Circulatory in different

regulation during exercise ambient temperatures

ETHAN R. NADEL, ENZO CAFARELLI, MICHAEL F. ROBERTS, AND C. BRUCE WENGER John B. Pierce Foundation Laboratory and Departments of Epidemiology Physiology, Yale University School of Medicine, New Haven, Connecticut

and Public 06519

Health

and

NADEL,ETHAN R., ENZOCAFARELLI,MICHAEL F. ROBERTS, in the heat, both the skin and muscle demands for blood AND C, BRUCEWENGER.CircuZatory regulation during exerflow are high and the body is presented with the problem cise in different ambient temperatures. J. Appl. Physiol.: Res- of providing sufficient circulation to both of these vaspirat. Environ. Physiol. 46(3): 430-437, 1979.-Three relatively cular beds. fit subjects performed duplicate 20- to 25-min cycle ergometer Accompanying the increase in SkBF with body heating exercise bouts at moderate and heavy intensities (40 and 70% is an increase in the peripheral venous volume because VO2 max ) in ambient temperatures of 20, 26, and 36OC. They cutaneous vasodilation enhances the rate at which the approached a steady state in internal body temperature (T,,) in capacitance vessels fill (17). Further compounding the all but the heavy exercise in the heat, where T,, rose consistproblem of a lowered central circulatory volume due to ently, averaging 38.84”C at the termination of exercise. Cardiac venous pooling, there is a loss of plasma from the intraoutput (Q), estimated by a rebreathing technique, was proporvascular volume. Plasma volume decreases in excess of tional to Vo2 and independent of the body temperatures,.except during the lower exercise intensity in the heat, where Q aver15% have been seen following 50 min of moderate exercise aged 1.3 Lmin higher throughout. In any environment, fore(9). If uncompensated the progressive reduction in cenarm blood flow was linearly related to T,, above the T,, threshtral blood volume would lead to progressive reductions old for vasodilation, but during heavy exercise in the heat this in cardiac filling pressure, resulting in tachycardia and relationship was severely attenuated above a T,, around 38.0°C, ultimately, circulatory collapse. However, the body does when forearm blood flow exceeded 15 mlmin-’ 100 ml? compensate to some degree: the muscle pump squeezes Plasma volume decreases during exercise were primarily a function of the intensity of exercise. During heavy exercise in a part of the cutaneous venous volume back toward the heart and perfusion of inactive tissues is reduced in the heat the relative vasoconstriction contributes to the maintenance of an adequate stroke volume preventing a fall in Q. In proportion to the increased demand from the skin (18). Furthermore, we (13) and Brengelmann et al. (3) have this case, circulatory regulation has precedence over temperature regulation. recently shown a reduction in the rate of SkBF increase per unit of internal temperature rise in conditions of high body temperatures. Presumably this modification in temperature regulation; cardiac output; forearm blood flow; plasma volume; heat SkBF control is the result of the baroreceptor reflex superimposing its influence upon the thermoregulatory drive. It is unclear, however, whether the relative cutaDURING DYNAMIC EXERCISE such as cycling or running, neous vasoconstriction is associated with the achievethe increase in cardiac output must be partitioned be- ment of a critical decrease in plasma volume, a critical level of internal body temperature, a critical level of tween the contracting muscles to meet the oxygen delivery requirements, and the skin to meet the heat transfer SkBF or, secondarily, a critical reduction in cardiac outrequirements of the temperature regulatory system. The put in the given conditions. It is not known how great rate of muscle blood flow is primarily related to the the reduction in plasma volume is in conditions of heayy intensity of exercise and, in simple analysis, can be con- exercise in the heat. Moreover, the only systematic study sidered to be a function of the local events at the level of of the central circulatory responses in these conditions the muscle, e.g., local oxygen tension (11). The rate of showed a reduction in cardiac output of around 2 1. skin blood flow (SBBF) is primarily determined by the min-’ in comparison to the same intensity of exercise in body temperatures. Above an internal body temperature cool conditions (19). If this were the case, exercise should threshold for vasodilation, the increase in SkBF is pro- be limited to relatively short durations. portional to the increase in internal temperature. IncreasThe present experiments were undertaken to attempt ing the average skin temperature causes a reduction in to answer the above questions. To do so, it was necessary the internal temperature threshold for vasodilation withto make serial measurements of cardiac output, plasma volume changes, and skin blood flow on subjects exercisout affecting the slope of the SkBF to internal temperature relation (23). During exercise in a cool environment, ing in various conditions, and to compare these physiothe vasodilatory threshold is high and the demand for logical responses to the thermal and metabolic characSBBF is relatively low. However, during heavy exercise teristics of the body over time. By studying these circul

430

0161-7567/79/0000-oooO$O1.25

Copyright

0 1979 the American

Physiological

Society

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

CIRCULATORY

REGULATION

DURING

latory parameters simultaneously, we hoped to be able to discern specific strategies for meeting conditions of multiple demand. METHODS

