Ž .Journal of the Autonomic Nervous System 65 1997 10–16

Simultaneous cardiac and renal sympathetic neural responses toactivation of left ventricular sympathetic afferents

Rajesh Malik, Anthony J. Minisi )

( )Department of Cardiology 111J , Medical College of VirginiarVirginia Commonwealth UniÕersity and Hunter Holmes McGuire Department of VeteransAffairs Medical Center, 1201 Broad Rock BÕld., Richmond, VA 23249, USA

Received 15 February 1996; revised 20 February 1996; accepted 26 March 1997

Abstract

In separate sets of experiments, we observed that activation of left ventricular sympathetic afferents by transmural myocardialŽ . Ž .ischemia TMI appeared to elicit greater reflex increases in efferent sympathetic nerve activity SNA to the heart than to the kidney. To

assess this observation more rigorously, we simultaneously measured changes in cardiac and renal SNA elicited by TMI and by epicardialŽ .and intracoronary bradykinin BK . Experiments were performed in 19 chloralose-anesthetized dogs with sinoaortic denervation and

vagotomy. TMI was created by a 2 min complete occlusion of the anterior descending coronary artery while a collateral flow limitingŽ 2.stenosis was present on the circumflex coronary artery. Epicardial BK was applied to small sponges 1 cm which were placed on the

anterior wall of the left ventricle. Intracoronary BK was injected into a branch of the anterior descending artery. We observed that meanŽmaximal reflex increases in SNA during TMI and intracoronary BK were significantly greater in cardiac than in renal nerves TMI;

.58"11% versus 36"9%, ps0.01; intracoronary BK; 144"48% versus 77"26%, ps0.05 . Epicardial BK elicited reflex increasesŽ .in cardiac and renal SNA which were not significantly different 167"44% versus 127"36%; ps0.72 . Our results indicate that

activation of left ventricular sympathetic afferents by TMI and intracoronary BK elicits greater reflex increases in sympathetic outflow tothe heart than to other end-organs such as the kidney. We speculate that these augmented excitatory responses are most likely related toengagement of cardio–cardiac spinal sympathetic reflexes during intense stimulation of sympathetic afferent endings. q 1997 ElsevierScience B.V.

Keywords: Neural control of the circulation; Sympathetic nerve activity; Cardiac receptors; Sympathetic afferents

1. Introduction

Coronary occlusion and myocardial ischemia elicit re-flex cardiovascular responses that are the result of activa-tion of sensory receptors located in left ventricular my-ocardium. The left ventricle contains two distinct types ofsensory receptors. The first group comprises receptorssubserved by afferent fibers that travel to the central

Ž .nervous system via vagal nerves vagal afferents . Activa-tion of these receptors elicits sympatho-inhibitory and

w xvasodepressor responses 35 . A second group of ventricu-lar receptors is subserved by afferent fibers that travel tothe spinal cord and central nervous system via the sympa-

Ž .thetic nerves sympathetic afferents . Stimulation of thesereceptors results in reflex vasopressor and sympatho-exci-

w xtatory responses 7 . In addition to being activated by

) Corresponding author. Tel.: q1 804 6755448.

myocardial ischemia, these receptors are also activated bywseveral chemicals, including bradykinin 3,15,16,25,28,

x32,33,36 .Previous studies from our laboratory in anesthetized

dogs indicated that reflex excitatory responses mediated byleft ventricular sympathetic afferents are most apparentduring ischemia which is transmural and involves the

w xepicardium 23 . This finding is consistent with otherexperimental data which indicate that left ventricular sym-pathetic afferent fibers are located mainly in the superficial

w xepicardial layers 4 . In our previous experiments, thereflex effects of sympathetic afferent activation were quan-tified by direct, recording of either efferent renal sympa-

Ž .thetic nerve activity ns20 or efferent cardiac sympa-Ž .thetic nerve activity ns5 . The reflex changes in renal

and cardiac sympathetic nerve activity during transmuralŽ .ischemia were directionally similar i.e. excitatory . How-

ever, the magnitude of reflex sympatho-excitation ap-

0165-1838r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.Ž .PII S0165-1838 97 00029-5

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–16 11

peared to be greater in the cardiac nerves compared to therenal nerves.

