only 2.4 l/

Rest and During Exercise. In preliminary min. at a time when the oxygen consumption experiments the changes in cardiac output produced by excitement in the ‘resting’ dog

was only slightly less than in the first ‘resting’

were frequently noted to be as large as those determination. Ventilation was now only 7.3 l/min. with an R.Q. of 0.75 and heart rate 78.

induced by moderate exercise. Furthermore, the effects of excitement were apparent during mild exercise as well as in the ‘resting’ dog, making comparisons of circulation and metab- olism during increasing work loads meaning- less. The magnitude of the effects produced by excitement are illustrated in figure I. In this experiment, cannulations having been com- pleted, the dog was placed on the treadmill for the resting measurements. The initial ‘resting’ cardiac output of 4.4 l/min. seemed high for this well-trained dog. Under these conditions it was usual to find several clear indica-

same work rate as in the first bout (3 mph on the level) cardiac output rose to a value that was slightly less than that of the first ‘rest’ period, and only $3 % of the output during the first bout of exercise, while oxygen consumption was 90% of that measured in the first bout of work. Heart rate was 148 (in contrast to 180) and ventilation only 50% of

When the animal was again exercised at the

that noted previously. Even at 3 mph and IO’

when the average oxygen consumption was 740 ml/min., cardiac output was 12 % less than that observed during the first bout of exercise

tions of excitement, e.g. a) restlessness, b) relative tachycardia (90 beats/min.), and c)

at 3 mph on the level. Many examples of this

hyperventilation (respiratory minute volume sort made it clear that for dependable results, measures must be taken to exclude the excite-

by 10.220.32.247 on April 5, 2017


ownloaded from


CIRCULATION DURING EXERCISE 617

FIG. 2. Relation between work and oxygen consumption in the dog exercising on the treadmill. S&Q? line was fitted to the data by the method of least squares. Dotted lines are modified from the data of

on Smith (9) for man exercising the treadmill (figs. 18 and 19).

ment with which many dogs . P

approach the treadmill and the first bout or exercise.

1.400 -

1.200 - f E

1.000 - :

g F -800 - a E 2

-600 - $

E .400 - z

.200 -

WORK - KG-M/MN.

OO I I I I I I I 1 I 1 \

50 100 150 200 250 300 350 400 450 500

Simply prolonging the time of preparatory standing on the treadmill was not effective because this enforced inactivity simply en- hanced the excitement in those animals that were eager to run. Far more efficacious was a short bout of mild exercise followed by a rest period. Results were more uniform and lower during rest and subsequent exercise than those obtained after a prolonged rest period alone. For example, in one dog (dog I-table I) standing on the treadmill cardiac outputs were 3.6 and 4.0 l/min. on one occasion, with values of 4.1 and 3.9 one month later. These outputs were, on the average, only slightly higher than three consecutive measurements performed on the same dog under pentobarbital anesthesia. When the stage of excitement in early anesthesia had passed the cardiac outputs were 3.4, 3.9 and 4.3 l/min. Therefore, for the later experiments, the dogs were first walked at 3 mph-o” for 3 minutes and then allowed to rest for 30 minutes before the first resting cardiac output determination was started.

‘Steady State’. Determination of the cardiac output by the Fick principle assumes that the arterial and venous specimens have sampled the blood which absorbed the oxygen removed from the gas phase in the lungs. Theoretically, therefore, blood flow through the lungs can be determined under most conditions by simulta- neous measurements of oxygen absorbed, and the arteriovenous oxygen difference. In prac-

tice instantaneous samples are difficult, if not impossible, to obtain. The validity of the values obtained by the Fick method can only be ac- cepted with complete confidence when the subject is in the steady state, i.e. when the rate of oxygen consumption is relatively constant, and when the arteriovenous oxygen difference is not varying significantly. Furthermore only under these conditions is the measured oxygen consumption equal to the oxygen consumed by the tissues. Moreover, in order to compare values obtained in various laboratories for specified degrees of activity, the determinations should be done under comparable conditions. Again, this is feasible only in the ‘steady state.’

To determine the length of time necessary to attain the steady state in exercising dogs, two consecutive cardiac output determinations were made during each work load. The first sampling was started after the dog had been running for 3 minutes, and the second some 3-5 minutes after the completion of the first measurement. Agreement between these values was good at all work intensities (table 2) and the respiratory quotients were normal, indi- cating that the steady state was reached in less than 3 minutes even during work at 5 mph--IoO. In addition cardiac output varied relatively little even during a do-minute run at 3 mph on the level in one dog (dog 6-283, 288

and 263 ml/kg/min.-table 2).

