human cardiovascular adjustments to rapid change in...

Human Cardiovascular Adjustmentsto Rapid Changes in SkinTemperature during Exercise

By Loring B. Rowell, Ph.D., John A. Murray, M.D.,

George L. Brengelmann, Ph.D., and Kenneth K.. Kraning, I I , Sc.D.

ABSTRACTIn 11 normal men, central circulatory responses were measured while skin

temperature was changed in a square-wave pattern during uninterruptedexercise (26% to 64% maximal oxygen consumption). Skin temperature waschanged at 30-minute intervals, beginning at 32°C. On raising it to 38.2°C atlow oxygen consumption (Vo2), cardiac output increased 2.5 liters/min, andcentral blood volume, aortic mean pressure, and stroke volume fell (7%, 7%, and11%, respectively). Right atrial mean pressure fell 2.2 and 2.3 mm Hg duringcontrol and heating periods, respectively. All variables returned to controllevels when skin temperature was reduced toward 26.9°C. Raising it to 40°Creproduced these changes with a more clear-cut drop in right atrial meanpressure. Results indicated reduced peripheral venous tone and cutaneouspooling of blood during heating and rapid reversal on cooling. On raising skintemperature to 38.7°C at high Vo2, cardiac output increased 19% (3.1liters/min), stroke volume decreased 14%, and central blood volume rose slightly.Aortic mean pressure fell during the control period and was maintained or roseduring heating periods. On cooling, central blood volume and stroke volumerose, cardiac output remained elevated, and aortic mean pressure fell. Increasesin cardiac output during heating were related to skin temperature and not toVo2 or body temperature. At high Vo2, circulatory adjustments favor metabolicrather than thermoregulatory demands.

ADDITIONAL KEY WORDS cardiac output stroke volumeperipheral vascular resistance heart rate central blood volumecentral venous pressure aortic and arterial blood pressurecirculatory regulation peripheral circulation

• Previous studies of human cardiovascularadjustment to superimposition of competitive

From the Departments of Medicine, Physiology,and Biophysics, University of Washington School ofMedicine and School of Nursing, Seattle, Washington98105.

This work was supported in part by U. S. PublicHealth Service Grant HE-09773 from the NationalHeart Institute. A part of this study was conductedthrough the Clinical Research facility of theUniversity of Washington, supported by the NationalInstitutes of Health (Grant FR-37).

Dr. Rowell was an Established Investigator of theAmerican Heart Association and was supported by theWashington State Heart Association.

A preliminary report of this work was presented atthe annual meeting of the American Heart Associa-tion, November 23, 1968, Miami, Florida.

Received August 19, 1968. Accepted for publicationMarch 25, 1969.

thermal and metabolic stresses have dependedupon manipulation of environmental tempera-ture rather than skin temperature. But, wherenatural conditions are simulated, skin temper-ature may vary throughout the exposure (1).It will also vary over the body surface andmay well be reduced, the more severe theexercise, due to an increased sweating re-sponse (1). In short, skin temperature is nottightly controlled by exposing subjects tocontrolled environments, but it is important inboth direct local control and reflex control ofthe vasoactive state of the skin (2). Accord-ingly, it is important to control this variablewhen attempting to understand the mecha-nisms governing human cardiovascular re-sponses to metabolic and thermal stresses.

Circulation Research, Vol. XXIV, May 1969 711

by guest on April 22, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

712 ROWELL, MURRAY, BRENGELMANN, KRANING

TABLE 1

Characteristics of Subjects and Oxygen Uptake While Wearing Water-Perfused Suits

Subject

P.S.Ki.F.W.D.R.G.Z.B.B.M.H.CM.G.T.P.H.P.K.

Age(yrs.)

2321272223212821222722

Height(cm)

174185180175178185183175182178177

Weight(kg)

72.783.672.464.571.890.278.680.676.768.972.4

Max. Vo»(L/min)

3.574.163.803.053.614.333.584.535.153.863.73

Worklo:(% grade)

000000

7.57.57.57.57.5

id

(mph)

2.72.72.72.73.53.53.53.53.53.53.5

O2

(liters/min)

0.941.261.000.941.321.492.292.372.291.982.28

uptake*(% max. Vo»)

26.330.326.330.836.634.464.052.344.551.361.1

'Oxygen uptake wasoccasion.

Our experimental design was based oncontrolling skin temperature as closely aspossible and manipulating it in a square-wavepattern between extremes during uninterrupt-ed exercise. It was hoped that this wouldenable (1) production of substantial changesof skin blood flow superimposed upon severallevels of constant metabolic rate and, pre-sumably, constant flow to exercising musclesand (2) possible clarification of the relativeroles of central and skin temperatures in thecardiovascular response to thermal and exer-cise stresses. Further, we hypothesized thatsudden, marked elevation of skin temperatureafter establishment of the normal circulatoryresponse to exercise would elicit cardiovascu-lar adjustments in both central and peripheralcirculations which heretofore have beenmissed. This hypothesis was based on thefinding of Bevegard and Shepherd (3, 4) thaielevating skin temperature abolishes the nor-mal increments in the tone of resistance andcapacitance vessels of the limbs which developin response to exercise.

Procedures and MethodsThe subjects were 11 university students, 21

to 28 years olds. None was acclimatized to heatnor had any been in active physical training. Onlytwo men (B. B. and P. K.) were physicallyactive. Their physical characteristics are de-scribed in Table 1. All volunteers were verythoroughly familiarized with all details of the

experiment. They were given a thorough physicalexamination, including electrocardiogram, chestx-ray, and routine laboratory examination of bloodand urine. All were in excellent health.

