control of hear t rat bye th autonomie c nervous...

Control of Heart Rate by the AutonomicNervous System

STUDIES IN MAN ON THE INTERRELATION BETWEEN

BARORECEPTOR MECHANISMS AND EXERCISE

By Brian F. Robinson, Stephen E. Epstein, G. David Beiser, and

Eugene Braunwald

ABSTRACTThe control of heart rate by the autonomic nervous system was investigated

in conscious human subjects by observing the effects of /3-adrenergic blockadewith propranolol, of parasympathetic blockade with atropine, and of com-bined sympathetic and parasympathetic blockade. The increase in heart ratewith mild exercise in supine men was mediated predominantly by a decreasein parasympathetic activity; at higher levels of work, however, sympatheticstimulation also contributed to cardiac acceleration. When the response to 80°head-up tilt was compared with the response to exercise in the same subjectsupine, it appeared that the attainment of an equivalent heart rate was asso-ciated with a significantly greater degree of sympathetic activity during tiltingthan during exercise. Although heart rate was always higher at any givenpressure during exercise than it had been at rest, the changes in heart rate thatfollowed alterations in arterial pressure were found to be of similar magni-tudes at rest and during exercise; it was therefore concluded that the sensi-tivity of the baroreceptor system was not altered during exercise. Investiga-tion of the efferent pathways concerned in mediating the baroreceptor-inducedchanges in heart rate suggested that the relative roles of the sympathetic andparasympathetic systems were nearly equal in the resting state. During exer-cise, on the other hand, changes in sympathetic activity appeared to be thepredominant mechanism by which speeding and slowing of the heart wasachieved. It thus appears that baroreceptor-induced alterations in heart ratemay be mediated by increased or decreased activity of either efferent system;the ultimate balance, however, is critically dependent on the preexisting levelof background autonomic activity.

ADDITIONAL KEY WORDS propranololsympathetic nervous system tiltingheart rate control

atropinebeta-adrenergic receptors

baroreceptor reflex

• The central nervous system controls heartrate by varying the impulse traffic in sympa-thetic and parasympathetic nerve fibers ter-minating in the sinoatrial node. Although therehas been considerable interest in the mech-anisms by which the heart rate is altered inresponse to exercise, postural changes, andstimulation of the baroreceptors, the relativeroles of the two divisions of the autonomic

From the Cardiology Branch, National Heart In-stitute, Bethesda, Maryland.

Dr. Robinson is a Nuffield Medical Fellow.Accepted for publication April 4, 1966.

nervous system in mediating these changes inrate in conscious man have not been clarified.Furthermore, little information is availableconcerning the manner in which the heart rateresponse to baroreceptor stimulation is modi-fied during exercise.

In the present study, the relative roles ofthe efferent pathways which mediate the heartrate response to exercise in supine man wereinvestigated by observing the separate andcombined effects of /J-adrenergic blockadeand parasympathetic blockade on cardiacacceleration. The cardiac acceleration result-ing from head-up tilting was studied in a

400 CircmUtion Rutnb, Vol. XIX, A»t»st 1966

by guest on May 8, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

CONTROL OF HEART RATE IN MAN 401

similar way, thus allowing a comparison ofthe mechanisms of speeding under conditionsof exercise and changes of posture. The sensi-tivity of the control of heart rate by thebaroreceptor system was then compared atrest and during exercise, and the efferentpathways utilized were studied by observingthe effects of the autonomic blocking agents.

MethodsFive normal volunteers and one patient (AS)

who had undergone a successful mitral commis-surotomy five months earlier were studied; theirages ranged from 19 to 28 years. All of the sub-jects were males and in good physical condition,but none was a trained athlete. All studies werecarried out in the postabsorptive state.

1. EXERCISE IN SUPINE POSITION AND CHANGE OFPOSTURE

Before any measurements were made, each sub-ject was thoroughly familiarized with the use ofthe bicycle ergometer in the supine position andwith the tilt table. In the definitive investigation,the subject was first placed on the tilt table inthe supine position and his heart rate was de-termined from the ECG for 30 sec. He was thentilted, first to 45° and then to 80° from thehorizontal, and the heart rate was recorded ineach position when the cardiotachometer andECG tracings showed that the rate was stable.Heart rate was then recorded with the subjectat rest on the bicycle ergometer in the supineposition and during the fourth minute of exerciseat each of a series of work levels which werecarried out consecutively without interveningrest periods. Oxygen uptake (Vo2) was recordedthroughout the period of exercise using a con-tinuous flow system.1 The protocol outlined abovewas repeated on three subsequent days; the car-diac sympathetic nerves were blocked on one,the parasympathetic nerves on another, and bothon the third. Stimulation of the heart throughthe sympathetic nerves and by circulatory cate-cholamines secreted by the adrenals was inhibitedby the /3-adrenergic blocking agent propranolol*given in a dose of 0.25 mg/kg iv; it has previous-ly been shown that 0.15 mg/kg reduces die effec-tiveness of infused isoproterenol at least 90%.2

Throughout this paper this is referred to as sym-pathetic blockade, although only the /3-receptorswere blocked. No untoward effects of propranololwere observed in any of the subjects at this higherdosage. Parasympathetic blockade was induced

* l-Isopropylamino-3- (1-naphthyloxy )-2-propanolhydrochloride.CircnUtion Rtsemrcb, Vol. XIX, Atgujl 1966

with 2 mg of atropine given intravenously. Com-bined sympathetic and parasympathetic blockade(double blockade) was produced by the simul-taneous administration of both drugs.