Oxygen Consumption, Cardiac Output and Other Circulatory Measurements at Rest and

by 10.220.32.247 on April 5, 2017


ownloaded from



I OXYGEN CONSUMPTION littn/minu'tt

OO 1 I I I I i I

0.2 0.4 0.6 0.6 1.0 1.2 1.4

40-g s Y

eo-*$

e 5 OXYGEN CONSUMPTION - L/MN.

OO I I I I I I

0.200 0.400 0,600 0.600 1.000 1.200 1.400

FIG. 3. Relation between oxygen consumption and cardiac output of the dog at rest and during exercise. The

line is fitted to the data by the method of least squares. The equation of the line is ob = 6.95 voz + 1.85, where

(jb is the cardiac output in l/min., and ~OZ is the oxygen consumption in l/min. FIG. 4. Arteriovenous oxygen difference of the dog during graded exercise, using the oxygen consumption as the

measure of work intensity.

During Graded Exercise. With each increase in work intensity oxygen consumption rose in linear fashion (table I and fig. 2). The increased oxygen requirements of exercising muscle were met by a) a rise in cardiac output and b) by an increase in arteriovenous oxygen difference. For example, with an elevation of oxygen consumption from 200 ml/min. at rest to 1400 ml/min. at 5 mph----IO’ (dog z-table I) the sevenfold increase of oxygen consumption was met by a threefold rise of cardiac output (4.0 to 12.1 l/min.), and a 2.3-fold increase in arteriovenous oxygen difference (so.5 to I 15

ml/l). The linear relation between cardiac output

and oxygen consumption for six dogs at rest and during exercise is illustrated in figure 3. Over the range studied, the equation of the line as determined by the method of least squares is:

. Qb = 6.95 v02 + 1.85

where ob is cardiac output in liters per minute and To, is oxygen consumption in liters per minute.5 The curve does not pass through the origin, but with each increment in oxygen consumption there is a constant rate of increase in cardiac output, i.e. cardiac output goes up 7 liters for each liter rise in oxygen consumption.

6 Since the standard error of the slope is 0.44 there is only one chance in IOO that the slope differs by as much as rt 1.20.

However, this rise in cardiac output alone was not sufficient to supply the oxygen consumed. As indicated above, the arteriovenous oxygen difference increased in hyperbolic fashion as illustrated in figure 4, taken from the data of dog I.

Heart rate rose progressively with increased work load (table I and fig. s), but varied considerably from animal to animal at rest and with each work load. For example, heart rates ranged from 72-11 I during standing, and 148- 204 at 3 mph and IO'.

This variability of heart rate is reflected in the stroke volume (table I and fig. 6). The stroke output of these well-trained dogs of approximately the same size varied from 25-47 ml/beat during quiet standing. The changes observed during exercise again differed considerably from animal to animal. Stroke volume increased 72 % (25 ml/beat to 43 ml/beat) in one dog (4) going from rest to 4 mph--IO’, while that of dog z increased 36 % (31 ml/beat to 42 ml/beat) for the same work span. In dog 3 no change in stroke volume was observed with exercise up to 3 mph-IO’. Stroke output remained about 30 ml/beat while cardiac output rose from 2.4 l/min. to 6.1 l/min., with heart rate increasing from 78 to 204. It is note- worthy that in this same dog excitement was associated with a stroke volume of 49 ml/beat. Figure 6 shows a plot of stroke volume against oxygen consumption to compare with heart

by 10.220.32.247 on April 5, 2017


ownloaded from



250

200

150

100

50

0

I I I I I I I

OXYGEN CONSUMPTION - L/MN. I I L I I

0.200 0.400 0.600 0.800 1.000 1.200 1.400

FIG. 5. Effect of graded exercise on the heart rate of the dog, using the oxygen consumption as the measure of work intensity.

rate (fig. 5). A small cardiac output could be met by a small stroke volume and a high heart rate, or vice versa. As the work intensity rises and the cardiac output increases the spread of the stroke volumes tends possibly to become smaller.

In the few determinations made, the changes in mean, systemic blood pressure were not large (table I). Total peripheral resistance (table I) fell with the onset of exercise (one exception) and fell progressively as the intensity of work increased. Mean pulmonary arterial pressure (measured against atmos- pheric pressure) remained relatively constant even during exercise requiring a three to four- fold increase in cardiac output.

