effects of unilateral stellectomy upon cardiac performance during

637

Effects of Unilateral Stellectomy uponCardiac Performance during Exercise

in Dogs

PETER J. SCHWARTZ AND H. LOWELL STONE

SUMMARY The role of the right and left stellate ganglion (RSG, LSG) in the cardiac response toexercise was determined in 34 chronically instrumented dogs. The dogs were divided into three groups:control, left stellectomized (LSGx), and right stellectomized (RSGx). Heart rate (HR), left circumflexcoronary flow velocity (CF), and left ventricular pressure were measured during a graded submaximalexercise program on a motor-driven treadmill. At the greatest work load, 6.4 kph and 16% grade, HRin the control group was 235 ± 6 beats/min, after LSGx was 258 ± 6 beats/min (P < 0.05), and afterRSGx was 157 ± 7 beats/min (P < 0.01). The maximal derivative of left ventricular pressure increasedto 7373 ± 501 mm Hg/sec in the control group; after LSGx it reached 8233 ± 759 mm Hg/sec [notsignificant (NS)], and after RSGx, 5524 ± 305 mm Hg/sec (NS). Control CF increased to 48 ± 3 cm/sec,after LSGx it increased to 57 ± 3 cm/sec (P < 0.05), and after RSGx to 42 ± 5 cm/sec (NS). After RSGxmost dogs did not reach the greatest work load, and the comparison was made at the level attained.Arrhythmias during exercise appeared in 8% of control dogs, in 11% after LSGx, and in 86% after RSGx.It is concluded that: (1) sympathetic control of HR is mediated primarily by the RSG; (2) LSGx does notimpair myocardial contractility because of compensatory mechanisms exerted by the RSG, and (3)RSGx increases the likelihood of arrhythmias during exercise. Ore Res 44: 637-645, 1979

RECENTLY, it has been shown that unilateralright and left stellectomy have profound and op-posite influences on cardiac arrhythmias duringmyocardial ischemia (Schwartz et al., 1976c), ven-tricular vulnerability to fibrillation (Schwartz et al.,1976b), ventricular excitability (Schwartz et al.,1977), and myocardial reactive hyperemia(Schwartz and Stone, 1977). With one exception(Schwartz and Stone, 1977), these studies had theshortcoming that the conditions under which theeffects of unilateral stellectomy were investigatedwere artificial because of anesthesia (Schwartz etal., 1976a, 1976c, 1977) and vagotomy (Schwartz etal., 1976b, 1977). In the latter studies, vagotomy wasjustified by the obvious need to eliminate the tonicafferent vagal inhibition on sympathetic nerves(Guazzi et al., 1962; Mancia et al., 1975). Exerciserepresents a condition in which conscious animalsmay be studied repeatedly and in which the highlevel of sympathetic activity attained suggests thatchanges may be observed after removal of eitherthe right or left components of cardiac sympatheticinnervation.

The data presently available on the role of car-diac sympathetic innervation during exercise origi-nate from studies in which either complete dener-

From the Department of Physiology and Biophysics, University ofOklahoma Health Sciences Center, Oklahoma City, Oklahoma, and theIstituto Ricerche Cardiovascolari, Universiti di Milano, and Centro Ri-cerche Cardiovascolari, C.N.R, Milano, Italy.

Supported by Grants NHLBI-22518 and 18798.Dr Schwarti is an Adjunct Associate Professor on leave from the

University of Milan.Address for reprints: Peter J. Schwartx, M.D., Istituto Ricerche Car-

diovascolari, Unrversita di Milano, Via F. Sforxa, 35, 20122 Milano, Italy.Received November 18, 1977; accepted for publication November 13,

1978.

vation (Donald and Shepherd, 1963) or bilateralsympathectomy (Brouha et al., 1936; Hodes, 1939)were performed or beta-blocking drugs were used(Donald et al., 1968).

We have investigated the effects of unilateralright or left stellectomy on various aspects of car-diac performance during exercise to understandbetter the specific role of cardiac sympatheticnerves in the complex cardiac adjustments associ-ated with exercise and to gain insights into thepuzzling effects of unilateral stellectomy previouslyreported.

Methods

Surgical ProcedureForty mongrel dogs ranging in weight from 17 to

25 kg were used. They were free of heart worms andin good health prior to surgery. They were anesthe-tized with sodium pentothal (30 mg/kg, iv) andintubated. The level of anesthesia was maintainedwith a mixture of oxygen, nitrous oxide, and halo-thane. The heart was exposed through the 5th leftintercostal space. The left circumflex coronary ar-tery was dissected free of surrounding tissue fromits origin for approximately 2.5-3 cm. Usually suf-ficient lengths of artery were obtained for the im-plantation of the flow probe without sacrificinglarge epicardial branches. Great care was exercisedin the dissection of this vessel to maintain theadventitia intact; in so doing, adipose tissue sur-rounding the vessel was not removed. A Dopplerultrasonic flow probe was placed around the vesseland secured in place. A polyvinyl catheter waspositioned in the left atrium through the left atrialappendage. A solid state pressure transducer (Kon-

by guest on April 12, 2018

http://circres.ahajournals.org/D

ownloaded from

http://circres.ahajournals.org/

638 CIRCULATION RESEARCH VOL.44, No. 5, MAY 1979

igsberg, model P20) was positioned in the left ven-tricular cavity through an incision in the apicaldimple. All lead wires were brought out of the chestand tunneled to the dorsal surface of the neck wherethey exited from the skin.