2. THE BARORECEPTOR SYSTEM

The studies at rest were carried out with thesubject seated in a chair and those during exer-cise while he walked on a motor-driven treadmill.The subjects were completely familiarized withwalking on the treadmill before the study com-menced, and a speed and grade were found whichconsistently resulted in a heart rate between 90and 125 beats/min. The oxygen requirements atthis level of exercise ranged from 960 to 1,950ml/min. Arterial pressure was recorded througha short teflon catheter that had been introducedpercutaneously into the brachial artery and con-nected to a Statham P23AA pressure transducer;heart rate was determined for 30-sec periodsfrom the simultaneously recorded ECG. Increasesof arterial pressure were produced by the intra-venous infusion of phenylephrine, a pressor agentthat in the doses used is without measurabledirect cardiac effect''; the rate of administrationwas controlled by a constant speed pump.

In each experiment, the subject _was_ firststudied at rest. When mean arterial pressure andheart rate had stabilized, the arterial pressure wasraised in steps by infusing phenylephrine indoses varying from 0.05 to 0.40 mg/min, andheart rate was recorded when the pressure hadstabilized at the new level for at least 1 min. Afterdiscontinuing the phenylephrine, the subjectwalked on the treadmill at the predeterminedspeed and grade; arterial pressure and heart ratewere recorded after 5 min and again after 7 minto ensure that a steady state had been reached.The effects on heart rate of elevating arterialpressure with graded doses of phenylephrinewere then determined during exercise as at rest,but larger doses of the drug, up to 0.80 mg/min,were required to produce elevations of arterialpressure during exercise comparable to thoseachieved at rest. After discontinuing the phenyl-ephrine and allowing its effects to wear off, nitro-glycerin (0.4 mg or 0.8 mg) was given sublingual-ly while the subject was still exercising, andarterial pressure and heart rate were recorded atthe peak of the drug effect. The effects of reduc-ing arterial pressure were studied again after theheart rate had stabilized at rest. The entire studywas repeated on three subsequent days in thesame way as in the posture studies. The action ofcardiac sympathetic nerves was inhibited on oneoccasion, the parasympathetics on another, andboth on the third, utilizing the doses of blockingagents described above. The experiments wereusually completed within 45 min of administra-



ownloaded from


402 ROBINSON, EPSTEIN, BEISER, BRAUNWALD

tion of the blocking drug; when the study con-tinued longer, a supplemental dose (0.1 mg/kg ofpropranolol; 1 mg atropine) was given. The sub-jects were in sinus rhythm under all of the ex-perimental conditions studied.

Results and Interpretations1. RELATIVE ROLES OF PARASYMPATHETIC ANDSYMPATHETIC EFFERENTS IN THE HEART RATEA. Response to Exercise in Supine Position

Four normal subjects were investigated andall showed qualitatively similar responses(table 1); the results in one are illustratedin figure 1, left. In the control studies, carriedout with both the sympathetic and parasympa-thetic efferents intact, the heart rate at restaveraged 52/min in the four subjects and rosein an approximately linear manner with in-creasing V02. Following blockade of both sym-pathetic and parasympathetic efferents (dou-ble blockade), the resting heart rate roseabove the control level in all subjects (averageincrease of 42/min), but there was little orno further increase with mild exercise (Vo2 upto 500 ml/min); small increases in rate, aver-

aging 10/min, occurred at higher levels ofwork (Vo2 900 to 1300 ml/min). When thecardiac sympathetic system alone was blocked,the parasympathetic remaining intact, the rest-ing heart rate fell by an average of 4/min.The rate increased normally with mild exercisein every subject, but there was less speedingat the higher levels of work in three of thefour (GM, RR, and RJ: table 1, fig. 1).When the parasympathetic system was in-hibited, leaving the sympathetic intact, rest-ing heart rate increased by an average of46/min above control; cardiac accelerationin response to mild exertion was reduced andin two subjects was abolished, but the raterose normally at the higher levels of exercise.