Since these recordings were made in different groups ofanimals, no definitive conclusions can be made aboutwhether this apparent discrepancy represents a true physio-logic finding. To our knowledge, Weaver et al. haveperformed the only experiments in which cardiac and renalnerve activity were recorded simultaneously during activa-

w xtion of cardiac sympathetic afferents 40 . They observedthat sympathetic afferent activation elicited greater reflexincreases in cardiac sympathetic nerves than in renal sym-pathetic nerves, but this difference was significant onlyafter section of the spinal cord. However, their studieswere performed in anesthetized cats and sympathetic affer-ents were stimulated by epicardial bradykinin only. Simul-taneous neural responses to myocardial ischemia were notassessed. Confirmation that sympathetic afferent activationresults in augmented reflex sympatho-excitation to theheart would have important physiological implications,particularly during transmural myocardial ischemia. In thisclinical situation, the levels of autonomic tone to the heartare known to be important determinants of myocardialelectrophysiological properties and mechanical function.Thus, to determine whether stimulation of ventricular re-ceptors with sympathetic afferent fibers actually elicitsgreater reflex increases in sympathetic outflow to the heartduring transmural myocardial ischemia, we studied a groupof animals in whom sympathetic nerve activity to the heartand kidney were measured simultaneously.

2. Materials and methods

Experiments were performed in nineteen anesthetized,mechanically-ventilated dogs. The animals were anes-

Ž .thetized with thiamylal sodium 15–25 mgrkg followedŽ .by alpha-chloralose 80 mgrkg i.v. . Additional doses of

Ž .chloralose 10 mgrkg i.v. were administered hourly. Theanimals were ventilated with a combination of oxygen androom air. During nerve recordings, pancuronium bromideŽ .2 mg i.v. was given to prevent movement. Body tempera-ture was monitored and animals were externally warmedwith a heat blanket and heat lamp. A cannula placed in thefemoral vein was used for the administration of drugs andfluids. Fluid losses were replaced by 0.9% saline intra-venously. The level of anesthesia was monitored in severalways. Femoral arterial and left atrial pressures were moni-tored continuously using conventional fluid filled cathetermanometer systems. Any changes in arterial pressure asso-ciated with painful stimuli were considered indicative ofinadequate anesthesia and supplemental doses were pro-vided. In addition, toe pinch, corneal and palpebral re-flexes were assessed every 15 min. When neuromuscularblocking agents were used, the effects of these drugs wereallowed to wear off before additional doses were given sothe level of anesthesia could be assessed. Our experimental

protocol was reviewed and approved by the InstitutionalAnimal Care and Use Committee of Virginia Common-wealth University.

2.1. Surgical preparation

In order to isolate the reflexes mediated by sympatheticafferents, cervical vagotomy and sinoaortic denervationwere performed. The carotid arteries and cervical vagiwere exposed bilaterally with a midline incision. The vagalcardiopulmonary receptors and the aortic arch barorecep-tors were denervated by sectioning both cervical vagi. Thecarotid baroreceptors were denervated by ligating and sec-tioning all structures which coursed between the internaland external carotid arteries. Abolition of the reflex in-creases in arterial pressure and sympathetic nerve activityduring transient bilateral carotid occlusion were consideredindicative of complete carotid sinus denervation in eachexperiment. Previous studies indicate that the technique ofcervical vagotomy abolishes reflex responses mediated bythe cardiopulmonary vagal afferents and the aortic arch

w xbaroreceptors 21,34 .