In 22 dogs, the 3rd, right, or left intercostal spacewas entered subsequently. The left (LSG) or right(RSG) stellate ganglion, the ansa subclavia, and therami communicantes from T\ to T* were identified.A length of nylon monofilament suture was placedaround the caudal portion of the ganglion near theentry of the T3 ramus and a second suture wasplaced around the ansa subclavia at the cranialportion of the ganglion. Both monofilament sutureswere buried subcutaneously near the vertebrae onthe left or right side. Care was exercised to makecertain that the sutures remained loose around theganglion to prevent undue tension. In five dogs,both stellate ganglia were removed at the time ofsurgery. In 13 additional dogs, the same surgicalprocedure was done except the stellate ganglia werenever removed during the study.

MeasurementsRecordings of the left circumflex coronary artery

flow velocity, left ventricular pressure, and the elec-trocardiogram (lead 2) were made on an eight-chan-nel Beckman direct-writing oscillograph. The firstderivative of left ventricular pressure (dP/dt) wasobtained by an analog differentiator with a timeconstant of 0.01 second and a frequency responselinear to 65 Hz. This signal was recorded on thedirect-writing oscillograph. The electrocardiogramwas used to trigger a cardiotachometer to measureheart rate. The signals also were recorded on mag-netic tape (Ampex FR-1300) for later analysis.

ProtocolStudies were started 3—4 weeks after implanta-

tion. Submaximal exercise was carried out on amotor-driven treadmill with six periods of 3-minuteduration at increasing work loads. The first andsecond periods were at 0% incline while the speedincreased from 4.8 to 6.4 kph. At the latter speedthe incline of the treadmill was gradually increasedto 16%.

Twenty-five of these dogs were anesthetized withsodium pentothal and the nylon monofilament su-tures around the LSG (nine dogs) or RSG (threedogs) were exposed. Both sutures were pulled si-multaneously, resulting in the destruction of theLSG or RSG. Two weeks later the study duringexercise was repeated. All dogs were afebrile andnone had arrhythmias at rest.

Four additional dogs were studied after the re-moval of the RSG, but no control studies were madein this group. In three of these, the LSG subse-quently was removed to produce bilateral stellec-tomy (BSGx) and the study repeated 2 weeks later.Five dogs were studied only after BSGx. This rather

complicated protocol and the conditions in whichthe various dogs were studied are shown in Figure1.

Data AnalysisThe resting and exercise data were analyzed from

the analog magnetic tape recording by a digitalcomputer (PDP 11/40) which was programmed tosample the data every 5 msec over 5 beats. Thecomputer averaged the data points for the 5 beatsand also calculated the derivative of left ventricularpressure (dP/dt). The submaximal exercise testswere taken at the end of each 3-mjnute segment.

It has to be specified that the control data referto the moment at which the dogs, standing on thetreadmill, were ready to begin exercise. This con-dition is already accompanied by some degree ofsympathetic activity.

Statistical analysis was done in two differentways. For the entire left stellectomy (LSGx) studyand for part of the right stellectomy (RSGx) study,each dog was used as its own control and data wereanalyzed by Student's f-test for paired observations(internal control study). The analysis was also car-ried out on the three experimental groups of dogs(RSGx, LSGx, BSGx) using multiple analysis ofvariance (group study). All values are expressed asmeans ± SE.

ResultsOf the 40 dogs instrumented, four died of post-

surgical complications including acute infection,and technical failure in two dogs made the contin-uation of the study impossible. The results reportedhere are based, therefore, on 34 dogs in which thevarious devices implanted performed well for thenecessary period of time, ranging from 50 to 70 days

M6S INSTRUMENTED

JO

ONLY CONTROL13

ONLY CONT5

CONTROL +

RSGx3

RS6x

RSGx +

BSGx3

CONTROL + LS6x9

ONLY RSGx

1

FIGURE 1 Protocol summary. The numbers indicatehow many dogs were instrumented and under whichspecific conditions they were studied.



ownloaded from


UNILATERAL STELLECTOMY AND EXERCISE/Schwartz and Stone 639

MCF

cm/iec

6 0 -

6 0

4 0

3 0

LSGi

Control

C S 4.3/0 6.4/0 6.4/4 6.4/8 6.4/12 6.4/16

Kmph/porcent slop*

FIGURE 2 The average heart rate (HR) response tosubmaximal treadmill exercise in a group (n =» 25) ofcontrol dogs (control), dogs (n = 9) with the left stellateganglion removed (LSGx), and dogs (n - 7) with theright stellate ganglion removed (RSGx). The barsthrough the points represent 1 standard error of themean.

following surgery. Thirteen of this group were stud-ied only during the control period.

Heart Rate

In 25 intact dogs (control group), heart rate (HR)increased during exercise from 116 ± 5 to 235 ± 5beats/min at the highest level of exercise, whichwas 6.4 kph at 16% incline (Fig. 2).

In the nine dogs studied after LSGx, the increasein HR from 120 ± 10 to 258 ± 6 beats/min (Fig. 2)

was significantly higher (P < 0.05) at the threehighest levels of exercise when compared to thecontrol data (n — 25). Equally significant was thisdifference when the data before and after LSGxwere compared (internal control study), as shownin Table 1.

In seven dogs after RSGx, HR was already sig-nificantly lower when they were standing on thetreadmill before beginning exercise and, during ex-ercise, increased only from 86 ± 8 to 157 ± 7 beats/min (Fig. 2). The latter values refer to 6.4 kph at8% incline because only one of the seven dogscompleted the exercise up to 16% incline. Six dogswould not run and exercise was terminated. Thedifference with the control group (n = 25) wassignificant (P < 0.05) when the dogs were standingon the treadmill and became highly significant (P< 0.01) during the entire exercise period. Similardata with the same significant difference were ob-tained for the three dogs studied before and afterRSGx (internal control study), as shown in Table 2.