It thus appears that in the supine restingstate, parasympathetic restraint is the domi-nant influence on heart rate (fig. 1, Pr), andthe accelerating effects of sympathetic stimu-lation (fig. 1, Sr) are minor. The speedingof the heart in response to mild exerciseappears to result largely from withdrawal of

TABLE 1

Effects of Autonomic Blockade on Heart Rate Response to Supine Exercise

GM

RR

PY

RJ

Subject

RestExercise

RestExercise

RestExercise

RestExercise

VO,ml/mln

233455700913

271385499770

1033

259418701919

1069

234436677755

1289

ControlHR/min

597798

108

4958658191

4959727992

49657782

115

Sympatheticblockade

VOi HRml/min /min

242317509651994

299405510774

1038

343510686898

1038

229418627940

1358

5970788196

4359657583

4760708084

4454667990

Paritympath eticblockade

Vo» HRml/min /min

275355514691922

284408479745

1011

309507705877980

225425656940

1248

93101107113134

949696

110122

115112107111127

848894

110122

Doubleblockade

VOi HRml/min /min

330369515661953

292389486797

1030

290457651845

1012

244410634878

1255

85868994

104

8686868996

113108108111113

90878893

100

Control = autonomic nervous system functioning normally; double blockade = combined sympathetic and para-sympathetic blockade; Vo2 = oxygen consumption; HR = heart rate.

CirttUliom Research, Vol. XIX, Amgxit 1966



ownloaded from



parasympathetic inhibition, since cardiac ac-celeration is essentially unimpaired by sympa-thetic blockade but tends to be inhibited byparasympathetic blockade and double block-ade. At higher levels of exercise, however,

400 600 800 1000

V02 HI I MINUTE

FIGURE 1

SUPIHE 45" JO-

POSITION

A comparison of the effects of various forms of auto-nomic blockade on the response of heart rate insubject RR to exercise when supine and to head-uptilt. The level of exercise in the control (no blockade)exercise study which would have produced a heartrate equal to that observed during the control 80°tilt is determined by interpolation (lines X and Y, seetext). Effects of parasympathetic blockade on heartrate are indicated by Pr at rest, Po during exercise,and Pt during tilting. Effects of sympathetic blockadeare indicated by Sr, Se and St.

cardiac acceleration must result in part fromsympathetic stimulation, since sympatheticblockade reduces the augmentation of heartrate in comparison to the control study. Thefinding that the increments in rate duringheavy exercise are smaller after doubleblockade than after parasympathetic blockadealone further identifies the contribution of thesympathetic system. These results were ob-tained in young healthy male adults, and itis possible that a different balance of auto-nomic effect might obtain in a different groupof subjects.

B. Response to Tilting

The heart rate rose by an average of 29/minduring the 80° head-up tilt in the controlstudy with the subjects in the unblockedstate (table 2, fig. 1). Following doubleblockade, the increase in rate induced bytilting was almost abolished and averagedonly 5/min. During parasympathetic blockade,however, heart rate increased by an averageof 36/min and during sympathetic blockadeby 17/min. Thus, in contrast to light exercisein supine position, substantial speeding couldstill occur during tilting when the parasympa-thetic system was blocked.

The results were then analyzed to charac-terize in greater detail the differences be-tween the mechanisms of cardiac speeding

TABLE 2

Effects of Autonomic Blockade on Heart Rate Response to Tilting

Subject

TUtangle

degreesControlHR/min

SympatheticblockadeHR/min

ParasympatheticblockadeHR/min

DoubleblockadeHR/min

GM

RR

PY

RJ

Mean

0458004580045800458004580

617288486179506879536680536782

576875445662465767444857485765

961141309711913411613815196113134101121137

888692899293108118114848690929697

Abbreviations as in table 1.

OrcmUtwn Rtsmrcb, Vol. XIX, Amguii 1966



ownloaded from



TABLE 3

Comparison of Effects of Sympathetic and Parasympathetic Blockade on Heart Rate Responseto Supine Exercise and Tilting

Subject

GM

RR

PY

RJ

Avg

Exercise80° tiltExercise80° tiltExercise80° tiltExercise80° tiltExercise80° tiltP

ControlHR/mln

8888797979797878

Sympatheticblockade

HR/mln

8075746278676956

AHR

- 8-13- 5-17- 1-12- 9-22- 6-16

<.O5

ParaBympatheticblockade

HR/mln

110131110134111152103132

AHR

+ 22+ 43+ 31+ 55+ 32+ 73+ 25+ 54+ 27+ 56

<.O2

AHR = change in heart rate compared to control; Avg = average change in heart rate com-pared to control. Other abbreviations as in table 1.