Studies were conducted during the wintermonths in a room maintained at 24° ± 2°C.Maximal oxygen uptake was determined in eachsubject, using either the method of Taylor et al.(5) or multistage exercise on a treadmill toexhaustion (6). The subjects were divided intotwo groups. Six subjects worked at low workloadswhich required 26.3% to 38.9% of maximal oxygenuptake (Table 1). The work consisted of walkingon the treadmill at 0% grade at either 2.7 or 3.5mph. The five others performed heavy exercise(7.5% grade, 3.5 mph) which required 51.3% to64.0% of maximal oxygen uptake. The averageenergy requirements of these three levels ofexercise were 14.5, 17.2, and 29.8 ml O2/kg/min., respectively.

Before the final experiments in which subjectswere catheterized, all had performed the entireexperimental procedure at least once. Thesepreliminary studies served to standardize thesubjects' thermal and metabolic responses and tofamiliarize both the subjects and investigatorswith the former's responses and capacities. Thecollection of expired air was omitted from thefinal experiment to eliminate any effect of therespiratory apparatus on right atrial pressure.Previous studies have shown oxygen uptakeduring exercise to be unaffected by the presenceof intravascular catheters (7, 8). Briefly, 105minutes of continuous exercise at a fixed loadwere divided into three 30-minute periods and afourth and final 15-minute period. During theseperiods the temperature of water perfusing aspecial garment was 21°C (period 1); 45°C and

Circulation Research, Vol. XXIV, May 1969



ownloaded from


CIRCULATORY RESPONSES TO HEATING AND COOLING 713

Ts °C

40 0%, 2.7 3.5 mph

33

36

20 40 60M i nutes

100

Ts "C

40 7.5%, 3.5mph

20 40 60Minutes

80 100

FIGURE 1

Skin temperatures (Ts) averaged for each subject from five points on the body surface. Dataon the left are from six subjects at the low workloads (0% grade on the treadmill). F.W. andB.B. on the left had a lower water temperature (40°C) perfusing the suit from 30 to 60minutes (broken lines). Data on the right are from five subjects at the high workload (7.5%grade).

40°C at low and high works, respectively (period2); 10°C (period 3); and 50°C (period 4). Intwo subjects working at the lowest workload (F.W. and 13. B.), water at 40°C was used duringperiod 2. The hydraulic plus thermal lag at thesuit-skin interface at the time of temperaturechanges is illustrated in Figure 1. Use of the twodifferent perfusion temperatures during period 2provided similar skin temperature at the high andlow workloads. Skin temperature was lower in thetwo subjects (F. W. and B. B.) whose suits wereperfused with water at 40°C during period 2(Fig. 1).

For the final experiment, subjects reported tothe laboratory at noon. All had eaten breakfast atleast 3 hours earlier. A Teflon catheter 70 to 80cm long (18-gauge, thin-wall) was introducedpercutaneously into the left brachial artery usinga modified Seldinger technique. This catheter waspositioned under fluoroscopic guidance in thedescending aorta just below the origin of the left


subclavian artery. A no. 7 double-lumen catheter100 cm long was introduced through a smallincision into the left antecubital vein. Thiscatheter was positioned in the middle of the rightatrium.

The plan of the experiment was to measureaortic blood pressure and cardiac output every 5minutes. These measurements were made every 2minutes for 6 minutes after each change in suittemperature. Thereafter, the regular 5-minuteschedule was followed. Heart rate, right atrialmean pressure, and blood, rectal, and skintemperatures were measured continuously. In fivesubjects (three at the low and two at the highworkloads), 1 ml of arterial blood was withdrawnat the tenth and final minute of each thermalperiod for lactate determination. Forty-five milli-liters of arterial blood were drawn at thetwentieth minute of each period except the lastfor calibration of the densitometer. A total of from500 to 900 ml of 0.5N heparinized saline was



ownloaded from



slowly infused into the aortic catheter betweendetermination of cardiac output and bloodpressure. At the end of the experiment, the suitwas perfused with cold (10°C) water while thesubject recovered.

Control of skin temperature was accomplishedby means of a suit1 of form-fitting coarse nylonmesh lined with small vinyl tubes spaced 1 to 2cm apart. The suit and tubing covered the entirebody surface except the face. The weight of thissuit filled with water plus the rubberized nyloncoverall was 3 kg. The purpose of the coverallwas to minimize evaporative cooling and unevenheating of the skin.

Skin temperature was measured from fivethermocouples (26-gauge, insulated, copper-con-stantan wires) attached to three points on thesubject's back and two on the front of the torso.Sites of placement were the same on all subjects.These thermocouple junctions were 2 cm longand placed on the skin at a right angle to thetubing in the suit. This meant that only a smallfraction of the thermocouple junction was ever indirect contact with the water-filled tubing. Thefive junctions were connected in parallel toaverage their outputs (resistances of the five leadswere equal). Temperatures were referenced to acold junction maintained at 0.0°C in a stirred icebath. Thermocouple voltages were electricallyoffset to 33 °C by a precision voltage source andsuitably amplified.2

The temperature of right atrial blood wasmeasured from an insulated thermocouple junc-tion (40-gauge, teflon-insulated, copper-constant-an wire) positioned at the tip of the longerlumen of the double-lumen catheter, but notoccluding it. These wires were sealed into one armof a Y connector at the hub of the lumen. Thereference junction was held at a constanttemperature of 38.5° ± 0.01°C. Thermocoupleoutput was amplified by a Medistor Model A-60amplifier. The other arm of the Y connector wasconnected to a pressure transducer.