Myocardial ContractilityIn 22 control dogs, dP/dt max increased from

3655 ± 273 to 7379 ± 501 mm Hg/sec with exerciseat a level of 6.4 kph and 16% incline (Fig. 3).

In nine dogs after LSGx, dP/dt max increasedfrom 3841 ± 407 to 8233 ± 305 mm Hg/sec (at 6.4kph and 16%). The difference between the controlgroup and the LSGx group was not significant; thisalso was true for the internal control comparison(Table 1).

In six dogs after RSGx, dP/dt max increasedfrom 3750 ± 378 to 5524 ± 305 mm Hg/sec at 6.4kph and 12% grade, and the difference from 6885± 354 mm Hg reached by the control group at thesame level of exercise was not statistically signifi-cant. There was no difference at all between cdntrolcondition and after RSGx in three dogs studiedbefore and after RSGx, with the only exceptionbeing in the standing condition when dP/dt maxwas greater after RSGx (4033 ± 263 mm Hg/sec)than in control (3280 ± 490 mm Hg/sec). These

TABLE 1 Effect of LSGx in Nine Dogs on the Response to Various Levels of Exerciseon a Motor-Driven Treadmill

(kph)

04.86.46.46.46.46.4

(%)

00048

1216

4027570761636368674874328248

(mm

C

±588± 702±699±693±674±863± 1153

PHg/«ec)

LSGx

3936 ±5204946 ± 5155350 ±3416096±3866565±5597319 ± 6567946 ±860

25364042454649

C

±±±±±±±

MCF(cm/sec)

2233436

LSGx

25±235±337 ±243 ± 347 ± 351 ±4*56 ± 7*

C

113 ±180 ±198 ±211 ±218 ±233±243±

HR(beata/min)

86788710

LSGx

120 ± 10172 ± 8191 ± 8206±8229 ± T245 ±6*258 ±6*

C = control, LSGx = left Btellectomy, P - maximal rate of rise of left ventricular pressure, MCF•• mean left circumflex coronary flow velocity, HR — heart rate. All values are mean ± 1 standarderror.

• Paired P < 0.05, experimental compared to control.



ownloaded from


640 CIRCULATION RESEARCH VOL. 44, No. 5, MAY 1979

TABLE 2 Effect of RSGx in Three Dogs on the Response to Various Levels of Exerciseon a Motor-Driven Treadmill

(kph)

04.86.46.46.4

(%)

00048

32804390478349405165

P(mm Hg/sec)

C

±849± 1023±1299± 1103± 1379

RSGx

4033 ±455*4548 ±5544696 ± 12474878 ±9444927 ± 671

1930313334

MCF(cm/sec)

C

± 3± 8± 4± 8± 9

RSGx

19 ± 131 ± 932 ± 1138 ± 1 039 ±11

116189196201215

C

±±±±±

1HR(beata/min)

35101346

RSGx

9 0 ±143 ±152 ±153 ±160 ±

17*12*14*3*11*

RSGx - right stellectomy. Other abbreviations and 'footnote are as in Table 1.

data are shown in Figure 3 for the group study andin Table 2 for the internal control study.

Coronary Blood FlowIn 25 control dogs mean coronary flow velocity

(MCF) increased from 26 ± 1 to 48 ± 3 cm/secduring exercise at a level of 6.4 kph and 16% incline(Fig. 4).

In nine dogs after LSGx, MCF increased from 25± 2 to 56 ± 7 cm/sec (Table 1). The values for MCFwere significantly different after LSGx at the twohighest levels of exercise in a comparison with

HRb /mtn

280

2 6 0

2 0 0

180

160

140

120

100-

8 0

LSG,

RSG,

S 4.8/0 6.4/0 6.4/4 &4/8 6.4/12 6.4/16Kmph/pirctnt slop*

FIGURE 3 The average maximal rate of rise of leftventricular pressure (dP/dt) response to submaximalexercise in the same three groups of dogs shown inFigure 1 [control (n - 22), LSGx (n •= 9), and RSGx (n= 6)J. The bars through the points represent 1 standarderror of the mean.

either the internal control data or control groupdata.

In six dogs after RSGx, MCF increased duringexercise from 24 ± 3 to 42 ± 5 cm/sec (Fig. 4), andthe difference from the control group was not sig-nificant. In the three dogs studied before and afterRSGx, the MCF was slightly higher after RSGx,but the difference was still not significant. Thesedata are shown in Table 2.

Left Ventricular (LV) PressureLV pressure was analyzed in nine dogs before

and after LSGx. LV systolic pressure increased un-der control conditions from 126 ± 11 to 156 ± 17mm Hg and after LSGx, LV systolic pressure in-creased from 122 ± 12 to 149 ± 20 mm Hg and thedifference was not significant. An exception to thiswas the first 3-minute period (4.8 kph and 0%incline) when under control conditions LV systolicpressure was significantly (P < 0.05) higher thanafter LSGx (143 ± 12 vs. 132 ± 11 mm Hg). LVdiastolic pressure averaged 4 ± 1 mm Hg in thecontrol group and increased to 5 ± 2 mm Hg at the

mmHg/ilc

9,000-1

8,000

7,000-

8,000

5,000

4,000-

LSG,

Control

RSG.