Energy Expenditure and Efficiency During Exercise. The energy cost for the dog of trans- porting I kg I m horizontally has been calculated by subtracting the energy cost of standing from that measured while the dog walked at 3 mph on the level (80.5 m/min.). The energy cost, in terms of heat production per kilogram meter ascribable to walking on the level varied from 0.42-0.92 Cal. with a mean of 0.67 Cal/kg m. The energy cost of performing I kg m work of elevation on the grade at 5” and IOO at 3 mph (work of ascent) was found to vary between 4.17 cal. and 7.58 cal. with an average value of 6.37 cal. The net climbing efficiency (caloric equivalent of work ascent divided by the difference between the energy cost of walking on the level and the energy cost of walking on the grade at the same speed) ranged from 31% to 56 %, with an average net climbing efficiency of 38 %.

DISCUSSION

The linear relation between the rise in cardiac output and the increase in oxygen consumption in the dog during exercise is apparent only if the effects of excitement are carefully avoided, and if the animals are in the ‘steady state.’ Even in well-trained animals, external evidence of excitement was associated, to a striking degree, with a high ‘resting’ cardiac output, as well as with an abnormally high output in relation to oxygen consumption during exercise. The dog’s circulatory response to excitement is thus similar to that observed in man (3)*

The circulatory effects of excitement may persist for some time in the dog standing on the treadmill, and as has been indicated, a short bout of very mild exercise followed by a rest period is far more efficacious in producing lower and more uniform results during rest and during exercise than a prolonged rest period alone. The effectiveness of mild exercise in over- coming the hyperactivity of the circulation of the apprehensive subject has also been observed in man (4). The lowered cardiac output in relation to oxygen consumption produced by preliminary exercise may be one of the benefits derived by atheletes from ‘warming up’ before a race.

The oxygen consumption of the dog standing on the treadmill is relatively high compared to basal oxygen consumption. Zuntz has pointed out (5) that he found the oxygen consumption of the standing dog approximately 40 % above that measured in the dog lying at

I OXYGEN CONSUMPTION - L/MIN. 1 I I I I

0 0.200 0.400 0.600 0.800 1.000 1.200 1.400

FIG. 6. Effect of graded exercise on the stroke volume - of the dog, using the oxygen consumption as the measure of work intensity.

by 10.220.32.247 on April 5, 2017


ownloaded from



70 r CARDIAC OUTPUT AND OXYGEN CONSUMPTION

DURING EXERCISE

HORSE (ZUNTZ 6 HAGEMANNI

MAN (BOCK & DILL)

DOG (OUR DATA)

20

IO

I OO

I I , I I 1 I I I 2 3 4 5 8 7 8

OXYGEN CONSUMPTION

(LITERS/MIN)

FIG. 7. Comparison of the relation between cardiac output and oxygen consumption in the dog, in man (9) and in the horse (IO).

rest, and that the increase measured on the assumption of the upright position was much larger than that seen in man. The average oxygen consumption of our dogs standing on the treadmill is 36 % above the average basal values found by Spencer et al. (6) in their 1%

zo-kg dogs, which are in close agreement with Zuntz’s observations. Similarly, our figures for oxygen consumption during exercise in dogs of equivalent size are only slightly lower than those of Zuntz.

th Because of the dog’s speed and endurance .e assumption has been made th .at his effi-

ciency is considerably higher than that of man. Actually, as indicated in figure 2, the oxygen

consumption for various work loads (in kg m/min.) is not much lower than the earlier data of Smith (7) for man (broken lines), and the calculated net climbing efficiency of our dogs was only slightly above the more recent values reported by Erickson et al. (8) for trained men. It seems likely that the d iff erence in performance in exercise of man and dog can be attributed to the differences in the circulatory systems. For a given oxygen consumption the cardiac output of the dog is approximately 3 l/min. lower than that of man (fig. 7) and this may account for the dog’s superior performance. With the exception of the lower cardiac output for a given oxygen consumption the circulatory changes observed in the dog during graded exercise are similar to those reported by Bock and his colleagues (9) on a

well-trained marathon runner. There is a linear correlation between the rise of cardiac output and oxygen consumption, a large increase in heart rate, with smaller changes in arteriovenous oxygen difference, and the smallest percentage change in stroke volume. The relative magnitude of the circulatory changes for one dog (I) has been schematically summar- ized in figure 8. For the dog, the relation between cardiac output and metabolic rate over the range studied is represented by the equation:

Qb = 6.95 \~OZ -I- 1.85

The slope of the line is almost identical with the average slope found by Bock (9) for man . (Qb = 7 & + 5). The equation indicates that cardiac output rises 7 liters for each liter increase in oxygen consumption. This relation apparently holds for work performed by man on the treadmill or bicycle ergometer, and is true regardless of the method used for deter- mining cardiac output, as shown in figure 9. These data are gathered from the references cited in the legend of figure 9.