during exercise and tilting. The level of Vo2

during exercise which would have producedthe same degree of cardiac acceleration asthat actually observed during the 80° tilt wasdetermined by interpolation (fig. 1, lines Xand Y). The effect of sympathetic and para-sympathetic blockade on the heart rate atthis level of Voo was then estimated bydetermining the intercepts of line Y with theappropriate heart rate-Vc>2 relationship. Theeffects of autonomic blockade could then becompared under conditions of exercise andtilting which had identical effects upon theheart rate when the autonomic nervous systemwas functioning normally. If the mechanismof speeding had been identical during exerciseand tilting, then the effects of autonomicblockade should also have been similar. Infact, the responses to sympathetic and toparasympathetic blockade were consistentlydifferent during the two interventions (fig. 1,table 3). During exercise, parasympatheticblockade increased heart rate by an averageof 27/min (fig. 1, Pe), but the increase duringtilting (Pt) was more than twice as greatand averaged 56/min (P<.02). Similarly,during exercise, sympathetic blockade de-creased the heart rate by an average of only6/min (Se), whereas the decrease duringtilting (St) was significantly greater (P < .05),averaging I6/min. It is thus apparent thatachievement of the same heart rate involved

a different balance of autonomic activity,depending on whether the stimulus was pro-vided by exercise in supine position or bytilting. Exercise resulted in a greater degreeof parasympathetic withdrawal than did tiltingas is shown by the lesser degree of speedinginduced by atropine during exercise. Converse-ly, exercise produced a smaller degree ofsympathetic stimulation than did tilting, asis shown by the lesser degree of slowinginduced by /3- adrenergic blockade. Thus, thecardiac acceleration in response to mild supineexercise in these subjects appeared to dependpredominantly on parasympathetic withdraw-al, whereas that produced by tilting involveda relatively greater degree of sympatheticstimulation.

2. EFFECTS OF EXERCISE ON CONTROL OF HEART RATEBY THE BARORECEPTOR MECHANISM

In these experiments, changes in heart ratewere related to the changes in mean arterialpressure with which they were associated.It is appreciated that the use of mean arterialpressure as an index of baroreceptor stimula-tion ignores the passible influence of alterationsin pulse pressure; however, since mean pres-sure and pulse pressure always changed in thesame direction, the basic pattern of the resultsis unaffected by the fact that mean arterialpressure alone was considered. Although therelation between heart rate and mean arterialpressure was not a linear one, it was never-

CircuUtwa RutarcB, Vol. XIX, August 1966



ownloaded from



theless found useful to express the results interms of average values when making broadcomparisons between different conditions.

At rest, graded elevations of mean arterialpressure in the six subjects studied causedprogressive decreases in heart rate, while re-ductions of pressure uniformly resulted inincreases of heart rate (table 4, fig. 2). Onthe average, a change of 40 mm Hg in meanarterial pressure was associated with an in-verse change in rate of 50/min. During exer-cise, the heart rate was always higher atany given arterial pressure than it was atrest, but the absolute magnitude of changesin rate that followed alterations in pressurewere similar to those observed at rest (fig.2);on the average, a change of 37 mm Hg inmean arterial pressure was associated with achange in rate of 42/min. One subject (GM)was studied at rest in both the sitting andstanding positions, as well as at two levelsof exercise; the relation between heart rateand mean arterial pressure could be expressedby four similar curves, each increase in stressleading to a further upward displacement ofthe entire curve (fig. 3).

60 70 13080 90 100 110

UEAN ARTERIAL PRESSURE

FIGURE 2

Relation between heart rate and mean arterial pressurein subject RR sitting at rest and during exercise inupright position. Arrows identify the baseline observa-tions made before phenylephrine or nitroglycerinwas given to modify arterial pressure. Slopes of thelines relating heart rate to mean arterial pressure aresimilar at rest and during exercise, but the heart rateis always greater at any given arterial pressure duringexercise than it is at rest.

140

120

100

40

"1 1 T 1 I 1 1 1—

o Enron Wj -1140 ml/Hit

• titn'ot fCj • 9S0 ml /Hit

A SHtiif Rill

A Slllitf Rill

50 60 70 60 90 100 120 M0 160 ISO

UEAN ARTERIAL PRESSURE » K |

FIGURE 3

Relation between heart rate and mean arterial pres-sure in subject GM studied at rest, in the sitting andstanding positions and at two levels of exercise inupright position. Arrows identify the baseline observa-tions before arterial pressure was altered. Relationsbetween heart rate and mean arterial pressure aresimilar under all four conditions, but the curve isdisplaced upward with the change from sitting tostanding, and there is further upward displacementat each of the two levels of exercise.

180

160 -

140 -

120 -

100 -

60 -

60 -

40

1REST

- \

_

A

A

1

1

\ !

1 1 1 1

0 CcnlnlA Pamjmpalhtlic Bhettdi ~• S/mfttiitic BlKlliiA DttUt Blutidi -

>—

40 60 80 100 120 140

MEAN ARTERIAL PRESSURE a n

160 180

FIGURE 4

Effects of varying types of autonomic blockade onthe relation between heart rate and mean arterialpressure in subject RR sitting at rest. Arrows denotethe baseline observations before arterial pressurewas modified under each circumstance. Heart ratecontinues to show some inverse variation in responseto changes in mean arterial pressure when eitherdivision of the autonomic was blocked. When bothdivisions were blocked, changes in rate were almostcompletely prevented.