Water was perfused through the suits at a

1 Generously lent by NASA Manned Space CraftCenter, Houston, Texas.

2Separate studies were made on six subjects toexamine whether temperatures of the extremities andthe torso were similar and the general magnitude ofvariation over the body surface. Simultaneouslymeasured outputs from the five paralleled leads on thethorax and five paralleled leads on the legs agreed towithin 0.5° to 1.0°C and 0.0° to 0.5°C during thecontrol and heating periods, respectively. In three menthe individual outputs from ten thermocouplessituated between the suit tubing on the anterior andposterior surfaces of arms, legs, and torso agreedwithin 1°C with the simultaneously measured outputfrom the five paralleled leads on the torso.

constant rate of 8 liters/min from one of two 40-liter, constant-temperature baths. Temperature

W B 21"

120-

110-

100-

40 60MINUTES

100

FIGURE 2

Circulatory and temperature data from subject P.S.during mild exercise at 2.7 mph (0% grade). Themagnitude, direction, and variability of this subject'sresponses typify those of the five other subjects studiedat the two lower workloads. The temperatures ofwater perfusing the suit are shown in °C at the topof the columns separating the 30-minute periods. TR

= rectal temperature in °C. TB = temperature ofright atrial blood in °C. CO = cardiac output in liters/min. HR = heart rate in beats/min. SV — strokevolume in ml. CBV = central blood volume in liters.AoMP = aortic mean pressure in mm Hg. RAMP =right atrial mean pressure in mm Hg. TPR = totalperipheral resistance in arbitrary units, mm Hg/liter/min.




ownloaded from



was changed almost instantaneously by switchingto a second bath previously set at the desiredtemperature.

Techniques for calibration of the densitometer(modified Gilford 103IR) and calculation ofcardiac output, mean transit time, and centralblood volume were identical to those describedearlier (8). Calibrations for the concentration ofindocyanine green in blood taken at differenttimes during each experiment agreed within 1% to3% within any subject. The densitometer hadconstant sensitivity to indocyanine green withbackground levels of dye up to 9 mg/liter.Background levels of dye did not exceed 5mg/liter during the study. Five milligrams of dyewere rapidly injected (in 0.5 seconds) from acalibrated 1-ml syringe electrically equipped torecord the beginning and end of injection.Injection was by displacement from the side hole,located 2 cm behind the tip of the double-lumencatheter. This eliminated any prolonged coolingof the thermocouple junctions (located at the tipof the other lumen) by the cool dye. Aortic bloodpressure was recorded with a Statham P23Gbpressure transducer. The entire manometric system(catheter, gauge, Honeywell 131-2C carrier ampli-fier, and 24-Hz galvonometer) had unity gain to20 Hz with 64% of critical damping. Right atrialmean pressure was measured with a StathamP23Db transducer and Honeywell 131-2C carrieramplifier and a previously described (9) electricalaveraging circuit (critically damped, 0.16 Hz, lowpass active filter). Both manometric systems werestatically calibrated with a mercury manometer(aortic pressure) or a water manometer (venouspressure) at multiple levels before and after eachexperiment. Zero pressure was referred to aleveling bottle set at the level of the subject'sfourth interspace. Heart rate was continuouslymeasured using a heart-rate meter (QuintonInstruments) which was calibrated at eachexperiment with a pulse generator and wasaccurate to within ±1 beat/min.

All data were recorded at full scale on aHoneywell 1508 Visit-order and on a Sanborn3905 FM tape recorder. Calibrations for all datachannels were repeated immediately after the endof the experiment. Electrical drift in all channelswas negligible. All data, except cardiac output,which was measured by rectangulation from theoriginal records (10), was processed by handfrom tape-recorded data replayed at very slowrecorder paper speeds. Aortic mean pressure wasmeasured from electrically averaged (as above[8]) pulsatile pressure replayed from magnetictape.

Blood lactate was determined using the methodof Strom (11) and previously described proce-dures (7).


TR.TB 21*

38

60MINUTES

FIGURE 3

Average circulatory and temperature data from sixsubjects during mild exercise (0% grade). All dataoccurring within any given 5-minute interval of timehave been averaged. For comparison with rectaltemperature (TB) and temperature of right atrialblood (TB), average skin temperature (Ts) is shownby a dashed line enclosing a shaded region. Otherabbreviations as in Figure 2.

ResultsRESPONSES AT LOW OXYGEN CONSUMPTIONS

Data from one of the six subjects studied atlow oxygen consumption3 (0.9 to 1.5 liters/

3Oxygen consumption at all workloads remainedconstant throughout an earlier performance of thesame procedures by these subjects.



ownloaded from



min) are shown in Figure 2.4 The averageresponses for this group are shown in Figure3. Qualitatively, the responses to heating orcooling for any subject were strikingly repre-sentative of the entire group.

Cardiac output, heart rate, stroke volume,central blood volume, and the temperature ofright atrial blood were relatively constantduring the first 30-minute period (referred toas the control period) during which skintemperature averaged 32.9°C (33.9° to 32.3°C). Aortic mean pressure, right atrial meanpressure, and total peripheral resistance fellslightly during the control period. On heatingthe skin to 38.3°C (average at 60 minutes),cardiac output and heart rate rose, whilestroke volume, central blood volume, aorticmean pressure, right atrial mean pressure, andtotal peripheral resistance all fell. The tem-perature of right atrial blood fell about 0.5°Cdue to an increase in the amount of bloodreturning to the right atrium from the skinwhich at that time was well below the rectaltemperature (Fig. 1). The changes in eachvariable at the end of any 30-minute periodare given in appropriate units and as percent-ages of that variable at the end of the previousperiod (or as percent of the entire controlperiod at the end of the second 30-minuteperiod) in Table 2.