The horse is the only other species for which data on the cardiac output and oxygen consumption at rest and during exercise are available. The figures of Zuntz and Hagemann (IO)

for the horse yield a similar relation between cardiac output and oxygen consumption (Qb = 7 vop + 11.35)/ The similarity of the slopes of the lines (fig. 7) for dog, man and horse suggests that the regulation of oxygen consumption and circulation in exercise may be similar in the three species. The details of this regulation are not known. However, since increases in tissue oxygen consumption during exercise are met by increases in blood flow and by increases in oxygen extraction from the blood, it is of interest to see how each is related to oxygen consumption.

The relation between blood Aow (cardiac output) and oxygen consumption is linear, as has been shown (fig. 3). It is possible to ex- amine also the relation between arteriovenous oxygen difference and oxygen consumption by solving simultaneously the Fick equation for cardiac output, and the equation for cardiac

6 Several of the values which appear to be in error, were excluded in this calculation. If these figures were included the slope would be 8.

by 10.220.32.247 on April 5, 2017


ownloaded from



output derived from our data @b = 7 \j02 + 1.85).

then

.

Qb

iroz x IO3 = --

- cvo, Ca0,

cjb = 7iro2 + K,

( > I

0 2

voz x IO3 = 76% + K (3)

Ca0, - c30,

or

Cao, - cvo, jriToz + K (4)

IO3 Ca0, - cvo, E------

K

7fs (5)

2

The relation between arteriovenous oxygen difference and oxygen consumption is hyperbolic, and the data for one experiment plotted in figure 4 (dog -r) fits this equation with the best fit when K is 2.3. The above relation also indicates that as vo2 becomes larger the arteriovenous oxygen difference approaches 143

ml/l. or 14.3 vol. % (limit Cao&o, = 103/7 = 143 ml/l.). This figure, as a limiting value for arteriovenous oxygen difference in hard exercise agrees with the data of Christensen for man (I 2).

At first glance it may appear difficult to reconcile the linear relation between cardiac output and oxygen consumption, with the hyperbolic relation of the arteriovenous oxygen difference and oxygen consumption. How- ever, these observations could be reconciled if the changes in blood flow during exercise were largely the increase in muscle blood flow with a constant arteriovenous oxygen difference of approximately 14 vol. % in this blood. Thus, arterial-mixed venous oxygen difference would approach 14 vol.% as a limit. The data presented here would be consistent with such a process. For example, if the assumption were made that 20 % of the cardiac output goes to muscle (skeletal and cardiac) at rest (13, 14),

and 80% to the other tissues of the body, and

? - Qb = Cardiac output in l/min.; VOW = oxygen Consumption in l/min.; Caoz = concentration of oxygen in arterial blood in ml/l.; Cvo2 = concentration of oxygen in mixed venous blood in ml/l. (II).

400* I I I I 1 I I

300-ga

s:'i /gs;

E Q PERCENT OF RESTING OXYGEN CONSUMPTION

. OO

1 I I 1 . 1 IO0 200 300 400 500 600 700

FIG. 8. Schematic representation of the relative magnitude of the circulatory changes in the dog during

exercise.

that during exercise the increase in cardiac output is entirely augmented muscle flow with a constant arteriovenous oxygen difference of 143 ml/l., the calculated mixed venous oxygen content agrees well with the observed values. Thus in experiment 2 of dog I the resting cardiac output averaged 4 l/min. Twenty per cent of this going to muscle would be 0.8 l/min., with 3.2 l/min. to the remainder of the body. If the latter portion remained relatively constant* then at 3 mph at o” the muscle blood flow would be 5.5 l/min. minus 3.2 l/min ., leaving 2.3 l/min. This is 42 % of the total cardiac output now going to muscle. Hence, if we assume that the arteriovenous oxygen difference in the nonmuscular tissue remains the same at rest (50 ml/l.), the mixed arteriovenous oxygen difference would be :

Caoz - cvo:! = .42 X 143 + .58 X 50 = 89 ml/l.

The observed value was 81. For the other determinations in this experiment the calculated and observed values were:

Work Load Calculated A-V Observed A-V Oxygen Diff. Oxygen Diff.

ml/l. ml/l.