Circulation Rtiircb, Vol. XIX, August 1966



ownloaded from



TABLE 4

Effects of Autonomic Blockade on Heart Rate Response to Changes in Arterial Pressure

GM

RR

PY

RJ

JW

Subject

Rest

ExerciseV o 2 = 960

Rest

ExerciseVo2 = 1430

Rest

ExerciseVo2 = 1950

Rest

ExerciseVo2 = 1406

Rest

BP

N

BP

BNBP

N

BP

BN

BP

N

BP

BNBP

N

BP

BN

BP

Control

MAPmmHg HR/mln

8395

11248

858896

104

65879498

11378

86879195

110

727583909598

1037671818590

10478718391

10511875709398

110120977575808898

715850

104

11811610694

1467062554892

1101081019286

13814454474038366980

12311610910614615655514443769897949290

108130136757265

Sympathetic blockade

MAPmmHg

7585

10068668490

110121

6296

104

8175848792

103

797573839195

72

8487

10010990708592

105

76758495

1051158784

HR/min

676349677396868378

965646

627088868380

869495494338

63

95948885

100102464439

5164847978787888

Paruympatheticblockade

MAP mm Hi HR/mln

951321387565829997

127

599097

127

6660738388

100

68

7587

112

60

84100128

65618192

126

48

6297

1121225250

1109496

108132131118116108

139118106103

126137141134120119

178

140120116

141

165142136

163181110

94101

114

120104105114128142

Double blockade

HAP mm Hi HR/min

8298

14260

6383

112

5291

113

67657797

126

7670628083

105118

57

697589

107675558

103122

58527892

1051187774

(Table 4, continued on

87848884

10210098

1048988

8392

1009790

103108118999999

101

97

120120115109121126849391

8791

101989696

104110

next page.)

CirctUiio* Risurcb, Vol. XIX, August 1966



ownloaded from



3. RELATIVE ROLES OF PARASYMPATHETIC ANDSYMPATHETIC EFFERENT* IN MEDIATING BARORECEPTOR-INDUCED ALTERATIONS IN RATE

A. Rest

In the 4 subjects studied, a change of 41mm Hg in arterial pressure in the unblockedstate was associated on average with changein rate of 49/min (table 4, fig. 4). Followingdouble blockade, the heart rate at rest re-mained essentially constant when arterial pres-sure was altered (fig. 4), a change of 65 mmHg in mean arterial pressure being associatedwith an inverse change in heart rate of only7/min; autonomic stimuli resulting fromchanges in pressure would therefore appearto have been effectively inhibited by thedrugs used. When the parasympathetic systemalone was blocked, substantial changes inheart rate occurred in response to changesin arterial pressure, but tended to be lessthan during the control period; on the aver-age, a change of 63 mm Hg in mean arterialpressure was associated with a change in heartrate of 29/min. During parasympathetic block-ade, when rate was controlled primarily bythe sympathetic system, the heart not onlyspeeded when arterial pressure was reducedbelow control, but also slowed as the pressurewas elevated; this indicates that in a resting,

sitting subject the heart rate can be changedreflexly by either increase or decrease in theactivity of the sympathetic nervous system.

When the sympathetic system alone wasblocked, the heart rate continued to vary in-versely with arterial pressure, a change of 29mm Hg being associated with an averagechange in heart rate of 25/min. During sympa-thetic blockade, when rate was controlledprimarily by the parasympathetic system, ratecould not only be slowed by an intensificationof vagal restraint as pressure was elevated,but could also be speeded by a decrease invagal activity when pressure was decreasedbelow control.

B. Ex«rcb«

In the four subjects in whom the effectsof autonomic blockade on the response tobaroreceptor stimulation were studied duringexercise in upright position, an average changein arterial pressure of 40 mm Hg was associatedwith a change in heart rate of 52/min duringthe control studies (table 4, fig. 5). In thesame subjects, parasympathetic blockadecaused only a slight impairment of the rateresponse, an average change in arterial pres-sure of 60 mm Hg being associated with a

Subject

N

BP

BNBP

N

BP

BN

MAP mm

11680759196

120807566738067607475857065

TABLE 4 continuedControl Sympathetic blockade

H i HR/min MAP mm Hg HR/mli

568896

10510394

10010681695098

11394856895

103

P a raiy mpa th et i cblockade Double blockade

AS

Exercise

Rest

Exercise

MAP = mean arterial pressure; B = baseline; P = phenylephrine given in increasing doses; N = nitroglycerine.Other abbreviations as in table 1.