Three relatively fit young men (mean physical characteristics: ht 180 cm; wt 79 kg; age 30 yr; maximal aerobic power 49 ml 02 emin-’ kg-‘) volunteered to participate in the present experiments. Each had experienced similar types of studies in this laboratory and was fully informed about all procedures. Experiments on any individual were performed at the same time of day to avoid variability attributable to the circadian rhythm of body temperature (24). Experiments were conducted primarily in fall and winter; presumably, none of the subjects was heat acclimatized. Each subject performed duplicate 20- to 25min bouts of cycle ergometer exercise at moderate and heavy intensities in ambient temperatures of 20, 26, and 36°C with water vapo r pressure always less than 16 Torr. Moderate and h.eavy exercise required . approximately 40 and 70% of maximal aerobic power (VOW max), averaging 100 and 180 W for the three subjects. Any exercise bout at a given ambient temperature provided a range of internal temperature at approximately a steady mean skin temperature (Tsk). Varying the exercise intensity between experiments gave two thermal and metabolic loads at similar environmental conditions, and allowed us to examine the circulatory responses to the change in load at a given level of Tsk. Performing the exercise bouts in different ambient conditions provided comparable internal loads at different levels of Tsk. By having the subjects perform duplicate experiments in each condition, somewhat greater precision was attained by averaging the data from two identical experiments and using the averaged data as representative of the given condition. The differences between duplicate experiments were small. In the few instances where a given variable differed considerably from that in the duplicate experiment, a third identical experiment was performed and the data from the three experiments analyzed. Subjects dressed in shorts and athletic shoes entered the experimental chamber and rested for approximately 30 min prior to the day’s test while seated in the contour chair of the cycle ergometer; the arrangement of the contour chair behind the pedals, with legs nearly horizontal (2), allowed subjects to exercise at heavy intensities while leaving their arms free for blood flow measurements or for blood samplin .g withou t movement interference. During the 30 min of rest a t the experimental temperature, all probes and measurement equipment were attached Internal body temperature (T,,) was continuously recorded from a thermocouple in the esophagus at heart level. Mean skin temperature was calculated each minute from the thermocouple temperatures at eight skin sites as previously described (12). Oxygen uptake (voz) and carbon dioxide production #co,) were calculated from continuous recordings of the fractions of 02 and CO2 in the expired air and continuous l

431

EXERCISE

recording of the expired ventilatory volume. Interruptions in these records were made during cardiac output determinations. Cardiac output was estimated five times during each exercise bout by a simple one-step rebreathing technique (8). A mouthpiece is attached to the common arm of a wide-bore three-way stopcock, with two solenoid valves controlling the direction of airflow. In the normal position, the flow is directed to and from a Collins triple-J low-resistance valve: expired air can be collected and/or sampled from a mixing chamber downstream from the triple-J valve. When the solenoid switches are activated, the airflow is directed into a 5 liter anesthesia bag and the subject is forced to rebreathe from the closed system. The rebreathing procedure is as follows: the bag is flushed and emptied and a predetermined volume (subject-specific, dependent on the subject’s tidal volume during exercise at the given intensity) of a high-02, COafree gas mixture is introduced into the bag. The subject is alerted that a rebreathing procedure is imminent and attempts to normalize his breathing pattern (note that subjects must be well trained in the entire maneuver to provide records that are consistent and artifact free). Following a signal and precisely at end expiration (again, training is essential) the subject depresses a button which activates the solenoids and redirects airflow into and from the closed system. The subject’s first respiratory move following the switch is an inspiration and he continues to rebreathe at 50 breathsmin-’ fQr about 15 s. A needle in the mouthpiece is connected by a short length of polyethylene tubing to the head of a Beckman LB-2 infrared analyzer. During the rebreathing maneuver respired air is continuously sampled (500 ml* min-‘) for determination of the fraction of COZ. Time delay of the entire system is less than 1 s. The infrared analyzer is calibrated against gases of known CO2 concentrations (checked against Haldane analysis). After 12-15 s, a second signal is given and the subj ect releases the button and resumes normal ventilation. The details of analysis of the rebreathing curve, as well as the justifications and comparisons with other methods for estimation of cardiac output, have been recently reported by Farhi et al. (8). All of the variables for solving the Fick equation are derived from the rebreathing curve and the four breaths just prior to rebreathing. In our hands this technique has proved to work quite well. We have validated the technique against literature values of cardiac output over a wide range of oxygen uptakes during exercise and have been able to obtain high reproducibility from measurement to measurement during steady-state exercise at a given oxygen uptake (Table 1). In addition, serial estimates of cardiac output may be made within a few minutes of each other without contamination from the indicator, as washout of CO2 is rapid. Because of these attributes, as well as the attribute of noninvasiveness, we have found this to be the technique of choice. Heart rate recorded immediately prior to rebreathing

was used to deduce stroke volume.