Rest 50 3 mph--o” 85 85 3 mph-10’ IO4 108

109 100

5 mph---IO’ 119 117 119 112

Thus, the data in this paper would be consistent with the hypothesis suggested some years ago by Christensen (16), that the venous oxygen content in muscle is relatively con-

8 Herrick ef al. (IS) did not find any significant de- crease in visceral blood flow in the dog during exercise ux) to a work intensitv of mph---q’.

by 10.220.32.247 on April 5, 2017


ownloaded from



stant, and with increased metabolic demands the flow through muscle capillaries is increased with arteriovenous oxygen difference remain- ing relatively unchanged. It would seem un- likely that the flow through muscle would be determined by the arteriovenous oxygen difference per se. The controlling factor more likely is the venous oxygen tension, and the figure of 14 vol. % as the arteriovenous oxygen difference is merely the resultant of this factor in the presence of an oxygen capacity of approximately 20 vol. %. In the dog, assuming a PH of 7.4 the venous oxygen tension at 30% saturation would be 22 mm Hg (17), a figure close to that found for mixed venous oxygen tension by Bock et al. (9) in strenuously exercising man. Such an interpretation, as indicated above, would also explain the findings of Bishop, Donald and Wade (18) in patients with anemia. In these individuals with a low oxygen capacity of the blood there is also a linear relation between cardiac output and oxygen consumption during exercise but with

30-

CARDIAC

OUTPUT

LITERS/MN

FIG l 9. Relation between oxygen on the bicycle ergometer, or pushing

a steeper slope, since the maximum arteriovenous oxygen difference through nutrient vessels of muscle would be smaller. Thus, cardiac output would increase more with each increment in oxygen consumption.

Cardiac muscle has been shown to have a relatively constant arteriovenous oxygen difference of 13-15 vol. %, with flow rising approximately sevenfold for a unit rise in oxygen consumption (19). However, direct experimental evidence for a constant arteriovenous oxygen difference in skeletal muscle has been lacking. The arteriovenous oxygen difference is stated to be small at rest, and to increase with exercise (20, 2 I). Recently we (22) have noted that the arteriovenous oxygen difference in the hindleg of the unanesthetized dog was small at rest (4-7 vol. %), but with relatively mild exercise rose to I 2-15 ~01% and remained relatively constant as the work intensity was progressively increased. Furthermore, Andres, Cader and Zierler (23) have found that the arteriovenous oxygen difference of blood from

0 b 0

b

b l m

l b b b b

b b b

5 b ‘0 ’

b ’

l b

l e

l ee .’ : ’ l

l l , b

l

OXYGEN CONSUMPTION

LITERS/MIN

Oo 1 I L I I I 1 L I I ,

0,5 I .o I.5 2.0 2.5 3.0 3.5

consumption and cardiac output of man at rest, exercising on the treadmill, against footrest. The data were obtained from sources in references g, 12,28-34.

by 10.220.32.247 on April 5, 2017


ownloaded from



forearm muscles of man (brachial artery- median cubital profunda vein) rises to 9-14

vol. % with light exercise. Similar observations have also been made by Love (24). These findings in the dog and man, of a small arteriovenous oxygen difference in resting muscle, with sudden, large increase with mild exercise, could be explained on the basis of two channels for blood flow through skeletal muscle: a) non- nutrient shunts which carry most of the flow at rest, and which have a small arteriovenous oxygen difference, and b) nutrient muscle capillaries with a large, constant arteriovenous oxygen difference, and which carry a larger and larger percentage of the muscle blood flow as exercise intensity increases. Such non- nutrient shunts in muscle have been postulated by Pappenheimer (25) and Barcroft (13) on physiological grounds. Recently Zweifach (26)

has described two channels for blood flow in muscle. At rest most of the muscle blood flow is through large interfascicular vessels, with only sporadic flow through nutrient muscle capillaries. In exercise the nutrient muscle capillaries open and more of the muscle blood flow is diverted through these vessels. The presence of two distinct channels for blood flow through muscle may well explain the discrep- ancy between the radioactive sodium clearance and the directly measured flow through muscle during body heating, and the correlation during exercise (2 7).

We wish to acknowledge with sincere appreciation the stimulating and constructive criticism of Drs. E. M. Landis, R. H. Kellogg and F. E. Yates.

REFERENCES

I. LARGER, A. C., V. RICHARDS, J. METCALFE AND

B. GUNTHER. Federation Proc. II: 8, 1952. 2. VAN SLYKE, D.D. AND J.M. NEILLJ. Biol. Chem.

61: 553, 1924. 3. HICKAM, J. B., W. H. CARGILL AND A. GOLDEN.