2.2. Coronary instrumentation

A thoracotomy was performed in the left fifth inter-costal space to expose the heart. A small opening wasmade in the pericardium. The proximal segments of theanterior descending and circumflex coronary arteries wereisolated. A snare occluder was placed around the circum-flex artery. A hydraulic occluder and a Doppler velocitytransducer were placed around the anterior descendingartery. In eight experiments, a small branch of a diagonal

Ž .artery was isolated and a small cannula PE-50 wasinserted for intracoronary injection of bradykinin. Carewas taken to keep the epicardial surface moist with warmsaline. Epicardial temperature was monitored and externalwarming was used to maintain temperature between 36and 388C.

2.3. Sympathetic nerÕe recordings

Through an incision in the left flank, the left kidney andthe left renal artery were exposed. Using a binocularoperating microscope, a branch of the left renal sympa-thetic nerves was isolated and dissected free from thesurrounding connective tissue. The nerve was cut distallyand the neural sheath was removed. The exposed nervewas then immersed in mineral oil and placed onplatinum–iridium electrodes for the recording of action

w xpotentials as previously described in detail 9 . Briefly, theŽsignal was amplified by a Grass band-pass amplifier P511;

.Grass Instruments Co., Quincy, MA with high and lowfrequency filters set at 1000–3000 and 30–100 Hz, respec-tively. The output of this amplifier was fed into an audio

Žamplifier and a spike counter University of Iowa Model

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–1612

.706c Nerve Traffic Analysis System, Iowa City, IA . Thespike counter counted and integrated all nerve spike activ-ity whose amplitude exceeded a preselected voltage levelwhich was positioned just above the background noiseband.

For cardiac nerve recordings, the left ventrolateral car-diac sympathetic nerve was identified adjacent to the leftthoracic vagus and dissected free from surrounding con-nective tissue. The nerve was cut distally and placed on adissecting stage in a mineral oil pool. After removing thenerve sheath, the nerve was placed on electrodes for actionpotential recordings as described above. This nerve wasselected for recording because it has been shown to con-tain mainly efferent sympathetic fibers rather than sensory

w xfibers 2 .

2.4. Experimental protocol

Measurements of arterial pressure, mean left atrial pres-sure, renal sympathetic nerve activity and cardiac sympa-thetic nerve activity were made at baseline and in responseto the three separate stimuli designed to activate ventricu-lar sympathetic afferent fibers. The stimuli consisted of a 2min period of inferoposterior transmural myocardial is-

Ž .chemia ns15 , application of bradykinin to the epi-Ž .cardium ns17 and intracoronary injection of bradykinin

Ž .into the anterior descending artery ns8 . Transmuralinferoposterior myocardial ischemia was created by totalocclusion of the circumflex coronary artery after placing acollateral flow limiting stenosis on the anterior descendingartery. The stenosis was adjusted to abolish coronary va-sodilator flow reserve without reducing basal levels ofcoronary flow. Coronary flow reserve was assessed byobserving the hyperemic flow velocity responses whichwere elicited by a transient complete occlusion of thestenosed artery. The stenosis was tightened until there waselimination of the reactive hyperemia. Previous experi-ments from our laboratory have demonstrated that thismethod results in myocardial ischemia which is moretransmural and involves the epicardial layers to a greater

w xdegree than does simple coronary occlusion 22,23 .ŽEpicardial bradykinin mean dose, 253"25 mg; range,

.100–500 mg was applied to 1.5=1.5 cm gauze spongeswhich were placed on the anterior surface of the left

Žventricle. Intracoronary bradykinin mean dose, 62"8.mg; range, 50–100 mg was injected into the anterior

descending artery through the coronary cannula. The orderin which these maneuvers were performed was random-

Ž .ized. Adequate time 30–45 min was allowed betweeneach manoeuver for stabilization.