S 4.8/0 6.4/0 6.4/4 6.4/8 6.4/12 6.4/16Kmph/p•rc•nt 11op•

FIGURE 4 The mean left circumflex coronary arteryflow velocity (MCF) changes to submaximal exercise inthe same groups of dogs given in Figure 1 [control (n «•25), LSGx (n = 9), and RSGx (n = 6)J. The bars throughthe points represent 1 standard error of the mean.



ownloaded from



highest work load. Following LSGx, LV diastolicpressure was 4 ± 2 mm Hg and increased to 5 ± 3mm Hg at the highest work load. Neither the in-crease in LV diastolic pressure with exercise norcomparison before exercise between control andLSGx was significantly different. LV systolic anddiastolic pressures did not change during exerciseafter RSGx in the six dogs in comparison to thecontrol group and the internal control data.

ArrhythmiasArrhythmias, mostly ventricular premature

beats, never were present before exercise in the 34dogs studied. They appeared during exercise in twoout of 25 control dogs (8%), in one of nine leftstellectomized dogs (11%), and in six of seven rightstellectomized dogs (86%). In most dogs, arrhyth-mias appeared during the first phase of exercise; asexercise progressed, however, the pattern was lost.In some dogs they continued and in some they didnot. Also, the incidence of premature beats variedfrom few to many. With one possible exception,they never seemed so frequent as to interfere withcardiac function and to limit exercise performance.Particularly, in the right stellectomized dogs whichdid not complete exercise, arrhythmias were verysporadic in the 3 minutes preceding the momentwhen they quit running.

Bilateral Stellectomy (BSGx)Five dogs were studied only after BSGx, and the

data from these are given in Table 3. Heart rate anddP/dt max increased significantly less than in thecontrol group (191 ± 4 vs. 235 ± 5 beats/min and5014 ± 476 vs. 7373 ± 501 mm Hg/sec), yet coronaryflow increased slightly above that observed in thecontrol group (51 ± 5 cm/sec vs. 48 ± 3 cm/sec).The LV systolic pressure increased from 92 ± 6 to120 ± 4 mm Hg at the highest level of exercise,while LV diastolic pressure increased from 4 ± 1 to7 ± 3 mm Hg. The increase in LV systolic pressurewith exercise was significant (P < 0.05), whereasthe increase in LV diastolic pressure was not. Thesystolic LVP was higher in the control group thanin the BSGx group. The comparison of the BSGxgroup to the control group showed a significant (P< 0.01) difference in pre-exercise systolic pressure

and the systolic pressure (P < 0.05) at the maximallevel of exercise. The LV diastolic pressures of thetwo groups were not different during the pre-exer-cise or exercise periods.

The effects of unilateral and bilateral stellectomywere compared, using as a reference the controlvalues. For RSGx (n = 3) and LSGx (n - 9) theinternal control study was used, whereas for BSGx(n = 5) the lack of internal control made necessarya comparison with the control group (n = 25). It isimportant to recall that in the RSGx group thenumber of dogs was small, and thus the comparisonwas made at the level of exercise attained (6.4 kph,8%). HR was 26% lower after RSGx, 19% lowerafter BSGx, and 6% higher after LSGx; dP/dt maxwas 5% lower after RSGx, 4% lower after LSGx, and32% lower after BSGx; MCF was higher in all threegroups (15% after RSGx, 14% after LSGx, and 6%after BSGx).

DiscussionThe rapid changes in heart rate and myocardial

contractility associated with initiation of exerciseare dependent upon the presence of an intact car-diac sympathetic innervation (Donald, 1974; Vatnerand Pagani, 1976). Without cardiac nerves thesechanges, although delayed, still occur and allow anapparently normal cardiac performance during ex-ercise because of changes in stroke volume andcirculating catecholamine levels (Donald, 1974).

Current knowledge of the role of the sympatheticnervous system during exercise stems mostly fromstudies in which either a bilateral sympathectomyor a more extensive cardiac denervation was per-formed (Donald and Shepherd, 1963; Brouha et al.,1936; Hodes, 1939). The studies with cardiac dener-vation yielded information concerning the rapidityin the adjustments to the increased circulatory de-mands, and the studies with bilateral stellectomyshowed increased fatigue during exercise (Hodes,1939) and impairment in HR control (Brouha et aL,1936; Hodes, 1939; Samaan, 1935). These classicstudies, however, considered the sympathetic nerv-ous system as a whole and do not provide infor-mation on possible specific roles of the variouscomponents of cardiac sympathetic innervation.

There has been, in fact, a growing and by now

TABLE 3 Effect of BSGx in 5 Dogs on the Response to Various Levels of Exercise on aMotor-Driven Treadmill

Speed(kph)

04.86.46.46.46.46.4

Grade(%)

00048

1216

P(mm Hg/aec)

2682 ±3683619 ± 7944127 ± 4314100 ± 3134786 ±3094622 ± 114*5014 ± 476*

MCF(cm/sec)

30±442 ±646±745±745 ±648±551 ±5

HR(beata/min)

88 ±2*146 ±3*159 ±2*164 ±3*174 ± 3*182 ±4*191 ± 4*

BSGi = bilateral stellectomy, other abbreviations and 'footnote are as in Table 1. Statisticalcomparison was made between the control group of dogs and BSGi dogs.



ownloaded from



definitive body of evidence that the sympatheticnervous system is highly specific in both its efferentand afferent components. Randall and Rohse(1956), using electrical stimulation, showed thatsympathetic control of HR is primarily mediatedby RSG, whereas both ganglia contribute to myo-cardial contractility, possibly the left more than theright. Malliani and associates, by electrophysiolog-ical studies, showed that various hemodynamicstimuli affect receptors, strategically located in dif-ferent parts of the cardiovascular system, whichrelay information to the spinal cord and elicit highlyspecific sympathetic reflexes (Malliani et al., 1972,1973a, 1973b; Pagani et al., 1974; Lioy et al., 1974).Furthermore, they showed reciprocal and synergis-tic effects of the cardiac sympathetic afferent fiberson the vagal outflow directed to the heart (Schwartzet al., 1973).