Forearm blood flow

was measured twice each minute by venous occlusion plethysmography with a Whitney mercury-in-Silastic

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

432

NADEL,

1. Serial cardiac output estimations individual experiments - -_____ ~.---------~-~~~~ _ - ~~.

TABLE

f;-om ---

Time, min Subj

Mean 3

7

12

17

ML RS EN

14.3 15.0 10.7

40% vo:!maw (1.55 Lomin-7 13.2 13.1 13.2 14.1 14.5 14.5 10.7 10.0 12.0

ML RS EN

19.8 18.4 18.4

17.0 19.1 21.9

70% VO2

Estimations are given in a 20°C environment.

max

17.2 20.5 in liters

f SD

22

11.5 14.2 9.3

13.1 * 1.0 14.5 t 0.4 10.5 t 1.0

18.8 16.8 18.2

18.4 t, 1.2 18.0 t 0.9 19.4 t 1.7

(2.82 I* min-‘) 18.1 18.4 18.0 per minute.

Experiments

took

place

strain gauge (25). As noted earlier, movement artifacts during exercise were eliminated by having the subjects seated behind the pedals in a contour chair and suspending the arm in a sling from the ceiling. Forearm skin temperature, which has an important effect on the absolute value of forearm blood flow (3), was not controlled as in previous studies (23). The forearm instead was allowed to arrive at its own temperature. Because the changes in forearm blood flow during leg exercise or body heating are limited to the skin (10, 15), changes in forearm blood flow can be interpreted as representative of changes in skin flow over most of the body when the local skin temperature is not artificially clamped. Plasma volume changes during exercise were estimated from the changes in hemoglobin (Hb) concentration in samples of venous blood drawn from a Zl-gauge butterfly catheter placed in an arm vein. The l-ml samples were taken immediately prior to exercise and around 2, 6, and 20 min of exercise after blood in the dead space of the catheter was drawn and discarded. Assuming that the red cell volume is constant during this short interval (14, Zl), the relative change in plasma volume at the different times of exercise can be estimated from the change in either Hb (cyanmethemoglobin technique) or Hct-concentration (1)) taking into account differences between central and peripheral sampling (5). This technique has recently been validated by simultaneous determination of radioiodinated serum albumin changes during exercise and heat exposure by Harrison et al. (9).

ROBERTS

AND

WENGER

best-fit linear regression analyses of the raw data from each subject. Because of the great many discrete blood flow measurements per experiment (35-40 measurements) and the continuous recording of Tes, regression analysis is the method of choice. The correlation coefficient of SkBF against T,,, above the Tes threshold for vasodilation, was always greater than 0.9. Individual responses of the three subjects were similar to each other in every case. In the steady state of exercise, cardiac output (&> was proportional to Vo2 and independent of the time of measurement, the body temperatures, and the other circulatory variables (skin blood flow or plasma volume). The cardiac output was nearly always at its steady-state level by the time the first estimate was made, between 2 and 3 min following the onset of exercise. At the lower exercise intensity both heart rate (HR) and stroke volume (SV) had reached a steady-state level by the 2-3 min measurement and remained relatively constant throughout exercise. There were no important differences according to ambient (or skin) temperature, with the sole exception that HR and & were significantly (P < 0.01, using paired sample analysis) higher (15 beats min-l and 1.3 Lmin-‘) in the 36OC environment than in the 20 or 26OCconditions. During exercise at 70% VOW max, SV was always at its highest level at 2-3 min of exercise and tended to fall over the course of exercise. & was maintained by a persistent increase in HR that compen-

2. Thermal and circulatory data of 3 subjects exercising at 40% V02 max(1.55 19min-I) -

TABLE

t,min T, = 20°C

2.9

‘f’sk = 32.O”C

6.8 11.3 16.8 21.3

T, = 26°C

2.7

T,k = 33.7”C

6.7 11.2

RESULTS

Subjects were able to approximate or approach a steady state in internal body temperature within the ZO25 min in all but the heavy exercise in the heat, where T,, rose consistently, averaging 38.84OC at the termination of exercise. During heavy exercise in the 20 and 26OC environments, T,, averaged 38.18 and 38.12”C at the end of exercise. Clearly, regulation of internal body temperature was impaired in conditions of high internal and external thermal loading. Tables 2 and 3 give the pertinent thermal and circulatory data over the course of exercise in the 20, 26, and 36OC environments. Most of the data are shown as means of individual determinations &SE. However, the T,, and SkBF data are the mean values taken directlv from the

CAFARELLI,

16.3 21.4

T, = 36°C

2.6

‘ii& = 35.2”C

7.3 11.7 16.3 21.7

Values condition.

09 1.min

’

HR, beatsmin ’

SV, ndbeat ’

T

oc (1%

SkBF*

AFV, 76

13.3 kO.5 12.7 to.4 12.5 to.5 13.2 kO.4 12.3 kO.4

111 tl 112 tl 111 tl 114 tl 115 *l

120 tl 113 t1 112 tl 116 tl 107 kl

37.10

1.4

37.31

1.7

37.54

2.5

37.67

2.7

37.67

3.4

-0.5 to.8

12.5 to.7 14.1 to.3 12.6 to.9 12.6 kO.5 14.4 to.6

112 k4 115 t,4 113 t3 114 t4 113 +_3

111 +3 123 +3 111

37.13

3.4

37.38

4.4

1.3 t1.2 -1.6 k1.0

37.57

6.1

111 +I 124 +,4

37.62

7.5

37.60

6.5

-1.6 t1.1

13.5 to.5 14.8 to.6 14.4 k0.6 14.2 to.4 14.6 to.6

118 tl 123 &I 126 tl 129 tl 131 tl

114 tl 119 t6 114 t7 111 t2 110 t5

37.19

6.0

37.45

9.5

-4.9 k0.8 -7.0 k1.0

37.60

12.3

37.67

13.1

37.77

14.8

are means & SE. Duplicate * ml-min-’ 100 ml-‘.