2.5. Data analysis

Arterial pressure, left atrial pressure, raw and integratedcardiac and renal nerve activity were recorded on an

Želectrostatic recorder Gould ES 1000; Gould, Greenbelt,

.MD . For the coronary occlusion experiments, baselinemeasurements of arterial pressure, left atrial pressure, car-

Ž .diac and renal sympathetic nerve activity impulsesrswere made following adjustment of the coronary stenosis.During the 2 min period of inferoposterior myocardialischemia, measurements of these parameters were madefor each 30 s interval. A recovery measurement was made5 min following the release of the occlusion and stenosis.Nerve activity was expressed as percent changes frombaseline levels. This normalization was performed becausewe recorded from multiunit nerve preparations in whichthe number of active fibers placed on the recording elec-trode may have varied widely between experiments. Statis-tical comparisons between experiments requires that nerveactivity changes be normalized to basal levels. Observa-tions from all dogs were combined and mean valuesŽ .SEM were calculated. A repeated measure analysis of

Žvariance General Linear Models Procedure, SAS, Cary,.NC was used to determine if the reflex changes in cardiac

and renal sympathetic nerve activity elicited by transmuralmyocardial ischemia were significantly different. In addi-tion, the maximal changes in cardiac and renal nerveactivity for each experiment were calculated and a paired

Ž .t-test SAS, Cary, NC was used to determine whether themean maximal values were significantly different. Pairedt-tests were also used to assess whether the maximalchanges in arterial and left atrial pressures during coronaryocclusion were significantly different from baseline values.

In the bradykinin experiments, measurements of arterialpressure, left atrial pressure, cardiac and renal sympatheticnerve activity were made immediately before bradykininadministration and at the time of maximal reflex pressureand nerve activity changes. Paired t-tests were used todetermine whether cardiac and renal nerve traffic changeswere different from each other and whether arterial andleft atrial pressure changes were different from baselinevalues. In all cases, a p value F0.05 was consideredstatistically significant.

3. Results

Fig. 1 shows a representative experimental record of theresponses to intracoronary administration of 50 mg ofbradykinin into the anterior descending artery. Arterial and

Ž .left atrial pressures are shown as well as both raw ENGand integrated cardiac and renal nerve activity. Intracoro-nary bradykinin elicited decreases in arterial pressure andreflex increases in both cardiac and renal nerve activity. Inthis particular experiment, intracoronary bradykinin re-sulted in a 291% increase in cardiac nerve activity and a160% increase in renal nerve activity.

Fig. 2 shows the responses of the sympathetic nerves totransmural inferoposterior myocardial ischemia. Reflex in-creases in neural traffic were observed in both the cardiacand the renal nerves during coronary occlusion. The mag-

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–16 13

Ž .Fig. 1. Original experimental record showing changes in arterial and left atrial pressures and in raw ENG and integrated renal and cardiac nerve activityŽ .in response to injection of bradykinin 50 mg into the anterior descending coronary artery.

nitude of reflex sympatho-excitation was significantlygreater in the cardiac nerves than in the renal nervesŽ .ps0.004 . The mean maximal responses of the sympa-thetic nerves to transmural inferoposterior ischemia areshown in Fig. 3. The mean maximal percent increase was

Žsignificantly greater in cardiac sympathetic nerves 58"

. Ž .11% than in the renal sympathetic nerves 36q9% .Hemodynamically, coronary occlusion was associated with

Ža significant fall in mean arterial pressure y30"3.mmHg and a significant increase in left atrial pressure

Ž .6"1 mmHg .

Ž .Fig. 2. Percent changes in sympathetic nerve activity SNA elicited byŽ .transmural myocardial ischemia. Mean "SEM values for renal

Ž . Ž .RSNA-solid symbols and cardiac CSNA-open symbols sympatheticŽ .nerve activity are shown. Measurements made at baseline B before

coronary occlusion and at 30 s intervals during a 2 min period ofŽ .transmural ischemia. Recovery measurement R made 5 min following

release of occlusion.

The responses of the sympathetic nerves to intracoro-nary bradykinin are shown in Fig. 4. Intracoronarybradykinin elicited reflex increases in both cardiac andrenal nerve activity. The degree of reflex sympathoexcita-

Žtion was significantly greater in the cardiac nerves 144"

. Ž .48% than in the renal nerves 77"26% . Hemodynami-cally, intracoronary bradykinin resulted in a significant

Ž .decrease in mean arterial pressure y54"7 mmHg . Therewere no significant changes in left atrial pressure.