When skin temperature was lowered at 60minutes, there was a rapid fall in the heartrate and the temperature of right atrial blood.Cardiac output fell after a 5-minute delay andreturned to control values at 90 minutes.Stroke volume, central blood volume, aorticmean pressure, right atrial mean pressure, andtotal peripheral resistance rose rapidly andovershot control values in four men (Fig. 2).Five to ten minutes after cooling began, thetemperature of right atrial blood began to riseagain. Undoubtedly, this was due to theaccumulation of metabolic heat, undissipatedbecause of cutaneous vasoconstriction. At 90minutes, the rapid rise in skin temperature

4Raw data from the other subjects were deletedfrom this paper by the editors to conserve space.Copies of these data are available from the author onrequest.

from 26.9°C toward 39.1°C brought a morerapid, and in three subjects a greater, responsein cardiac output. For other variables, re-sponses were similar to the first rise in skintemperature.

Measurements of the central blood volumeshould not be construed to represent accurateestimates of intrathoracic blood volume butrather to reliably describe directions of changein this volume. Since the aortic sampling sitewas distal to the subclavian arteries, theincrease in blood flow to the arms on heatingand decrease on cooling caused a falseincrease and decrease, respectively, in truemean transit time to the sampling site. Sincethese changes were of opposite sign to the truechanges, the decrease in central blood volumewith heating and the increase with coolingwere underestimated. This has been discussedpreviously (8).

Although right atrial mean pressure cannotbe considered as effective right ventricularfilling pressure because intrathoracic pressurewas not subtracted, the changes in the formermust have closely, paralleled the changes ineffective filling pressure. Ventilation rose only15% above control values by the end of heatingand was still elevated 10% between theseventieth and seventy-fifth minutes of coolingafter right atrial mean pressure had risen.Normally, right atrial mean pressure remainsalmost constant up to maximal work despitemarked rises in ventilation, while reductions inintraesophageal pressure are comparativelyminor (12).

Quantitatively, the variation among thesesubjects appeared to be related primarily tothe different levels of skin temperaturereached during the first 30 minutes of heating.The smaller changes in all variables duringheating in subjects F.W., D.B., and D.R.appear to result from their lower skintemperature. Raising perfusion temperature to45°C at the low workload in the remainingsubjects gave an almost equal skin tempera-ture and equal increase in cardiac output atthe two levels of oxygen consumption (Fig.

1)-Circulation Research, Vol. XXIV, May 1969



ownloaded from


JO

1o 30

TAB

LE 2

S Su

mm

ary

of A

vera

ge C

ardi

ovas

cula

r C

hang

es i

n Si

x Su

bjec

ts a

t L

ow O

xyge

n U

ptak

es

Ave

rage

s w

ere

take

n at

the

beg

inni

ng a

nd e

nd o

f ea

ch 3

0-m

inut

e pe

riod

(i

ndic

ated

by

the

tem

pera

ture

of

the

wat

er p

erfu

sing

th

e su

it w

hich

is

giv

en

atth

e to

p of

eac

h co

lum

n) a

nd t

he f

inal

15-

min

ute

peri

od.

"Cha

nges

(A

) ar

e gi

ven

in a

ppro

pria

te u

nits

(up

per

num

ber)

and

in

perc

ent

(low

er n

umbe

r)

whi

chw

as c

alcu

late

d fr

om t

he r

atio

of

the

chan

ge t

o th

e va

lue

at t

he b

egin

ning

of

that

per

iod.

Ini

tial

and

fin

al n

otat

ions

do

not

appl

y to

thi

s co

lum

n.

o I/I

Car

diac

out

put

(lit

ers/

min

)

Hea

rt r

ate

(bea

ts/m

in)

Stro

ke v

olum

e(m

l)

Cen

tral

blo

odvo

lum

e (l

iter

s)

Aor

tic m

ean

pres

sure

(m

m H

g)

Rig

ht a

tria

lm

ean

pres

sure

(mm

Hg)

Tot

al p

erip

hera

lre

sist

ance

(m

mH

g/li

ter/

min

)

init

ial

fina

l

initi

alfi

nal

initi

alfi

nal

initi

alfi

nal

initi

alfi

nal

init

ial

fina

l

initi

alfi

nal

21'C

11.0

11.0

105

107

105

105

1.18

1.20

100

94 6.0

3.8

9.4

8.8

A 0* 0%*

+2

+ 1.3

%

0 0

+.02

+ 1.7

%

-6

-6%

-2.2

-36

%

-0.6

-6.4

%

40°C

to

45

°C

11.0

13.5

107

146

105 94 1.

201.

12

94 87 3.8

1.5

8.8

6.7

A

+2

.5+2

2.7%

+3

9+3

6.4%

—11

-10

.5%

-0.0

8-6

.7%

-7

-7.4

%

-2.3

-60

%

-2.1

-23

.9%

io°c

13.5

11.2

146

107

94 106

1.12

1.22

87 98 1.5

4.5

6.7

8.9

A

-2.3

-17

%

-36

-26

.7%

+ 12

+12.

8%+0

.10

+8.9

%

+ 11

+12.