1.7 kO.7 0 to.8

t1

experiments

are given

-7.0 to.8 in each

l

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

CIRCULATORY

REGULATION

DURING

3. Thermal and circuhtory data exercising at 70% lb, max(2.82 I* min-‘) .-- _______-______ _-~--..__-----_ _lil__-~I_______ -_--.______~_- - -~-...~.. TABLE

t, min

T, = 20°C

3.0

Q, 1. min ’

bE2. min. ’

SV, mlbeat- ’

T,.,, “C

18.3

136

18.3 t0.6 19.0 kO.7 17.8 kO.2

134 tl 151 t1 155 z!El 162 *I

18.5

HI.4 = 32.0°C

Rk

7.8 12.0 17.0 22.3

T, = 26°C

3.1

Rk = 33.6Oc

7.8 11.4 16.3 21.5

T, =

36°C

2.6 6.6 10.4

15.0 19.3

* ml

l

min-’

_

AFT,%

37.1

2.1

-3.6

121 *2 122 t,2 110 &I

37.68

3.3

-12.2 zkO.8

37.95

6.3

38.09

8.0

38.18

*l

21.0

110

&l

Zkl

17.5 kO.9 16.8 21.2 18.8 +a9 17.6 kO.9

135 a 143 +;l 150 tl 152 *l

130 k3 117 21 125 kl 115 22

37.20

2.7

37.64

6.4

37.85

8.4

38.02

12.7

38.12

14.2

17.4

159

109

kO.9

tl

k3

17.8

147

121

,tl

kl

17.6 kO.4 18.0 to.4 18.3 kO.2

163 Zkl 172 Z!A 179 *I

19.0

kO.6

l

__-..-

SkBF+

168

10.6

-16.0

kO.9

-0.3 to.5

-8.5 AA.5

- 12.9

k1.5 37.34

7.4

108 &I

37.84

13.2

105

38.17

16.5

kl 102 +l

38.54

18.2

185

103

38.84

19.7

tl

kl

Values are means 2 SE. Duplicate condition.

of 3 subjects

to.2

HI.4 Tsk = 35.5*c

433

EXERCISE

-5.3 k1.1 -12.8 a.0

rise. This reduction in SkBF increase per unit of T,, rise

was consistent in all three subjects. The average data for both exercise intensities at the extreme environments are shown in Fig. 1. The data from the 26OC environment were omitted for clarity. Plasma volume decreases during exercise were primarily a function of the exercise intensity and its duration and were affected by the body temperatures only when Tsk was high, as shown in Fig. 2. Plasma volume changes were negligible during exercise at 40% VOW max, except in the 36OC environment, when plasma volume decreased by 7%, with nearly all of the decrease occurring at the beginning of exercise. During exercise at 70% v02max, volume decreases were relatively large, with the greatest amount of the decrease (up to 13% of the plasma volume) occurring within the first 8 min of exercise. Thereafter, the rate of volume loss diminished, and the total losses averaged 16.0, 12.9, and 16.5% of plasma volume at 20 min of exercise in the 20, 26, and 36°C environments, respectively. Figure 3 illustrates the thermal and circulatory data from one subject’s duplicate heavy exercise bouts in each of two environments. The data are plotted against esophageal temperature rather than time. Higher levels of T,, occurred in the warm environment, where the subjects were never able to achieve a thermal steady state during heavy exercise. Typically, cardiac output was relatively constant over the course of exercise and independent of T,, and Tsk. Stroke volume tended to fall with higher T,,

-16.5 % CO,

k1.1

40

experiments are given in each

MAX 70

100 ml-‘.

sated for the fall in SV with time. As with the lower exercise intensity, HR was considerably (P < 0.01) higher (~20 beats *mine’) in the 36OC environment, reaching 185 beats minD’ by the end of the exercise period. However, & was maintained at a level similar to that in cooler

S.E.E.

/

conditions. The elevated HR during heavier exercise in the heat compensated for the reduced SV, which stabilized around 100 ml* beat-’ over the final 10 min of exercise. Forearm blood flow was linearly related to T,, above the T,, threshold for vasodilation in all experiments

except the heavy exercise bout in the heat. The vasodilatory threshold was shifted to lower T,, by increasing ?& as we have shown previously (23), and changing TSk had no effect on the slope of the SkBF to T,, relation, as indicated earlier (23). From the pooled data of the three subjects, the average T,, threshold for vasodilation was 37.67”C in the 20°C environment, when TSk was 32.0°C. The slope of the SkBF to T,, relation was 13.4 ml l rein-1 g

sfk

6 t

I(-)() d-1. *c-l (r = 0.96). When ?i?& was 35.5”c, the vasodilatory threshold was 36.95*C and the slope was 13.1 rnl-min-lo 100 mPgo C-‘, independent of the exercise intensity and up to a SBBF of 15 ml0 min-’ . 100 ml-‘, which was achieved at a T,, of around 38.0°C. After this

juncture, during heavy exercise in the 36°C environment, the slope of the SkBF to T,, relation was reduced to 4.8 ml~min-l 100 ml-l. *c-l (r = 0.99) as T,, continued to l

FIG. 1. Forearm blood flow as a function of esophageal and mean skin temperatures during exercise. Data points are averaged from bestfit linear regression analyses of individual blood flow to T,, relations of 3 subjects in each condition (see text).