Circulation Riunb, Vol. XIX, August 1966



ownloaded from



200

180 -

160

5 140 -

120 -

100 -

60

1EXERCISE

s1

1

A

5 A

, \ A

V1

1 1 1 1

O CMIIII

A Parts/apsfielic Bleckodt ~

• Sjmptlktlic Bhcksdi

A Dot bit Blictnit —

^A^__

1 1 1

-

40 60 80 100 120 140

MEAN ARTERIAL PRESSURE mre

160 180

FIGURE 5

Effects of varying types of autonomic blockade onthe relation between heart rate and mean arterialpressure during exercise in subject RR. Arrows denotethe baseline observations for each. Response of heartrate to alterations of arterial pressure is only slightlyreduced after parasympathetic blockade, but it ismarkedly impaired following sympathetic blockadeand double blockade.

change in rate of 43/min. This finding suggeststhat during exercise the alterations in rate asa consequence of changing arterial pressuretook place largely as a result of variationsin sympathetic activity. This conclusion issupported by the finding that when the sympa-thetic system alone was blocked, the heartrate response to changes in arterial pressurewas much impaired, a change of 40 mm Hgbeing associated with an average change inrate of only 15/min. Moreover, when sympa-thetic blockade was added to parasympatheticblockade, the rate response that had beenobserved during parasympathetic blockadealone was markedly attenuated; during doubleblockade a change in pressure of 55 mm Hgwas associated with an average change in rateof 15/min.

At the levels of exercise employed, changesin the degree of sympathetic stimulation ofthe heart thus appeared to play a moreimportant role in mediating baroreceptor in-duced increases and decreases in heart ratethan did changes in parasympathetic activity.At rest, on the other hand, the roles of the

sympathetic and parasympathetic systems inmediating baroreceptor reflexes appeared tobe more nearly equal.

Discussion

Interest in the mechanism of exercise-in-duced tachycardia dates to 1914 when Gasserand Meek observed in the dog that aftervagotomy little augmentation of heart rateoccurred during exercise, although stellateganglionectomy did not greatly impair therate response.4 Bruce et al., on the other hand,studying anesthetized dogs, observed that thecardioacceleration produced by electrically in-duced muscular contraction was abolished notonly by vagotomy, but also by excision ofthe upper thoracic sympathetic ganglia.6

Donald and Shepherd, however, reported thateven after total cardiac denervation, animalsexhibited a large, though subnormal increasein heart rate during running.8 Relatively littleinformation is available concerning the rela-tive roles of the two components of the auto-nomic nervous system in effecting the tachy-cardia of exercise in man. S. Robinson et al.7

observed that at maximal levels of exerciseatropine did not elevate heart rate, suggestingthat vagal restraint is normally totally with-drawn at this level of activity. It is generallyagreed that sympathetic blockade lowers theheart rate during exercise, a finding whichsuggests that sympathetic stimulation contrib-utes to cardiac acceleration.2' 8"10 No informa-tion is available, however, on the interplaybetween the two divisions of the autonomicnervous system during exercise, and the rela-tive importance of each at varying levels ofexercise has not previously been defined.

In the present investigation we examinedthe response to exercise in the supine position,since only in this way could the effects ofexercise be separated from the effects of theupright position with its associated alterationsin the background of autonomic activity. Theresults, presented schematically in figure 6,indicate that the speeding of heart rate atrelatively mild exercise (Vo2 up to 500 ml/min) is effected primarily by a withdrawalof parasympathetic restraint on the sinoatrial

CircmUlion Resttnb, Vol. XIX, Anguii 1966



ownloaded from



Para s/epctki lie Bfecktde

\

\ -

\ „COM! rol

Ti

Sympathetic Blocked*

REST MAX ¥0

FIGURE 6

Schematic diagram showing the relative contributionsof the sympathetic and parasympathetic systems tocardioacceleration at various levels of exercise. Brokenlines are extrapolations based on published'- ? andunpublished data. Comparisons are between control(no blockade) state and parasympathetic or sympa-thetic blockade.

node. At higher levels of work, further with-drawal of parasympathetic restraint occurs, butincreases in sympathetic activity becomeprogressively more important in acceleratingthe cardiac rate. This interpretation is de-rived from a comparison of the rate responsesat varying levels of exercise in the control studywith those observed after either sympatheticor parasympathetic blockade; comparison ofthe rate response after blocking one divisionof the autonomic with the response afterdouble blockade leads to a similar conclusion(fig-7).

The possibility cannot be excluded thatinterference with the activity of one divisionof the autonomic nervous system might leadto compensatory changes in the other thatwould obscure to some extent the relativecontribution of each of the two components.Interpretation of the results could also becomplicated if the degree of blockade ofeither system was incomplete. This did notappear to be the case in the present experi-ments, however, since heart rate was con-stant, or increased only slightly at the higherlevels of exercise during double blockade. Itmust be borne in mind that the autonomicnervous system is probably not the only mech-

CircuUium Rti—rcb, Vol. XIX, August 1966

anism concerned in speeding the heart dur-ing severe exertion. As already mentioned, theheart rate of dogs continues to increase sub-stantially during heavy exercise even aftercomplete cardiac denervatdon,6 and the in-creases in rate seen in man during maximumexertion after double blockade (our labora-tory, unpublished observations) may be dueat least partly to nonautonomic influences.