J. Clin. Investigation 27: 290, 1948.

4. LANDIS, E. M. Personal communication. 5. ZUNTZ, N. Arch. ges. Physiol. 68: 191, 1897.

6. SPENCER, F.C.,D.L. MERRILL,~. R. POWERS AND

R. J. BING. Am. J. Physiol. 160: 149, 1950. 7. SMITH, H. M. Gaseous Exchange and Physiological

8.

99

IO.

II.

12.

I3*

14*

15

16.

=7*

18.

199

20.

21.

22.

23.

24,

25,

26,

27.

28.

29.

3Q

319

32.

33*

34*

Requiremwts for Level and Grade Walking. Wash- ington: Carnegie Inst., 1922.

ERICKSON, L., E. SIMONSON, H. L. TAYLOR, H. ALEXANDER AND A. KEYS. Am. J. Physiol. 145:

391, 1945-6. BOCK, A. V., C. VANCAULAERT, D. B. DILL, A. FRILLING AND L. M. HURXTHAL. J. Physiol. 66: 136, 1928.

ZUNTZ, N. AND 0. HAGEMANN. Stojwechsel des Pjerdes. Berlin: Paul Parey, 1898.

PAPPENHEIMER, J. R. Federation Proc. 9: 602, 1950. CHRISTENSEN, E. H. Arb. Physiol. 4: 470, 1931.

BARCROFT, H. AND H. J. C. SWAN. Sympathetic Con- trol of Human Blood Vessels. London: Arnold, 1953.

BAZETT, H. C. Factors Regulating Blood Presswe. New York: Macy, 1949, p. 61. HERRICK, J. F., J. H. GRINDLAY, E. J. BALDES AND F. C. MANN. Am. J. Physiol. 128: 338, 1939-40.

CHRISTENSEN, E. H. Ergebn. d. Physiol. 39: 348,

=937* DILL, D. B., H. T. EDWARDS, M. FLORKIN AND R. W. CAMPBELL. J. Biol. Chem. 95: 143, 1932.

BISHOP, J. M., K. W. DONALD AND 0. L. WADE.

c&z. SC. 14: 329, 1955. FOLTZ, E. L., R. G. PAGE, W. F. SHELDON, S. K. WONG, W. J. TUDDENHAM AND A. J. WEISS. Am. J. Physiol. 162: 521, 1950.

BARCROFT, J. AND T. KATO. Phil. Tr. Roy. Sot. Lon- don,B. 207: 149, 1915.

KRAMER, K., F. OBAL AND W. QUENSEL. Arch. ges. Physiol. 241: 717, 1939.

BARGER, A. C., V. RICHARDS, J. METCALFE AND B. GUNTHER. Unpublished observations. ANDRES, R., G., CADER AND K. L. ZIERLER. Per- sonal communication. LOVE, A. H. G. C&z. SC. q-:275, 195.5. PAPPENHEIMER, J. R. J. Physiol. 99: 182 and 283,

1941. ZWEIFACH, B. W. AND D. B. METZ. Federation Proc. 14: 168, 1955. BARCROFT, H. AND A. C. DORNHORST. Peripheral Circulation in Man. Boston: Little, 1954, p. 124. GROLLMAN, A. Am. J. Physiol. 90: 210, 1929.

DEXTER, L.,J.L. WHITTENBERGER, F.W. HAYNES, W. T. GOODALE, R. GORLIN AND C. G. SAWVER. J. Appl. Physiol. 3: 439, 1951. DOUGLAS, C. G. AND J. S. HALDANE. J. Physiol. 56: 69, 1922. DILL, D. B., J. H. TALBOTT AND H. T. EDWARDS. J. Physiol. 69: 267, 1930. HOCHREIN, M., J. H. TALBOTT, D. B. DILL AND L. J. HENDERSON. Arch. f. exper. Path. u. Pharma- kol. 143: 147, 1929. HICKAM, J. B., W. H. CARGILL AND A. GOLDEN. J. Clin. Investigation 27: IO, 1948.

RILEY, R. L., A. HIMMELSTEIN, H. L. MOTLEY, H.M. WEINER AND A. COURNAND. Am.J.PhysioZ. 152: 372,1948.

by 10.220.32.247 on April 5, 2017


ownloaded from


regulation of the circulation during exercise...400 yl./mih i ii 200 -i 0 (excited) (8wmn) (c&m)...

Documents