The responses of the sympathetic nerves to epicardialbradykinin are shown in Fig. 5. Epicardial bradykinin alsoelicited reflex increases in both cardiac and renal nerveactivity. The increases which were observed in the cardiac

Fig. 3. Maximal percent changes in SNA observed during two minuteperiod of transmural myocardial ischemia. Reflex changes in renalŽ . Ž .RSNA-solid bar and cardiac CSNA-open bar nerve activity are shown.Values represent means"SEM.

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–1614

Fig. 4. Maximal percent changes in SNA observed during intracoronaryŽ .bradykinin. Reflex changes in renal RSNA-solid bar and cardiac

Ž .CSNA-open bar nerve activity are shown. Values represent means"

SEM.

Fig. 5. Maximal percent changes in SNA observed during epicardialŽ .bradykinin. Reflex changes in renal RSNA-solid bar and cardiac

Ž .CSNA-open bar nerve activity are shown. Values represent means"

SEM.

Ž .nerves 167"44% were not significantly different fromthe increases which were observed in the renal nervesŽ .127"36% . Epicardial bradykinin was associated withsmall but significant increases in mean arterial pressureŽ .6"1 mmHg and with no consistent change in left atrialpressure.

4. Discussion

Our experimental results indicate that activation of leftventricular sympathetic afferents by either transmural my-ocardial ischemia or by intracoronary administration ofbradykinin elicits greater reflex sympatho-excitation to theheart than to the kidney. Although epicardial bradykininelicited increases in sympathetic nerve traffic, there was noaugmented reflex sympatho-excitation in cardiac nervescompared to renal nerves. This observation is similar to theresults reported by Weaver et al. which were described

w xabove 40 . We speculate that the failure of epicardialbradykinin to elicit augmented sympatho-excitation in car-diac nerves is related to the number of receptors which are

activated by this route of administration. Coronary occlu-sion and intracoronary bradykinin potentially stimulate allreceptors in the distribution of the coronary artery. This isin contrast to epicardial bradykinin which activates onlythose receptors in the limited area of bradykinin applica-tion. The strength of the stimulus which is provided byepicardial bradykinin may not be sufficient to augment thelevel of sympatho-excitation to the heart.

4.1. Experimental model

Our experiments were performed in animals withsinoaortic denervation and vagotomy. This experimentalmodel enabled us to isolate reflexes mediated by leftventricular sympathetic afferents and to eliminate otherreflex pathways which could potentially modulate sympa-thetic responses to myocardial ischemia and bradykinin.We have shown previously that reflex excitatory responsesto ischemia and bradykinin in animals with sinoaorticdenervation and vagotomy are eliminated following inter-

w xruption of cardiac sympathetic afferents 23 . This observa-tion confirms that these responses are mediated by leftventricular receptors with sympathetic afferent fibers.

Although this experimental model facilitates investiga-tion of reflexes mediated by sympathetic afferents in anes-thetized dogs, these excitatory reflexes have been morereadily demonstrated in other species and under differentconditions. For instance, in anesthetized cats, simple coro-nary occlusion has been shown to increase efferent sympa-thetic outflow in animals with all reflexes intact as well asin decerebrate animals, spinal animals, animals with vago-

w xtomy and animals with sinoaortic denervation 17,18,39 .The reasons for this species difference are unclear. Thefeline heart is known to have limited collateral circulation

w xcompared to the canine heart 12 . Due to this limitedcollateral circulation, we speculate that simple coronaryocclusion elicits greater transmural ischemia in cats than indogs and that this provides a stronger stimulus to thecardiac receptors with sympathetic afferents which are

w xlocated mainly in the epicardium 4 .The use of anesthesia also may significantly influence

the experimental evaluation of reflexes mediated by sym-pathetic afferents. In conscious dogs, intracoronary injec-tion of bradykinin elicited hemodynamic changes sugges-tive of an excitatory reflex mediated by sympathetic affer-

w xents 26 . This study used doses of bradykinin which weremuch smaller than the doses which were used in ouranesthetized dogs and direct neurographic evaluation ofreflex changes in sympathetic outflow were not performed.Coronary occlusion in conscious dogs also has been shownto elicit changes in the heart rate power spectra which are

w xsuggestive of an excitatory sympathetic reflex 29 . How-ever, these conscious animals had functioning sinoaorticbaroreceptors and coronary occlusion was associated withsmall but significant decreases in systolic arterial pressure.