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

434

NADEL, %

-201

I 5

0

1 15

1 IO

EXERCISE

V02

TIME,

mox

I 20

MIN.

2. Change in plasma volume during moderate and heavy cise in 20 and 36°C environments. Data points are average values duplicate experiments in 3 subjects. FIG.

exerfrom

. -

%k

. -

‘sk

1

37.0

L 0/ 0 I

1

38.0

I

1

39.0

1 es 8 OC FIG. 3. Circulatory responses of a representative subject during heavy exercise in a 20 and 36°C environment as a function of esophageal temperature. Data are averaged from duplicate runs in each case.

values and was lower in the presence of higher levels of Tsk. Plasma volume decreased with increasing T,,, but this probably reflects the relation of both T,, and plasma volume to the duration of exercise. Forearm blood flow increased as a linear function of T,,, above the threshold for vasodilation, but when Tsk was high the linear increase was compromised above a T,, around 38.O"C. DISCUSSION

The most in the heat the tissues portion of

serious problem encountered during exercise is that of distributing sufficient blood flow to that demand it. Selectively directing a large the blood flow away from the contracting

CAFARELLI,

ROBERTS

AND

WENGER

muscles to the skin to prevent an excessive rise in internal body temperature would result in anaerobiosis and ultimately to an inability to continue exercise. Maintenance of adequate flow to the muscles at the expense of the skin limits the ability to transfer heat to the site of dissipation, resulting in progressive hyperthermia and ultimately to cessation of exercise. Nearly twenty years ago Brouha (4) proposed a potential solution to the problem of providing an adequate circulation to muscle and skin: cardiac output should be increased during exercise in the heat with respect to cooler conditions, thus the increased circulatory requirements could be met and exercise could be prolonged. Our data support the hypothesis of Brouha for conditions of relatively low demand. During exercise at 40% I702 max in a 36OC environment, we found cardiac output to be elevated by 1.3 lmin-’ above that in a 20 or 26°C environment. However, during conditions of higher demand, 70% VOW maxin a 36OC environment, it becomes increasingly difficult to augment cardiac output as heart rate approaches its maximum, whereas stroke volume is limited due to the reduced cardiac filling pressure attending the peripheral venous pooling and loss of volume from the vascular space. An adequate arterial blood pressure (and, indirectly, cardiac filling pressure) is presumably maintained in these conditions by a vasoconstrictor influence (baroreceptor reflex) superimposed upon the high vasodilator drive accompanying elevated skin and core temperatures. We have previously observed this (13), and Brengelmann et al. (3), using a different protocol, have convincingly demonstrated the existence of this phenomenon. At least partially as a consequence of this relative vasoconstriction in conditions of high demand, the heat transfer from the body core to the skin did not balance the heat production in our subjects, and the rate of internal temperature rise was not checked throughout the exercise bout. In the 20 and 26°C environments, T,, tended to level off between 38.1 and 382°C after 20 min of exercise, whereas in the 36OC environment T,, reached 3884°C at 19.3 min with no sign of approaching a steady state. The most thorough test of Brouha’s hypothesis was made about a dozen years ago by Rowell et al. (19). They found that cardiac output was unaltered during 15 min of mild and moderate treadmill exercise in the heat and reduced by approximately 2 1. min-’ during heavy exercise when compared to cooler conditions. In the latter condition, the muscle demands for oxygen delivery were presumably met by increased extraction. Rowell et al. (18) had previously demonstrated that splanchnic blood flow was 300-400 ml. min-’ lower during exercise in the heat than in cooler conditions and speculated that this difference might be sufficient to compensate for the increased demands from the skin (16). Nevertheless, even with a redistribution of the cardiac output, it seems that a reduction in output would be counterproductive to the needs of the organism. In a brief report, Damato et al. (6) found that during mild (3-4 met) supine exercise, cardiac output was elevated by around 2 lmin-’ during exercise in a 52OC environment when compared to a 26°C environment. In the present study, we also found that cardiac output was somewhat elevated during moderate (40% VOW max) exercise in the heat, as a result of the increased heart