6%

+3.

0+2

00%

+2.2

+32.

8%

50°C

11.2

14.3

107

151

106

98 1.20

1.18

98 87 4.5

2.3

8.9

6.2

A

+3.

1+2

7.7%

+44

+41.

1%

-8 -7.5

%-0

.2-1

.7%

-1

1

-11

.2%

-2.2

-49

%

-2.7

-30

.3%

H O X m MING Z o n o o o

by guest on April 22, 2018http://circres.ahajournals.org/Downloaded from


00

TA

BL

E

3

Sum

mar

y of

Ave

rage

Car

diov

ascu

lar

Cha

nges

in

Fou

r Su

bjec

ts a

t H

igh

Oxy

gen

Upt

akes

21°C

40° C

io°c

50"C

Car

diac

out

put

initi

al

16.3

(lit

ers/

min

) fin

al

16.3

19

.4

Hea

rt r

ate

initi

al

127

+1

3 14

0(b

eats

/min

) fi

nal

140

+11

%

181

Stro

ke v

olum

e in

itia

l 12

3(m

l)

fina

l 12

3 10

9

Cen

tral

blo

od

init

ial

1.28

volu

me

(lit

ers)

fi

nal

1.28

1.

33

Aor

tic m

ean

init

ial

104

—15

89

pres

sure

(m

m H

g)

fina

l 89

-1

4.4%

92

Tot

al p

erip

hera

l in

itial

6.

6 —

0.8

5.8

resi

stan

ce (

mm

fi

nal

5.8

—12

%

4.9

Hg/

lite

r/m

in)

+3.

1+1

9%+

41+2

9.3%

-17

-13.

8%

+0.

05+3

.9%

+3

+3.4

%

-0.9

-15.

5%

19.4

17.9

181

148

109

122

1.33

1.41

92 108

4.9

5.2

-1.5

-7.7

%-3

3-1

8.2%

+13

+11.

9%

+0.0

8+6

%

+16 +17.

4%

+0.

3+6

.1%

17.9

20.1

148

187

122

108

1.41

1.34

108

96 5.2

4.6

+2

.2+

12.3

%

+3

9+2

6.4%

—14

-11.

4%

-0.0

4—

2.8%

—12

-11.

1%

—0.

6—

11.5

%

|- T

he r

esul

ts f

rom

sub

ject

P.K

. w

ere

omit

ted.

Pre

sent

atio

n of

dat

a is

as

desc

ribe

d in

Tab

le 2

exc

ept

for

card

iac

outp

ut,

stro

ke v

olum

e, a

nd c

entr

al b

lood

vol

ume

I' w

here

ave

rage

s fr

om t

he i

nitia

l pe

riod

(21

°C)

wer

e us

ed f

or c

ompa

riso

n w

ith

the

end

of t

he s

econ

d pe

riod

(4

0°C

).

I 79 W 50 Z z I

by guest on April 22, 2018http://circres.ahajournals.org/Downloaded from



39.5

38.5

37.5

JR.TB 21°

60MINUTES

90

FIGURE 4

Average circulatory and temperature data from joursubjects during heavy exercise (7.5% grade, 3.5 mph).Data from P.K. were omitted (see text and Fig. 6).Abbreviations as in Figure 2.

RESPONSES AT HIGH OXYGEN CONSUMPTIONS

Average responses of four subjects at thehigher oxygen consumptions (2 to 2.4 liters/min) are shown in Figure 4 and Table 3. Re-sponses of a representative subject are shownin Figure 5. The data from subject P.K. wereomitted from this summary because his re-


13Oi

120-

110-1

100-

20 40 60 80MINUTES

100

FIGURE 5

Circulatory and temperature data from subject P.H.during heavy exercise. This subject's responses typifythose of three other subjects at this workload. Notethe marked fall in aortic mean pressure (AoMP) asthe skin was cooled. Abbreviations as in Figure. 2.

sponses differed so from those of the other fourmen. Despite a 1.5 to 2 times greater rate ofendogenous heat production and higher tem-peratures of right atrial blood and rectum inthese four men, the increase of 3.1 liters/ min incardiac output (range 2.9 to 3.4 liters/min) dur-ing heating was similar to that seen at the lowerlevels of work (range 1.5 to 4.4 liters/min).The correlation between the increase in cardiacoutput and the increase in skin temperaturehad a coefficient of 0.74 for all measurementsbetween 30 and 60 minutes (except forsubject P.K.) at all levels of work. The majordifferences in the circulatory responses at thetwo levels of oxygen uptake during heatingwere the tendency for aortic mean pressure



ownloaded from



T R / % 2 1

12OI

21°C 45°C 1O"C 50°C

20 40 60MINUTES

8 0 100

FIGURE 6

Circulatory and temperature data from subject P.K.during heavy exercise (7.5% grade). Responses dis-tinguishing P.K. from the others were: extremely highheart rate, a drop in cardiac output and a very largefall in stroke volume whenever the temperature ofright atrial blood and skin temperature rose, and aprogressive and marked fall in aortic mean pressurethroughout the experiment with only a minor responseto cooling (see text). Abbreviations as in Figure 2.

and central blood volume to increase alongwith constant total peripheral resistance at thehigher oxygen uptake; the rise in heart rateand fall in stroke volume were only slightlygreater at this level. On sudden cooling, therewas always a marked drop in aortic meanpressure in contrast to the rise always seen atlow oxygen uptake (except subject G.Z. whoalso showed an initial fall).