Our present study, utilizing unilateral stellectomyas a tool, provides new insights into the interactionoccurring during exercise between the right and leftcardiac nerves and into their specific roles.

Exercise PerformanceWe found an unexpected and striking difference

between the LSGx dogs, all of which completed theexercise, and the RSGx dogs, six of seven of whichcould not complete the last portion of exercise. Wehave no explanation for this finding; yet an intri-guing observation has been made in an ongoingclinical study (Schwartz, unpublished observation)in patients with unilateral stellectomy. Dyspnea oneffort was observed in 19% of 25 patients with RSGxand in none of 11 patients with LSGx; all patientswere under 50 years of age. Thus, even if an expla-nation is still missing, a similar occurrence has beenfound in both dogs and humans.

For the sake of discussion, the possibility may beproposed that the lower HR achieved during exer-cise after RSGx might have limited the necessaryincrease in cardiac output, thus inducing dyspneaand interfering with exercise performance.

Heart RateThe most obvious finding was reduction in HR

in the standing position on the treadmill and adiminished increase in HR during submaximal ex-ercise after RSGx as compared to controls. Thisconfirms in conscious animals what was found inanesthetized preparations (Randall et al., 1956) and,by proving the crucial role of the RSG in thesympathetic control of HR, explains per se theimpaired HR response during exercise observed bySamaan (1935) and by Hodes (1939) in their bilat-erally sympathectomized animals. To our surprise,the LSGx dogs at the higher levels of exerciseconsistently attained a higher HR than did controls.A related observation, of anecdotal interest, wasmade by Jonnesco (1921) who found that removingthe LSG in a patient with angina pectoris produced

a marked increase in HR which was still evidentseveral years after surgery.

This finding might be explained in at least twoways. The first would be a postdenervation super-sensitivity to circulating catecholamines, and thesecond would be an increase in sympathetic activitythrough the contralateral stellate ganglion. The firstmechanism is highly unlikely, not only because ofthe anatomical distribution of cardiac nerves to thepacemaker areas, but especially because of veryrecent data indicating that suprasensitivity doesnot develop after LSGx (Stone et al., 1978). Thesecond possibility implies a reflex activation of de-scending sympathetic activity running through theintact ganglion after unilateral stellate ganglionec-tomy. Although not proven, this mechanism hassome data in its favor. The paradoxical effects ofRSGx (Schwartz et al., 1976b, 1976c, 1977) dependupon an intact LSG, and it previously was suggested(Schwartz et al., 1977) that a tentative explanationmight be a baroreceptive reflex. Following the abla-tion of one stellate ganglion, HR and/or systemicarterial pressure decrease, thus reducing carotidsinus baroreceptor activity. This results in in-creased sympathetic activity through the contralat-eral ganglion. Therefore, unilateral stellectomy,while suppressing sympathetic activity on the ipsi-lateral side, may increase it on the contralateralside. Whereas a baroreceptive mechanism is plau-sible under resting conditions, to accept such amechanism during exercise is more difficult becauseit has been shown that in this condition baroreflexesare attenuated (McRitchie et al., 1976). Anothermechanism by which sympathetic activity may beincreased following contralateral stellectomy is sug-gested by the work of McCall and Gebber (1977).They have recently found that spinal interneuronslocated in the vicinity of the intermediomedial nu-cleus directly inhibit part of the sympathoexcita-tory pathways located in the intermediolateralspinal nucleus. They have also demonstrated thatthese inhibitory interneurons receive part of theirinput from sympathetic afferents running in theinferior cardiac nerve. Thus, removal of these affer-ents, as occurs with stellectomy, would interruptthe inhibitory pathway and result in an increasedsympathetic activity through the contralateral stel-late ganglion.

The fact that by interrupting the ansa subclaviasome left vagal fibers are destroyed should, in ouropinion, play no role whatsoever in the observedresponse.

Myocardial ContractilityRemoval of either right or left SG did not result

in a major reduction of the contractility of the leftventricle. In addition to the possibilities just men-tioned for HR (i.e., increased sensitivity to circulat-ing catecholamines and reflex action of the contra-lateral ganglion), other mechanisms may be in-



ownloaded from



volved. A HR effect may account in part for themaintenance of dP/dt max after LSGx, but sincethis mechanism may not be involved in right stel-lectomized animals, this seems an unlikely expla-nation.

Left ventricular end-diastolic pressure did notincrease substantially after either RSGx or LSGxand, therefore, this potential mechanism did notplay a role in our studies. Another possibility whichcannot be discounted and which deserves furtherinvestigation is represented by an increase in car-diac dimensions (JewelL 1977). The fact that, afterBSGx, dP/dt max fell by 32% compared to controlwhile the decrease after unilateral right and leftSGx were, respectively, 5% and 4%, strongly sug-gests that a compensatory activation of the intactganglion might have played a decisive role. Thiswould coincide with other findings (Randall andRohse, 1956) that both right and left stellate gangliacontribute to myocardial contractility.

Coronary Blood FlowChanges in coronary blood flow result primarily

from an increase or decrease in myocardial oxygenconsumption (Sonnenblick et al., 1968). Myocardialcontractility, HR, and ventricular wall tension de-velopment have been considered to be the mostimportant factors influencing myocardial oxygenconsumption. During exercise, myocardial oxygenconsumption increases (Khouri et al., 1965; Stone,1977) since myocardial contractility, tension, andHR all increase. Thus, in the control study, theincrease in coronary blood flow could be explainedby the increase in myocardial oxygen consumption.