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

CIRCULATORY

REGULATION

DURING

EXERCISE

rate without a decrement in stroke volume (see Table 2). The elevation in cardiac output averaged 1.3 Lmin-’ throughout the exercise period and was apparently sufficient to provide for the excess skin blood flow during the period. In the 36OC environment the average forearm blood flow was around 6 ml. min. 100 ml-’ of arm tissue higher than during exercise at 26OC. Assuming that this entire increment was limited to the skin (10, 15), and assuming that blood flow to the arm skin adequately represents skin blood flow to the entire body surface, the total excess skin blood flow in the heat amounts to 1.2 1. min-’ (where 200 cm2 of skin surface enclose an arm volume of 100 ml). Thus the increment of 1.3 Lmin-’ of cardiac output may be entirely accounted for. This may be a fortuitous calculation, since the reduction of splanchnit blood flow should also be accounted for. It is likely that both of these circulatory adjustments occur during moderate supine or semirecumbent exercise in the heat, thereby providing for both an adequate muscle perfusion and the increased core-to-skin heat transfer, allowing internal body temperature to arrive at the same steady state in the different thermal environments. It is possible that upright (treadmill) exercise does not permit such an increase in cardiac output, because there is a considerable hydrostatic pressure head to overcome in the upright position and therefore a potential difficulty in maintenance of cardiac filling pressure (17). This may explain the difference between the findings of Rowell et al. (19) and our own. In subjects performing heavy upright exercise in the heat, Rowell et al. (19) found a decrement in cardiac output when compared to cool conditions. Our data, taken from subjects performing heavy exercise in the semiupright position, showed cardiac output to be approximately the same in cool and warm conditions. Again, differences can be attributed to the different hydrostatic pressure heads between the upright and semiupright positions. The circulatory adjustment that provided for the maintenance of cardiac filling pressure, and therefore prevented the stroke volume from falling below 100 n&beat-’ in our study, was the vasoconstrictor response that was superimposed upon the dilator drive above a SkBF of 15 mlmin-’ 100 ml-‘. In a series of pilot experiments intended to work out techniques, we found this response to be lacking in a relatively unfit subject (who was not subsequently used in the actual procedure). The data of this subject are shown in Fig. 4. In the warm environment the forearm blood flow response was not compromised as T,, continued to rise and forearm flow reached rates considerably in excess of 20 ml*mirP JO0 ml-‘. After a certain point the peripheral demands exceeded the ability of the circulatory system to maintain an adequate venous return, stroke volume fell to approximately 70 ml. beat-’ and heart rate being close to its maximum and unable to increase further, cardiac output fell and the subject was not able to continue. It should also be noted that hyperthermia was not excessive in this subject. Failure was due to an inability to provide adequate blood flow to the contracting muscles. Thus the peripheral vasoconstrictor response of our three moderately fit subjects during heavy exercise in the heat allowed for adequate muscle perfusion at the expense of the temperature regulatory system. Upright l

435

1l 8 FIG. 4. Circulatory responses of 1 subject during heavy exercise in a 20 and 36°C environment as a function of esophageal temperature. There was no attenuating of the SkBF to T,, relation and stroke volume and cardiac output fell dramatically above a critical juncture.

exercise in the heat either requires this vasoconstrictor response to a greater extent, or proceeds with somewhat lower cardiac outputs owing to stroke volumes between 85 and 95 ml. beat-’ (19). Brengelmann et al. (3) concluded that the attenuation in slope of the SkBF to internal temperature relation, occurring around a T,, of 38.O”C in their study, was the result of a progressive vasoconstrictor influence associated with arterial blood pressure regulation and not the result of the attainment of maximum blood flows. In their conditions, local heating of one arm resulted in a further increase in SkBF in that arm, whereas SkBF in the other arm was not altered. In order to test their conclusion, we compared the SkBF to T,, relation derived from semiupright exercise bouts in the heat to the relation from bouts in the supine posture where the hydrostatic head in the peripheral veins would be minimized and venous return would be maximized. In duplicate experiments we found that the proportionality between SBBF and T,, was maintained in the supine position until exercise was terminated, with T,, reaching 38.5OC and SkBF exceeding 26 ml.min-‘o 100 ml-‘, rates considerably greater than achieved in the semiupright position. These data are shown in Fig. 5. In the supine position there apparently was no difficulty in maintaining arterial blood pressure and the compensatory vasoconstrictor response was not required. It is generally acknowledged that there is relatively little change in plasma volume during mild cycle ergometer exercise (1, 22) and some loss of plasma from the intravascular space during moderate and heavy exercise (1, 7, 9). Our data are in general agreement with the