60MINUTES

FIGURE 7

120

Arterial blood lactate concentrations in two men (P.K.and P.H.) during heavy exercise (7.5% grade) and inthree men during light exercise (D.R. and Ki, 0%grade, 2.7 mph and G.Z., 0%, 3.5 mph).

All subjects, except P.K., experienced veryrapid and dramatic amelioration of symptomsof exhaustion after 10 minutes of cooling.Subject P.K. (Fig. 6) was unable to meet thecirculatory demands added by increased heatload. He showed an almost continuous fall inaortic mean pressure which was reversed onlyslightly by cooling. His drop in stroke volumefrom 120 to 83 ml/beat was the largest in thestudy. We do not know the reason for hisinability to maintain aortic mean pressure andtotal peripheral resistance. Our expectationwas that his performance would improve, as itdid in the other men at the high workload,when suit-perfusion temperature was reducedfrom 45°C to 40°C after the initial trial run.At the fifty-fourth minute of exercise P.K.experienced lightheadedness, and suit temper-ature was immediately reduced to 10°C. Aftera period of temporary symptomatic relief, heagain experienced lightheadedness, along withslight visual blurring at the sixty-fourthminute. As we were about to end theexperiment, aortic mean pressure rose and thesubject experienced relief. Because of theseproblems, the final heating period was re-duced to 8 minutes. His recovery from




ownloaded from



extreme fatigue after the experiment followedan entirely normal course over the next 24hours. Subject P.K. was apparently overdriv-en by thermal stress. Because of the exhaus-tion of chronotropic reserve and depressedstroke volume, cardiac output could be neitherincreased nor maintained. Similar responseswere observed in a previous study (8).

All subjects, including P.K., were wellhydrated during the experiment. The adminis-tration of 0.5N saline and the suppression ofsweating caused by the impermeable outersuit kept net water loss in these men down to0.5% (0.1% to 1.2$) of body weight at thelower oxygen uptakes and to 0.9% (0.6% to1.5%) at the higher levels.

Arterial blood lactate concentration showedno marked elevation suggestive of reducedmuscle blood flow during heating (Fig. 7).

Discussion

During moderate exercise, cardiac outputand heart rate increased with heating anddecreased with cooling of the entire bodysurface. Stroke volume and aortic pressuredecreased with heating and increased withcooling. During more severe exercise, al-though similar changes occurred in cardiacoutput, heart rate, and stroke volume, aorticmean pressure did not change during heating.When the skin was cooled, aortic pressure felltransiently, followed by a return to controllevels, compared with a sustained increaseduring milder exercise.

There is strong evidence that heating andcooling the skin caused marked changes in theperipheral circulation. Clearly, the rapid fallin temperature of right atrial blood beforecardiac output rose at the lower workloadsindicated a shift in blood flow from deep tosuperficial veins (in still relatively cool skin),which were dilated by heating. There areseveral reasons for assuming that the subse-quent increase in cardiac output was directedprimarily to skin. The progressive decrease inthe difference between temperatures of theskin and right atrial blood to an average of0.2°C at 60 minutes indicates an increase inskin blood flow. At rest, muscular flow in the


limbs is not increased with heating (10), andthe rise in cardiac output of 3 to 4 liters/minparallels the rise in skin blood flow (13).There are data suggesting that muscle bloodflow may even be reduced during exercise in ahot environment since arterial lactate concen-trations tended to be higher than normal (7,14) while oxygen uptake was maintained at(7, 8) or slightly below control values (14,15). However, our subjects showed no rise inblood lactate concentration.

Visceral organs are the only others likely toreceive any substantial fraction of an incre-ment in cardiac output. During heating, renal(16) and splanchnic blood flow (7) fall belowlevels normally found at the same intensity ofexercise in cool environments. Thus, totalcutaneous blood flow may exceed the incre-ment in cardiac output by quantities equal toor greater than those redistributed fromvisceral organs.

The observed responses of cardiac output,stroke volume, and heart rate may be theresultant of many factors, including reflexesfrom the skin (17) and humoral responseselicited by changes in temperature of the skinand right atrial blood per se. However,focusing on the periods immediately followingchanges in skin temperature reveals that thecentral responses were at least partly due toalterations in the peripheral circulation. Sincethe magnitude and direction of some changesdepended upon the level of metabolism,results from the high and low levels arediscussed separately.

RESPONSES AT LOW OXYGEN UPTAKE

We interpret the changes we observed asfollows. Raising skin temperature abolishedthe level of peripheral venoconstriction whichis normally increased in proportion to theseverity of exercise (3, 4). Recent evidencesupports such a response to heating the skin inman (3, 4) and blood in dogs (18). Alongwith dilatation of the resistance vessels to theskin, the volume of the more compliantcutaneous venules and veins presumablyincreased suddenly. This caused a transient lagin venous return so that the output of the leftventricle exceeded that of the right, causing a



ownloaded from



sudden reduction in central blood volume.Right atrial mean pressure either continued tofall or fell further as heating began, withfurther gradual pooling of blood in thecutaneous venous system. The increase incardiac output was not adequate to compen-sate for the fall in total peripheral resistance,and aortic mean pressure fell further. Ofcourse, other interpretations for these changescan be offered, but ours is strengthened by theresponses to cooling as discussed below.