After LSGx, coronary blood flow increased moreat the highest two levels of exercise than in thecontrol group. There may be at least two possibleexplanations for the difference. The first would bethat the increase in coronary flow was due to anincrease in myocardial oxygen consumption greaterthan during the control runs as a result of theincrease in HR during exercise above the prestel-lectomy controls. Previous data (Stone, 1977; Vat-ner et al., 1972) have indicated that HR alone cancontribute 27-33% of the increase in coronary flowand myocardial oxygen consumption during exer-cise. A second possible explanation would be theremoval of a vasoconstrictor tone on the coronaryvessels by LSGx (Schwartz and Stone, 1977; Feigl,1967). In a previous study (Schwartz and Stone,1977) it was shown that the coronary reactive hy-peremic response to a 10-second occlusion of thecircumflex artery was increased following LSGx andthat the same response could be mimicked by a-adrenergic receptor blockade in the conscious dog.The peak of the reactive hyperemic flow responsewas not altered following LSGx in that study, whichsuggests that the maximal downstream bed size hadnot changed. These data suggested for the first timea tonic vasoconstriction in the coronary vessels

during short periods of ischemia, which is known toelicit a reflex increase in cardiac sympathetic activ-ity (Malliani et al., 1969). Two recent studies inexercising dogs (Gwirtz and Stone, 1978; Murrayand Vatner, 1978) concluded that coronary bloodflow can be increased by a-adrenergic blockade,thus supporting the concept of a tonic sympatheticvasoconstrictor activity on the coronary vessels dur-ing submaximal and maximal exercise. It is difficultto separate these two possibilities without a mea-sure of myocardial oxygen consumption.

Another line of evidence that suggests a basicchange in the regulation of coronary flood flowduring exercise comes from the five dogs that weresubjected to BSGx. In this group, HR and contrac-tility were reduced at the maximal level of exercisestudied. Yet the coronary flow was slightly in-creased over that found in the control group at thesame level of exercise. Since a decrease in HR andcontractility would reduce myocardial oxygen con-sumption, coronary flow would be expected to de-crease below the control group data at this point.The data from this group strongly suggest thatstellectomy has removed some vasoconstrictor ef-fect that was still present in the coronary vessels atsubmaximal workloads in the control condition andindicate an important role of cardiac sympatheticnerves in limiting coronary flow during exercise.

ArrhythmiasRSGx clearly favored the development of ar-

rhythmias during exercise. This finding representsan important confirmation of what already hadbeen observed during myocardial ischemia(Schwartz et al., 1976), because anesthesia and va-gotomy can no longer be implicated. An obviouspossible explanation would be that the lower HRafter RSGx unmasked the arrhythmias not evidentin the control condition or after LSGx because ofoverdrive suppression. The fact that, after BSGx,despite the same low level of HR, arrhythmias didnot appear militates against this explantion; how-ever, the number of dogs with BSGx is small. Atvariance with what happens during myocardial is-chemia (Schwartz et al., 1976), where dorsal rootsection decreased the incidence of arrhythmias, itis unlikely that the section of sympathetic afferentfibers (Malliani et al., 1973b) may have played animportant role.

Once more (Schwartz, 1976b, 1976c) it seems thatunilateral sympathetic discharges mediated by theleft nerves facilitate arrhythmias. Two very recentsets of data in conscious animals suggest this. Catswere found to present arrhythmias during emo-tional stresses only after RSGx (Schwartz andStone, 1978). In dogs whose hearts are completelydenervated with the exception of the ventrolateralcardiac nerve, which originates from the left stellateganglion, arrhythmias are often produced by exer-cise (Randall et al., 1978). This latter model is



ownloaded from



limited by the absence of the vagi, but it also pointsagain to left-sided cardiac sympathetic nerves asarrhythmogenic.

This study has demonstrated in conscious dogsthat the sympathetic control of HR is primarilymediated by the RSG, that arrhythmias duringexercise are favored or unmasked by RSGx, andthat compensatory mechanisms—clearly evident onmyocardial contractility—are exerted after unilat-eral stellectomy by the contralateral ganglion. It isnot excluded that some of the cardiovascular ad-justments to exercise depend in part upon infor-mation relayed to the central nervous system bythe cardiac sympathetic afferent fibers (Malliani etal., 1973b) and that part of the results observeddepend on these fibers and therefore on the inter-ruption of cardio-cardiac reflexes (Malliani et aL,1972, 1973a; Schwartz et aL, 1973).

The data on HR, dP/dt max, and arrhythmiasindicate that sympathetic activity through rightand left cardiac nerves is not independent and thatimportant interactions may occur. The interactionsdepend either on baroreceptors reflexes (Schwartzet aL, 1977) or on other central mechanisms (in-cluding also the cardio-cardiac reflexes) still to bedefined.

Our findings, by showing the preservation of agood myocardial performance and even increasedcoronary flow after LSGx, may also have clinicalimplications because it has been recently suggested(Schwartz et al., 1975, 1976b; Schwartz and Stone,1978; Schwartz, 1978) that LSGx may play a pro-tective role in patients at high risk of sudden deathdue to ventricular fibrillation.

AcknowledgmentsWe would like to thank Gerald Todd, Hal Jackson, and

Delvin Knight for technical assistance in conducting these ex-periments and Dr. Virginio Mandelli for assistance in statisticalanalysis of the data.