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

NADEL,

436

CAFARELLI,

ROBERTS

AND

WENGER

the 36OC environment when SkBF exceeded a critical rate, approximately 15 ml. min. 100 ml-’ at a T,, around l SEMIUPRIGHT EX., 150 W isk * 35.5 “C 38.0°C, although these values are specific to the individ0 SEMIUPRIGHT EX, , 100 W f,, = 35.5 “c ual tested and the posture in which the exercise is undertaken. It is likely that both phenomena, a critical loss 0 of plasma volume and a critical rate of SkBF (and peripheral venous volume), are required to reduce arterial a blood pressure to the point at which reflex vasoconstriction is necessary. We found the changes in stroke volume to be independent of any changes in plasma volume at the moderate exercise intensity (Fig. 6). However, during heavy exercise the decrease in stroke volume corresponded well to the decrease in plasma volume at a given Tsk. Saltin (20) has previously demonstrated that reduced stroke volumes during exercise are a function of plasma volume when artificially lowered by dehydration prior to exercise. The effect of increasing Tsk by 3.5OC was to shift this relation toward lower stroke volumes. It is plausible 39.0 37.0 38.0 that the difference between these two relations, accounting for a difference in stroke volume of approximately 10 Tea R .C ml. beat-‘, can be ascribed to the difference in cutaneous FIG. 5. Forearm blood flow as a function of esophageal temperature venous volume between the cool and warm environduring supine and semiupright exercise in a 36°C environment. Data points are averaged from 1 subject’s duplicate exercise bouts in each ments. If this be the case it follows again that loss of fluid condition. from the intravascular space plus loss to the peripheral volume during exercise in a warm environment have an consensus. We found practically no change in plasma additive effect in the reduction of cardiac filling pressure, volume over 20 min of exercise at 40% 002 maxin a 20°C thereby limiting stroke volume. Physically fit individuals environment. However, exercise at this intensity in a may be better able to induce an effective baroreceptor 36OC environment was accompanied by an immediate reflex, decreasing the slope of the SkBF to T,, relation at (within 2-3 min) decrease in plasma volume of nearly 5%. high levels of T,, and reducing the rate at which the This decrease stabilized around 7% for the duration of dependent veins fill. Senay et al. (22) have presented the 20.min exercise bout. Heavy exercise (70% VOW max) evidence that heat-acclimated individuals are able to was accompanied by decreases in plasma volume which prevent a critical loss of plasma from the intravascular progressed to nearly 17% within 20 min of exercise. space by expansion of the plasma volume via retention Ambient (or skin) temperature had no effect on this of protein. A high initial plasma volume ensures an progression. These data are basically similar to the data of Harrison et al. (9), who followed plasma volume . changes during mild exercise (25% VOW max)in the heat % vo2 max 40 70 (42°C) and medium exercise (55% ~oZ max) in a warm 32.0 0 0 (30°C) environment. The large loss of fluid from the t 35.5 0 l vascular volume during heavy exercise is probably re'* lated to the increased osmotic pressure in the extravascular compartment, resulting from substrate mobilization and outward movement of protein, and to increased capillary filtration as a consequence of reductions in precapillary resistance. These changes are related to intensity of exercise and are independent of Tsk. The small changes in plasma volume during moderate exercise in the heat probably result from increased capillary filtration accompanying the reductions in cutaneous precapillary resistance. Because of the simultaneous measurement of plasma volume, skin blood flow, and cardiac output over the course of exercise in the present study, it is possible to relate changes in these variables to each other. It is clear, I I I I I for instance, that there is no direct relationship between 0 +5 -15 -10 -5 the decrease in plasma volume and the sharp attenuation PLASMA VOLUME, O/o CHANGE of the slope to T,, relationship. Plasma volume changes were independent of Q at the heavy exercise intensity, FIG. 6. Stroke volume as a function of plasma volume during moddecreasing by nearly 17% in both the 20 and 36°C envi- erate and heavy exercise in 20 and 36°C environments. Data points are ronments. The change in SkBF control was only seen in average values from 3 subjects. 0

SUPINE

EX.,

125

W

trk

= 35.7

‘C

r

P

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.

CIRCULATORY

REGULATION

DURING

437

EXERCISE

adequate venous return and stroke volume during exercise, even in the face of plasma volume decreases. In conclusion, the high circulatory demand from the skin in relatively fit individuals during moderate exercise in the heat is met by an increase in cardiac output, resulting from the elevation in heart rate that is itself a response of the arterial blood pressure regulating system. During 20 min of heavy exercise, as much as 17% of the plasma volume, amounting to as much as 500 ml in normal individuals, is lost to the extravascular space. Additionally, in the heat the drive to increase SkBF is high due to the high skin and core temperatures, and peripheral venous volume is also high. Despite these potential compromises to circulatory stability, cardiac output is maintained over 20 min of heavy exercise in the

heat at rates equivalent to cooler environments. Contributing to the maintenance of cardiac output are the elevated heart rate and a relative cutaneous vasoconstriction, which insures an adequate central circulating blood volume and ultimately prevents a precipitous fall in stroke volume. In the latter case, circulatory regulation has precedence over temperature regulation. This study was supported in part by National Institutes of Health Grants HL-20634, HL-17732, and ES-00123. A preliminary report of this study was presented at the Temperature Regulation Satellite Symposium of the 27th Congress of the International Union of Physiological Sciences, Lille, France, July, 1977. Present address of E. Cafarelli: Dept. of Physical Education and Athletics, York University, Downsview, Ontario, Canada. Received

21 August

1978; accepted

in final

form

5 October

1978.