In contrast to the mechanism of speedingduring mild exercise, the increase in heartrate induced by upright tilting involved asignificantly greater degree of sympatheticstimulation. This finding is relevant to theunderlying mechanism responsible for thetachycardia of exercise. It has been suggestedthat this tachycardia is mediated primarilythrough the baroreceptor system.11'12 Sincearterial pressure does not normally fall in manduring exercise, this hypothesis in its simp-lest form appears to be untenable. It could beargued, however, that the baroreceptor sys-tem is 'reset' during exercise and-responds asif the blood pressure were reduced. If thiswere the case, it might be expected that thetachycardia of exercise would be mediatedby a balance of autonomic activity similar tothat observed during tilting. The observation

ic Bltcktdt

REST MAX V 0 ,

FIGURE 7

Schematic diagram showing the relative contributionsof the sympathetic and parasympathetic systems tocardioacceleration at various levels of exercise. Brokenlines are extrapolations based on published'-7 andunpublished data. Comparisons are between doubleblocked state and parasympathetic or sympatheticblockade alone.



ownloaded from



that the mechanisms of speeding are in factquite different suggests that the speeding ofthe heart during exercise is not mediatedprimarily through the baroreceptor system,but depends, at least partly, on some othermechanism.

Interest in the control of heart rate by thebaroreceptor system dates back to the workof Marey, who in 1859 demonstrated the in-verse relation between arterial pressure andheart rate.18 Subsequent attempts to definethe mechanisms involved in the efferent limbof this reflex more precisely have yielded con-flicting results. Several investigations carriedout in experimental animals demonstrated areciprocal relation between the sympatheticand parasympathetic nervous systems in re-sponse to changes in arterial pressure.14"10

However, on the basis of studies carried outin this laboratory, primarily on anesthetizeddogs, it was concluded that when arterialpressure rises above control, the slowing ofheart rate results from an increase in para-sympathetic restraint, while when pressurefalls below control levels, the resultant- in-crease in heart rate is mediated primarily bythe sympathetic nervous system.3 The resultsof the present investigation seem to resolvethis apparent discrepancy. One of the majordifferences in the experimental conditions be-tween the investigations in which a reciprocalrelation was demonstrated and the study inwhich it was not was the level of backgroundsympathetic tone present. It has been demon-strated in this study that the backgroundautonomic activity existing prior to the in-duced change in arterial pressure profoundlyinfluences the results. In the sitting position,sympathetic activity is apparently of sufficientmagnitude to permit slight cardiac slowingby withdrawal of sympathetic stimulation aspressure is raised. In contrast, in the supineposition sympathetic activity appears to beminimal, since slowing does not occur con-sistently when arterial pressure is elevatedafter parasympathetic blockade.8 On the otherhand, during exercise in the upright position,when sympathetic activity is at a much higherlevel and parasympathetic activity is lessened,

cardiac slowing consequent upon increases inarterial pressure results predominantly fromreduction in sympathetic tone; increase inparasympathetic activity plays a relativelyminor role.

The observations on the effects of exerciseon the relation between arterial pressure andheart rate (figs. 2 and 3) are relevant to thequestion of the sensitivity of the baroreceptormechanism under differing conditions. It hasbeen suggested that the baroreceptor reflexesare suppressed during static muscular exer-cise.17 On the other hand, Bevegard andShepherd18 recently observed similar de-creases in heart rate at rest and during exer-cise when the baroreceptors were stimulatedby application of subatmospheric pressure tothe neck. Our findings are in accord withtheirs and indicate further that similar re-sponses occur at rest and during exercisewhen arterial pressure is decreased. It thusappears that the sensitivity of the barorecep-tor system to induced changes of pressuredoes not undergo any substantial alterationin the transition from rest to exercise: at anygiven level of arterial pressure, the heart rateduring exercise is higher than at rest, butwhen arterial pressure is changed, the mag-nitude of the alteration in heart rate is similarunder both conditions.

Since the baroreceptor system remains ac-tive during exercise, it undoubtedly continuesto play a modifying role in determining theheart rate, and presumably it also influencesthe level of inotropic stimulation of the myo-cardium. It might therefore be expected thatif the level of cardiac stimulation becameexcessive, the resultant increase in output andarterial pressure would lead to a reflex reduc-tion in sympathetic stimulation of the heartwhich would tend to lower the rate andcontractile state to more appropriate levels.In this way, the baroreceptor system mightbe visualized as providing the feedback neces-sary for adjusting the cardiac performanceduring exercise so that the output was pre-cisely adjusted to the fall in peripheral re-sistance, and so to the metabolic demand.Thus, even though the primary drive to in-

Rtsmrcb, Vol. XIX, August 1966



ownloaded from



creased cardiac activity during exercise aroseby some mechanism that was quite indepen-dent, the baroreceptor system might still beof crucial importance in determining the finallevel of cardiac stimulation attained.