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–16 15

Thus, it cannot be certain that these reflex changes weremediated solely by cardiac sympathetic afferents.

4.2. Etiological mechanisms

Although our experimental design does not enable us todetermine the exact mechanism for this augmented sympa-tho-excitation, one possible explanation is that cardio-cardiac spinal reflexes are engaged during activation of leftventricular sympathetic afferents. Unlike the ventricularvagal afferents which project to the brain, the ventricularsympathetic afferents project to the spinal cord where they

w xsynapse with neurons which travel to the brain 19 . As aresult, the heart is under the neural influences of sympa-thetic cardio–cardiac reflex pathways which exist at both

w xthe central and spinal levels 6,9,19,20 . Although there isevidence that cardiac spinal reflexes can be ‘multisegmen-

w xtal’ 40 , augmented increases in sympathetic outflow re-lated to engagement of these spinal reflexes would be mostapparent in the heart compared to other end-organs such asthe kidney.

The significance of the spinal reflex during myocardialischemia is uncertain. Several studies have shown thatinhibitory supraspinal mechanisms diminish the influence

w xof spinal sympathetic reflexes 8,41 . For instance, Felderand Thames demonstrated that myocardial ischemia elicitedreflex increases in cardiac sympathetic nerve activity only

w xafter the spinal cord was transected 9 . Their experimentalpreparation was somewhat similar to ours in that theystudied anesthetized dogs with the vagi sectioned and thecarotid sinuses isolated and perfused. However, myocar-dial ischemia was created in their study by simple occlu-sion of the coronary artery. Our previous experimentalresults would suggest that this technique resulted in non-transmural ischemia which provided a weak stimulus toreceptors which are located in the superficial epicardial

w xlayers 22,23 . Although spinal reflexes are clearly inhib-ited by supraspinal influences when the spinal cord isintact, our results suggest that a strong stimulus to leftventricular sympathetic afferents can overcome supraspinalinhibition of the sympathetic cardio–cardiac spinal reflex.

4.3. Pathophysiological significance of the findings

Our data adds to a growing body of information whichindicates that there can be major inhomogeneity in thecontrol of sympathetic outflow to different organ systems.In our experiment, activation of left ventricular sympa-thetic afferents by myocardial ischemia elicited reflexchanges in cardiac and renal nerve activity which were atleast directionally similar. Other studies have shown thatactivation of vagal afferents can result in reflex changes insympathetic outflow to different organs which are direc-

w xtionally opposite 38 . Furthermore, it has been demon-strated that coronary occlusion can elicit reflex changes insympathetic nerve activity which vary within the heart

depending upon whether the nerve supplies ischemic orw xnon-ischemic regions of the left ventricle 24 . This vari-

ability raises serious technical questions regarding the useof sympathetic nerve traffic measurements from one organas an estimate of sympathetic outflow to all other organs.

The finding of augmented reflex sympathoexcitation tothe heart in response to activation of left ventricular recep-tors with sympathetic afferent fibers may be particularlyimportant during pathologic states such as acute myocar-dial ischemia or infarction. Augmented reflex sympatho-excitation to the heart could promote cardiac electrical

w xinstability 10,11,13,14,30,31,37 . As a result, ventriculararrhythmias would be more likely to occur. Finally, aug-mented increases in sympathetic outflow to the heart couldincrease cardiac contractility and could vasoconstrict coro-nary arteries leading to increased myocardial oxygen de-

w xmand and increased coronary vascular resistance 1,5,27 .This would worsen myocardial ischemia and potentiallyincrease the extent of myocardial necrosis.