REFERENCES 1. BEAUMONT, W. VAN, J. E. GREENLEAF, AND L. JUHOS. Disproportional changes in hematocrit, plasma volume and proteins during exercise and bed rest. J. Appl. Physiol. 33: 55-61, 1972. 2. BIGLAND-RITCHIE, B., H. GRAICHEN, AND J. J. WOODS. A variablespeed motorized bicycle ergometer for positive and negative work exercise. J. AppZ. Physiol. 35: 739-740, 1973. 3. BRENGELMANN, G. L., J. M. JOHNSON, L. HERMANSEN, AND L. B. ROWELL. Altered control of skin blood flow during exercise at high internal temperatures. J. AppZ. PhysioZ.: Respirat. Environ. Exercise PhysioZ. 43: 790-794, 1977. 4. BROUHA, L. Physiologic effect of work on the heart. In: The Heart in Itidustry, edited by L. J. Warshaw. New York: Hoeber, 1960,47104. 5. COSTILL, D. L., AND B. SALTIN. Changes in ratio of venous to body hematocrit following dehydration. J. AppZ. Physiol. 36: 608-610, 1974. 6. DAMATO, A. N., S. H. LAU, E. STEIN, J. I. HAFT, B. KOSOWSKY, AND S. I. COHEN. Cardiovascular response to acute thermal stress (hot dry environment) in unacclimatized normal subjects. Am. Heart J. 76: 769-774, 1968. 7. EKELUND, L.-G. Circulatory and respiratory adaptation during prolonged exercise. Acta PhysioZ. Stand. Suppl. 292, 1967. 8. FARHI, L. E., M. S. NESARAJAH, A. J. OLSZOWKA, L. A. METILDI, AND A. K. ELLIS. Cardiac output determination by simple one-step rebreathing technique. Respir. PhysioZ. 28: 141-159, 1976. 9. HARRISON, M. H., R. J. EDWARDS, AND D. R. LEITCH. Effect of exercise and thermal stress on plasma volume. J. AppZ. Physiol. 39: 925-931, 1975. 10. JOHNSON, J. M., AND L. B. ROWELL. Forearm skin and muscle vascular responses to prolonged leg exercise in man. J. AppZ. Physiol. 39: 920-924, 1975. 11. MITCHELL, J. W., J. A. J. STOLWIJK, AND E. R. NADEL. Model simulation of muscle blood flow and oxygen uptake during exercise transients. Biophys. J. 12: 1452-1466, 1972. 12. NADEL, E. R., J. W. MITCHELL, AND J. A. J. STOLWIJK. Differential thermal sensitivity in the human skin. Pfluegers Arch. 340: 71-76, 1973. 13. NADEL, E. R., C. B. WENGER, M. F. ROBERTS, J. A. J. STOLWIJK, AND E. CAFARELLI. Physiological defenses against hyperthermia of exercise. Ann. NY Acad. Sci. 301: 98-109. 1977-

14. NYLIN, G. The effect of heavy muscular work on the volume of circulating red corpuscles in man. Am. J. Physiol. 149: 180-184, 1947. 15. RODDIE, I. C., J. T. SHEPHERD, AND R. F. WHELAN. Evidence from venous oxygen saturation measurements that the increase in forearm blood flow during body heating is confined to the skin. J. Physiol. London 134: 444-450, 1956. 16. ROWELL, L. B. Human cardiovascular adjustments to exercise and thermal stress. Physiol. Rev. 54: 75-159, 1974. 17. ROWELL, L. B. Competition between skin and muscle for blood flow during exercise. In: Problems with Temperature Regulation During Exercise, edited by E. R. Nadel. New York: Academic, 1977, p. 49-76. 18. ROWELL, L. B., J. R. BLACKMON, R. H. MARTIN, J. A. MAZZARELLA, AND R. A. BRUCE. Hepatic clearances of indocyanine green in man under thermal and exercise stresses. J. AppZ. PhysioZ. 20: 384-394, 1965. 19. ROWELL, L. B., H. J. MARX, R. A. BRUCE, R. D. CONN, AND F. KUSUMI. Reductions in cardiac output, central blood volume and stroke volume with thermal stress in normal man during exercise. J. CZin. Invest. 45: 1801-1816, 1966. 20. SALTIN, B. Circulatory response to submaximal and maximal exercise after thermal dehydration. J. AppZ. PhysioZ. 19: 1125-1132, 1964. 21. SENAY, L. C., JR. Movement of water, protein and crystalloids between vascular and extra-vascular compartments in heat-exposed men during dehydration and following limited relief of dehydration. J. Physiol. London 210: 617-635, 1970. 22. SENAY, L. C., JR., D. MITCHELL, AND C. H. WYNDHAM. Acclimatization in a hot, humid environment: body fluid adjustments. J. AppZ. Physiol. 40: 786-796, 1976. 23. WENGER, C. B., M. F. ROBERTS, J. A. J. STOLWIJK, AND E. R. NADEL. Forearm blood flow during body temperature transients produced by leg exercise. J. AppZ. Physiol. 38: 58-63, 1975. 24. WENGER, C. B., M. F. ROBERTS, J. A. J. STOLWIJK, AND E. R. NADEL. Nocturnal lowering of thresholds for sweating and vasodilation. J. AppZ. Physiol. 41: 15-19, 1976. 25. WHITNEY, R. J. The measurement of volume changes in human limbs. J. Physiol. London 121: l-27, 1953.

Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on October 17, 2018. Copyright © 1979 American Physiological Society. All rights reserved.