ReferencesBrouha L, Cannon WB, Dill DB: The heart rate of the sympa-

thectomized dog in rest and exercise. J Physiol (Lond) 87:345-349, 1936

Donald DE, Shepherd JT: Response to exercise in dogs withchronic denervatdon. Am J Physiol 205: 393-400, 1963

Donald DE, Ferguson DA, Milburn SE: Effect of beta-adrenergicreceptor blockade on racing performance of greyhounds withnormal and denervated hearts. Circ Res 22: 127-134, 1968

Donald DE: Myocardial performance after excision of the ex-trinsic cardiac nerves in the dog. Circ Res 34: 417-424, 1974

Feigl EO: Sympathetic control of coronary circulation. Circ Res20: 262-271, 1967

Guazzi M, Libretti A, Zanchetti A: Tonic reflex regulation of thecat's blood pressure through vagal afferents from the cardio-pulmonary region. Circ Res 11: 7-16, 1962

Gwirtz PA, Stone HL: Effect of alpha-blockade on cardiac per-formance during exercise (abstr). Fed Proc 37: 664, 1978

Hodes R: Exercise in the sympathectomized cat Am J Physiol126: 171-179, 1939

Jewell BR: A reexamination of the influence of muscle length onmyocardial performance. Circ Res 40: 221-230, 1977

Jonnesco T: Traitement chirurgical de l'angine de poitrine parla resection du sympathique cervico-thoracique. Presse Med20: 193-194, 1921

Khouri EM, Gregg DE, Rayford CR: Effects of exercise oncardiac output, left coronary flow and myocardial metabolismin the unanesthetized dog. Circ Res 17: 427-137, 1965

Iioy F, Malliani A, Pagani M, Recordati G, Schwartz PJ: Reflexhemodynamic responses initiated from the thoracic aorta. CircRes 34: 78-84, 1974

Malliani A, Schwartz PJ, Zanchetti A; A sympathetic reflexelicited by experimental coronary occlusion. Am J Physiol217: 703-709, 1969

Malliani A, Peterson DF, Bishop VS, Brown AM: Spinal sym-pathetic cardio-cardiac reflexes. Circ Res 30: 158-166, 1972

Malliani A, Parks M, Tuckett RP, Brown AM: Reflex increasesin heart rate elicited by stimulation of afferent cardiac sym-pathetic nerve fibers in the cat Circ Res 32: 9-14, 1973a

Malliani A, Recordati G, Schwartz PJ: Nervous activity of affer-ent cardiac sympathetic fibers with atrial and ventricularendings. J Physiol (Lond) 229: 457-169, 1973b

Mancia G, Shepherd JT, Donald DE: Role of cardiac, pulmo-nary, and carotid mechanoreceptors in the control of hindlimband renal circulation in dogs. Circ Res 37: 200-208, 1975

McCall RB, Gebber GL, Barman SM: Spinal interneurons in thebaroreceptor reflex arc. Am J Physiol 232: H657-H665, 1977

McRitchie RJ, Vatner SF, Boettcher D, Heyndricks GR, PatrickTA, Braunwald E: Role of arterial baroreceptors in mediatingcardiovascular response to exercise. Am J Physiol 230: 85-89,1976

Murray PA, Vatner SF: Alpha receptor attenuation of coronaryvascular response to severe, spontaneous exercise (abstr). FedProc 37: 664, 1978

Pagani M, Schwartz PJ, Banks R, Lombardi F, Malliani A:Reflex responses of sympathetic preganglionic neurons initi-ated by different cardiovascular receptors in spinal animals.Brain Res 68: 215-225, 1974

Randall WC, Rohse WG: The augmentor action of the sympa-thetic cardiac nerves. Circ Res 4: 470-475, 1956

Randall, WC, Thomas JX, Euler DE, Rozanski GJ: Cardiacdysrhythmias associated with autonomic nervous system im-balance in the conscious dog. In Neural Mechanisms in Car-diac Arrhythmias, edited by PJ Schwartz, AM Brown, AMalliani, A Zanchetti. New York, Raven Press, 1978, pp 123-138

Samaan A: Muscular work in dogs submitted to different con-ditions of cardiac and splanchnic innervations. J Physiol(Lond) 83: 313-331, 1935

Schwartz PJ, Pagani M, Lombardi F, Malliani A, Brown AM: Acardio-cardiac sympathovagal reflex in the cat Circ Res 32:215-220, 1973

Schwartz PJ, Periti M, Malliani A: The long Q-T syndrome. AmHeart J 89: 378-390, 1975

Schwartz PJ, Foreman RD, Stone HL, Brown AM: Effect ofdorsal root section on the arrhythmias associated with coro-nary occlusion. Am J Physiol 231: 923-928, 1976a

Schwartz PJ, Snebold NG, Brown AM: Effect of unilateralcardiac sympathetic denervation on the ventricular fibrillationthreshold. Am J Cardiol 37: 1036-1040, 1976b

Schwartz PJ, Stone HL, Brown AM: Effects of unilateral stellateganglion blockade on the arrhythmias associated with coro-nary occlusion. Am Heart J 92: 589-599, 1976c

Schwartz PJ, Stone HL: Tonic influence of the sympatheticnervous system on myocardial hyperemia and on coronaryblood flow distribution in dogs. Circ Res 41: 51-58, 1977

Schwartz PJ, Verrier RL, Lown B: Effect of stellectomy andvagotomy on ventricular refractoriness in dogs. Circ Res 40:536-540, 1977

Schwartz PJ: Unilateral stellectomy, cardiac arrhythmias andsudden death. In Primary and Secondary Angina Pectoris,edited by A Maseri, J Kla&sen, M Lesch. New York, Grune &Stratton, 1978, pp 417-425

Schwartz PJ, Stone HL: Unilateral stellectomy and suddendeath. In Neural Mechanisms in Cardiac Arrhythmias, editedby PJ Schwartz, AM Brown, A Malliani, A Zanchetti NewYork, Raven Press, 1978, pp 107-122