At both the low and high levels of oxygenuptake, interpretation of changes in responseto the initial elevation in skin temperature wascomplicated by the presence of drift in severalvariables (Fig. 3). Ekelund and Holmgren(19) found that if work was prolonged in acool environment, systemic and pulmonaryarterial mean pressures, right ventricular end-diastolic pressure, and stroke volume gradual-ly declined. The authors suggested that thesechanges might be due to gradual displacementof blood into peripheral veins consequent toperipheral vascular adjustments for thermo-regulation (19). In our study, similar changescould be rapidly reversed by cooling the skin.Also, when skin temperature was maintainedat low levels during the sixtieth to ninetiethminute, this downward drift was essentiallyeliminated. Thus, these changes do appear tobe associated with thermoregulatory adjust-ments in the peripheral circulation.

Elimination of this drift during the coolingperiod (60 to 90 minutes) established a morestable baseline for measuring the responses tothe second elevation in skin temperature (at90 minutes). Here, reductions in aortic meanpressure, right atrial mean pressure, and totalperipheral resistance were more obvious. Aftera 5-minute delay, stroke volume fell as beforebut somewhat faster.

We interpret the hemodynamic response tolowering skin temperature as follows. The fallin skin temperature caused cutaneous veno-constriction. There are also previous reports ofvenoconstriction with lowering of skin tem-perature (18, 20). Blood pooled in dilatedcutaneous venules was rapidly displacedcentrally. Right ventricular output momentari-

ly exceeded that of the left, resulting in thesudden increase in central blood volume andright atrial mean pressure.

Cutaneous vasoconstriction appeared tooccur gradually over about 10 minutes (60 to70 minutes) as indicated by the time course ofchanges in the temperature of right atrialblood, aortic mean pressure, cardiac output,and total peripheral resistance.

Cutaneous vasoconstriction appears to havebeen essentially complete after the tenthminute of cooling since the temperature ofright atrial blood then began to rise again,indicating reduced transcutaneous heat trans-fer.

Without data on blood catecholamines, pH,Pco2, etc., no comment can be made regardinga number of possible mechanisms which mayhave contributed to the changes in strokevolume. The similarity in the stroke volumeresponses in subjects who showed either verysmall or very large elevations in temperatureof right atrial blood suggests that myocardialtemperature was not an important factor. Theperipheral displacement of blood volume andfall in right atrial mean pressure could wellhave contributed to the fall in stroke volume.

It is important to note that the bloodvolume shifts are unlikely to be of primarycardiac origin. Were such the case, thetemporary imbalance between right and leftventricular output, necessary to decrease cen-tral blood volume, would have to have beencaused by differential changes of right and leftventricular contractile forces.

RESPONSES AT HIGH OXYGEN UPTAKE

Despite much greater increments in rectaltemperature and temperature of right atrialblood, the direction and magnitude of changesin cardiac output, heart rate, and strokevolume were similar at all levels of oxygenuptake. Again this suggests a major role ofskin temperature in determining the hemody-namic responses to combined exercise andthermal stresses. However, responses fromfour subjects who responded similarly (Fig.4) reveal several features distinguishing theirresponses from those at lower oxygen uptakes.

With heating (period 2), the temperature




ownloaded from



of right atrial blood and central blood volumeshowed a small transient dip and rosethereafter—central blood volume only slight-ly. This suggests that there was no sudden,major redistribution of blood flow or volumeto the skin. Judging from the magnitude ofdrift in aortic mean pressure and strokevolume during the control period, the majoradjustments in the cutaneous and possiblyother peripheral vascular beds may havealready occurred during that period. Thestriking stability in temperature of right atrialblood after 15 minutes supports this reasoning.

Although 30 minutes of heating appeared tohave altered flow through the skin, much thesame as at the lower workloads (assuming theincrement in cardiac output goes to the skin),this may have been accomplished withoutfurther significant displacement of bloodvolume peripherally. At least, behavior ofaortic mean pressure and central bloodvolume during that interval suggested this.Unfortunately, reliable measurements of rightatrial mean pressure could not be made athigher workloads because of large artifactsinduced by walking up a grade. The fourfoldincrease in the difference between tempera-ture of right atrial blood and skin tempera-ture, despite an only 1.5- to 2-fold increase inheat production, suggests skin blood flow mayhave been even less at higher than at lowerworkloads. A thermally induced increase inmuscle blood flow cannot be ruled out.

The dramatic drop in aortic mean pressureat high levels of work (and in subject G.Z., ata lower level) on cooling raises some perplex-ing questions. Where has vasodilatation oc-curred, and why have baroreceptor reflexesnot buffered the changes more effectively—particularly in subject P.K.?

Possible explanations of the reversal ofaortic mean pressure responses at the highoxygen uptake, that is, rise with heating andinitial fall with cooling, are as follows: (1)The increase in cardiac output was unaccom-panied by any further net vasodilatation sothat total peripheral resistance remainedconstant from 45 to 60 minutes as aortic meanpressure and flow rose proportionally. This


indicates that either there was not furthervasodilatation of skin once it was heated orany which did occur was compensated for byvasoconstriction elsewhere (7, 16). (2) Onlowering skin temperature, the drop in aorticmean pressure indicates a sudden reversal ofcompensator)' vasoconstriction. The fall inaortic mean pressure is clearly not attributablein any subject to a major fall in cardiac outputduring a time when skin was still vasodilated(rapidly falling temperature of right atrialblood indicated skin vasodilatation at thattime). Cardiac output was still close to peakvalues at that time.

A major feature of the cardiovascularresponse to competitive thermal and metabolicstress is the maintenance of aortic meanpressure during severe thermal stress whenmetabolic demands are very high in contrastto the fall seen at lower metabolic rates. Ourdata imply that when demands for oxygentransport are very high they will be main-tained at the expense of thermoregulation.