AcknowledgmentWe would like to thank Mrs. Frankie Hayes for

her expert nursing assistance. Dr. A. Sahagian-Edwards of Ayerst Laboratories, Inc., New York,kindly supplied the propranolol (Inderal; ICI).

References1. KAHLER, R. L., THOMPSON, R. H., BUSKTRK, E.

R., FRYE, R. L., AND BRAUNWALD, E.: Studieson digitalis: VI. Reduction of the oxygen debtafter exercise with digoxin in cardiac patientswithout heart failure. Circulation 27: 397,1963.

2. EPSTEIN, S. E., ROBINSON, B. F., KAHLER, R. L.,AND BRAUNWALD, E.: Effects of beta-adrenergicblockade on the cardiac response to maximaland submaximal exercise in man. J. Clin. In-vest. 44: 1745, 1965.

3. GLJCK, G., AND BRAUNWALD, E.: Relative rolesof the sympathetic and parasympathetic ner-vous systems in the reflex control of heart rate.Circulation Res. 16: 363, 1965.

4. GASSER, H. S., AND MEEK, W. J.: A study of themechanism by which muscular exercise pro-duces acceleration of the heart. Am. J. Physiol.34: 48, 1914.

5. BRUCE, T. A., CHAPMAN, C. B., BAKER, O., ANDFISCHER, J. N.: The role of autonomic andmyocardial factors in cardiac control. J. Clin.Invest. 42: 721, 1963.

6. DONALD, D. E., AND SHEPHERD, J. T.: Responseto exercise in dogs with cardiac denervation.Am. J. Physiol. 205: 393, 1963.

7. ROBINSON, S., PEARCY, M., BRUECKMAN, F. R.,NICHOLAS, J. R., AND MILLER, D. I.: Effectsof atropine on heart rate and oxygen intake in

working man. J. Appl. Physiol. 5: 508, 1953.8. BISHOP, J. M., AND SECEL, N.: The circulatory

effects of intravenous pronethalol in man atrest and during exercise in the supine and up-right position. J. Physiol. (London) 169: 112,1963.

9. CHAMBERLAIN, D. A., AND HOWARD, J.: Thehemodynamic effects of /3-sympathetic block-ade. Brit. Heart J. 26: 213, 1964.

10. SCHRODER, C , AND WERKO, L.: Hemodynamicstudies and clinical experience with nethalide,a beta-adrenergic blocking agent. Am. J. Car-diol. 15: 58, 1965.

11. WARNER, H. R., TOPHAM, W. S., NICHOLES,K. K.: The role of peripheral resistance incontrolling cardiac output during exercise.Ann. N. Y. Acad. Sci. 115: 669, 1964.

12. TOPHAM, W. S., AND WARNER, H. R.: Effect oncardiac output of brachiocephalic constric-tion during rest and exercise. Physiologist 8:289, 1965.

13. MAREY, J.: Recherches sur le pouls au moyend'un nouvel appareil enregistreur le sphyg-mographe. Mem. Soc. Biol. Paris, Ser. 3, 1:281, 1859.

14. ROSENBLUETH, A., AND FREEMAN, N. E.: Thereciprocal innervation in reflex changes ofheart rate. Am. J. Physiol. 98: 430,"1931."

15. WANG, S. C , AND BORISON, H. L.: An analysisof the carotid sinus cardiovascular reflex me-chanism. Am. J. Physiol. 150: 712, 1947.

16. GELLHORN, E.: The significance of the state ofthe central autonomic nervous system forquantitative and qualitative aspects of somecardiovascular reactions. Am. Heart J. 67: 106,1964.

17. LJND, A. R., TAYLOR, S. H., HUMPHREYS, P. W.,KENNELLY, B. M., AND DONALD, K. W.: Thecirculatory effects of sustained voluntary mus-cle contraction. Clin. Sci. 27: 229, 1964.

18. BEVEGARD, B. S., AND SHEPHERD, J. T.: Circu-latory effects of stimulating the carotid arterialstretch receptors in man at rest and duringexercise. J. Clin. Invest. 45: 132, 1966.

CircnUtion Rtiircb, Vol. XIX, Angus! 1966



ownloaded from


BRAUNWALDBRIAN F. ROBINSON, STEPHEN E. EPSTEIN, G. David BEISER and EUGENE

Interrelation Between Baroreceptor Mechanisms and ExerciseControl of Heart Rate by the Autonomic Nervous System: Studies in Man on the

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1966 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.19.2.4001966;19:400-411Circ Res.

http://circres.ahajournals.org/content/19/2/400World 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/19/2/400

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

http://www.lww.com/reprints

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


control of hear t rat bye th autonomie c nervous...

Documents