Sonnenblick EH, Ross J, Braunwald E: Oxygen consumption ofthe heart Newer concepts of its multifactorial determination.Am J Cardiol 22: 328-336, 1968



ownloaded from


RENAL NERVES MODULATE NONNEURAL RENIN SECRETION/Thames and DiBona 645

Stone HL: Myocardial oxygen consumption during graded ex- Vatner SF, Higgins CB, Franklin D, Braunwald E: Role ofercise (abstr). Fed Proc 36: 449, 1977

Stone HL, Dowell RT, Sordahl LA, Schwartz PJ: The role ofthe left stellate ganglion in cardiac performance during exer-cise. In Neural Mechanisms in Cardiac Arrhythmias, editedby PJ Schwartz, AM Brown, A Malliani, A ZanchettL NewYork, Raven Press, 1978, pp 221-236

tachycardia in mediating the coronary hemodynamic responseto severe exercise. J Appl Physiol 32: 380-385, 1972

Vatner SF, Pagani M: Cardiovascular adjustments to exercise:Hemodynamics and mechanisms. Prog Cardiovasc Dis 19: 91-107, 1976

Renal Nerves Modulate the Secretion of ReninMediated by Nonneural Mechanisms

MARC D. THAMES AND GERALD F. DIBONA

SUMMARY We investigated the role of efferent renal nerve activity in modulating the responses ofrenin secretion rate to suprarenal aortic constriction and to furosemide administration in female dogs.After the renal nerves had been sectioned, constriction of the suprarenal aorta decreased renalperfusion pressure (from 132 ± 5 to 51 ± 2 mm Hg) and renal blood flow (from 303 ± 21 to 149 ± 18 ml/min) and increased renin secretion rate (from 184 ± 49 to 2012 ± 499 ng/min). Stimulation of the renalnerves at very low frequency (0.25 Hz) had no significant effect on basal renin secretion rate, arterialpressure, renal blood flow, or urinary sodium excretion. Aortic constriction during this low level renalnerve stimulation resulted in a significant augmentation in the renin secretion rate response (from 358± 107 to 6988 ± 1600 ng/min), whereas the changes in renal blood flow and renal perfusion pressurewere similar to those observed without nerve stimulation. Similarly, low level renal nerve stimulation(0.25 Hz) was found to augment the renin secretion rate response to the intravenous administration offurosemide (1.0 mg/kg bolus followed by 0.017 mg/kg per min). These data show that a very low levelof renal nerve activity which by itself does not change arterial pressure, renal blood flow, urinarysodium excretion, or renin secretion rate, augments the release of renin in response to suprarenalaortic constriction or furosemide. Furthermore, these data provide direct evidence that the renal nervesmodulate renin release mediated through nonneural mechanisms. Additional data are presented whichshow how these observations account for previously reported differences in renin secretion rateresponses of innervated and denervated kidneys during aortic constriction and furosemide administra-tion. Circ Rea 44: 645-652, 1979

THE RELEASE OF renin from the kidney hasbeen shown to be largely mediated by threemechanisms (Davis and Freeman, 1976; Reid et al.,1978; Vander, 1967; Zanchetti and Stella, 1975). Thefirst of these is the intrarenal vascular receptor(baroreceptor), which senses changes in renal per-fusion pressure and induces an increase in reninsecretion when the renal perfusion pressure is de-creased (Davis and Freeman, 1976; Reid et al., 1978;Vander, 1967; Zanchetti and Stella, 1975). The sec-ond is the macula densa mechanism, mediated byspecialized cells in the distal tubule that are sensi-tive to changes in sodium chloride transport. Fac-tors which tend to reduce distal tubular sodium

From the Cardiovascular Center and Department of Internal Medi-cine, University of Iowa, and the Veterans Administration Hospital, IowaCity, Iowa.

Supported by Grants HL 21158, HL 14388, and AM 16843, and bygrants from the Iowa Heart Association and the Veterans Administration.Dr Thames is the recipient of a Research Career Development Awardfrom the National Heart, Lung, and Blood Institute.

Address for reprints. Marc D. Thames, M.D., University of IowaHospitals, Iowa City, Iowa 52242.

Received May 30, 1978; accepted for publication November 30, 1978.

chloride transport increase renin secretion via thismechanism (Davis and Freeman, 1976; Reid et al.,1978). The third mechanism for renin release in-volves the renal nerves (Davis and Freeman, 1976;Reid et al., 1978; Vander, 1967; Zanchetti and Stella,1975). Increases in renal sympathetic nerve activityhave been shown to evoke increases in renin secre-tion (Bunag et al., 1966), even in the absence ofchanges in renal blood flow or glomerular filtrationrate (LaGrange et al., 1973). To evaluate the role ofeach of these mechanisms in the control of reninsecretion, each has been studied in a fashion whichallowed it to be evaluated independently of theinfluence of the other two mechanisms. Recentevidence indicates that these vascular, tubular, andsympathetic neural mechanisms controlling reninsecretion can, under certain circumstances, functionindependently of each other (Osborne et aL, 1977).Since most investigations have examined one ofthese three specific mechanisms, there has beenlimited investigation of possible interactions amongthem.

Moreover, in spite of all the evidence for neurally



ownloaded from


P J Schwartz and H L StoneEffects of unilateral stellectomy upon cardiac performance during exercise in dogs.

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1979 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.44.5.6371979;44:637-645Circ Res.

http://circres.ahajournals.org/content/44/5/637.citationWorld 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/44/5/637.citation

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

http://www.lww.com/reprints

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


effects of unilateral stellectomy upon cardiac performance during

Documents