AcknowledgmentsWe are grateful for the assistance of Mrs. Fusako

Kusunii and Mrs. Evelyn C. Steen, R. N., and to Mr.P. E. Purser and Mr. A. G. Buck of NASA MannedSpace Craft Center, Houston, Texas, who madepossible the loan of the water-perfused suits.

References1. {CRANING, K. K. II, BELDINC, H. A., AND HERTIC,

B. A.: Use of sweating rate to predict otherphysiological responses to heat. J. Appl.Physiol. 21: 111, 1966.

2. GREENFIELD, A. D. M.: The circulation throughthe skin. In Handbook of Physiology, sec. 2,vol. 2, Circulation, edited by W. F. Hamiltonand P. Dow. Washington, D. C , AmericanPhysiological Society, 1964, p. 2615.

3. BEVECARD, B. S., AND SHEPHERD, J. T.: Changesin tone of limb veins during supine exercise. J.Appl. Physiol. 20: 1, 1965.

4. BEVECARD, B. S., AND SHEPHERD, J. T.: Reactionin man of resistance and capacity vessels inforearm and hand to leg exercise. J. Appl.Physiol. 21: 123, 1966.

5. TAYLOR, H. L., BUSKIRK, E., AND HENSCHEL, A.:Maximal oxygen uptake as an objectivemeasure of cardio-respiratory performance. J.Appl. Physiol. 8: 73, 1955.

6. DOAN, A. E., PETERSON, D. R., BLACKMON, J. R.,AND BRUCE, R. A.: Myocardial ischemia after



ownloaded from



maximal exercise in healthy man. Am. Heart J.69: 11, 1965.

7. ROWELL, L. B., BLACKMON, J. R., MARTIN, R. H.,MAZZARELLA, J. A., ANT) BRUCE, R. A.: Hepaticclearance of indocyanine green in man underthermal and exercise stresses. J. Appl. Physiol.20: 384, 1963.

8. ROWELL, L. B., MARX, H. J., BRUCE, R. A., CONN,R. D., KUSUMI, F.: Reductions in cardiacoutput, central blood volume, and strokevolume with thermal stress in normal menduring exercise. J. Clin. Invest. 45: 1801,1966.

9. ROWELL, L. B., BRENCELMANN, C. L., BLACKMONJ. R., BRUCE, R. A., AND MURRAY, J. A.:Disparities between aortic and peripheral pulsepressures induced by upright exercise andvasomotor changes in man. Circulation 37:954, 1968.

10. RODDIE, I. C , SHEPHERD, J. T., AND WHELAN, R.F.: Evidence from venous oxygen saturationmeasurements that the increase in forearmblood flow during body heating is confined tothe skin. J. Physiol. (London) 134: 444,1956.

11. STROM, G.: Influence of anoxia on lactateutilization in man after prolonged muscularwork. Acta Physiol. Scand. 17: 440, 1949.

12. HOLMGREN, A.: Circulatory changes duringmuscular work in man: With special referenceto arterial and central venous pressures in thesystemic circulation. Scand J. Clin. Lab. Invest.8 (suppl. 24): 48, 1956.

13. KOROXENIDIS, G. T., SHEPHERD, J. T., ANDMARSHALL, R. J.: Cardiovascular response to

acute heat stress. J. Appl. Physiol. 16: 869,1961.

14. ROWELL, L. B., BRENCELMANN, G. L., BLACKMON,J. R., Twiss, R. D., AND KUSUMI, F.:Splanchnic blood flow and metabolism in heat-stressed man. J. Appl. Physiol. 24: 475,1968.

15. WILLIAMS, C. G., BREDELL, G. A. G., WYND-HAM, C. H., STRYDOM, N. B., MORRISON, J. F.,PETER, J., FLEMING, P. W., AND WARD, J. S.:Circulatory and metabolic reactions to work inheat. J. Appl. Physiol. 17: 625, 1962.

16. RADIGAN, L. R., AND ROBINSON, S.: Effects ofenvironmental heat stress and exercise on renalblood flow and filtration rate. J. Appl. Physiol.2: 185, 1949.

17. COOPER, K. E., AND KERSLAKE, D. MCK.:Changes in heart rate during exposure of theskin to radiant heat. Clin. Sci. 14: 125,1955.

18. WEBB-PEPLOE, M. M., AND SHEPHERD, J. T.:Responses of superficial limb veins of the dogto changes in temperature. Circulation Res.22: 737, 1968.

19. EKELUND, L.-G., AND HOLMGREN, H.: Circulatoryand respiratory adaptation during long-term,non-steady state exercise, in sitting position.Acta Physiol. Scand. 62: 240, 1964.

20. WOOD, J. E., AND ECKSTEIN, J. W.: A tandemforearm plethysmograph for study of acuteresponses of the peripheral veins of man: Theeffect of environmental and local temperaturechange and the effect of pooling blood in theextremities. J. Clin. Invest. 37: 41, 1958.




ownloaded from


K. KRANING IILORING B. ROWELL, JOHN A. MURRAY, GEORGE L. BRENGELMANN and KENNETH

ExerciseHuman Cardiovascular Adjustments to Rapid Changes in Skin Temperature during

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1969 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.24.5.7111969;24:711-724Circ Res.

http://circres.ahajournals.org/content/24/5/711World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:



ownloaded from

http://circres.ahajournals.org/content/24/5/711

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


human cardiovascular adjustments to rapid change in...

Documents