In summary, our experiments reveal that activation ofleft ventricular receptors with sympathetic afferent fibersby either transmural ischemia, intracoronary bradykinin, orepicardial bradykinin elicits reflex sympatho-excitation inboth efferent cardiac and renal sympathetic nerves. Reflexexcitatory responses to transmural ischemia and intracoro-nary bradykinin are significantly greater in efferent cardiacsympathetic nerves than in renal sympathetic nerves. Wespeculate that engagement of cardio–cardiac spinal re-flexes during activation of left ventricular sympatheticafferents could contribute to these augmented increases insympathetic outflow to the heart.

Acknowledgements

We wish to acknowledge the technical assistance ofTheresa L. Cersley in the performance of these experi-ments and the clerical assistance of Shirley Y. McCray inthe preparation of the manuscript. We thank Marc D.Thames, MD, Bruce R. Stambler, MD, Kenneth A. Ellen-bogen, MD and P.K. Mohanty, MD for their thoughtfulreview of the manuscript.

References

w x1 A.J. Anzola, R.F. Rushmer, Cardiac responses to sympathetic stimu-Ž .lation, Circ. Res. 4 1956 302.

w x2 J.A. Armour, W.C. Randall, Functional anatomy of canine cardiacŽ .nerves, Acta Anat. 91 1975 510–529.

w x3 D.G. Baker, H.M. Coleridge, J.C.G. Coleridge, T. Nerdrum, Searchfor a cardiac nociceptor: Stimulation by bradykinin of sympathetic

Ž .afferent nerve endings in the heart of the cat, J. Physiol. 306 1980519–536.

w x4 M.J. Barber, T.M. Mueller, B.G. Davies, D.P. Zipes, Phenol topi-cally applied to canine left ventricular epicardium interrupts sympa-

Ž .thetic but not vagal afferents, Circ. Res. 55 1984 532–544.w x5 R.J. Barnes, E.A. Bower, T.J. Rink, Haemodynamic responses to

( )R. Malik, A.J. MinisirJournal of the Autonomic NerÕous System 65 1997 10–1616

stimulation of the cardiac autonomic nerves in the anesthetized catŽ .with closed chest, J. Physiol. 299 1980 55.

w x6 A.M. Brown, A. Malliani, Spinal sympathetic reflexes initiated byŽ .coronary receptors, J. Physiol. 212 1971 685–705.

w x Ž .7 A. Brown, Cardiac reflexes, in: R.N. Beme Ed. , Handbook ofPhysiology: The Heart, American Physiology Society, Washington,DC, 1983, pp. 677–689.

w x8 K. Dembowsky, J. Czachurski, K. Amendt, H. Seller, Bulbospinalinhibitory influence on sympathetic preganglionic neurons, in: C.M.

Ž .Brooks, K. Koizumi, A. Sato Eds. , Integrative Functions of theAutonomic Nervous System, Tokyo University Press, Tokyo, 1979,pp. 376–384.

w x9 R. Felder, M.D. Thames, Cardio–cardiac sympathetic reflex duringŽ .coronary occlusion in anesthetized dogs, Circ. Res. 48 1981 685–

692.w x10 R.A. Freedman, C.D. Swerdlow, D.S. Echt, R.A. Winkle, V. Soder-

holm-Difatte, J.W. Mason, Facilitation of ventricular tachyarrhyth-Ž .mia induction by isoproterenol, Am. J. Cardiol. 54 1984 765–770.

w x11 R.A. Gillis, Role of the nervous system in the arrhythmias producedŽ .by coronary occlusion in the cat, Am. Heart J. 81 1971 677–681.

w x12 D.E. Gregg, L.C. Fisher, Blood supply to the heart, in: Handbook ofPhysiology, vol. II, section 2, American Physiology Society, Wash-ington, 1963, pp. 1517–1584.

w x13 J.M. Herre, L. Wetstein, Y.L. Lin, A.S. Mills, M. Dae, M.D.Thames, Effect of transmural versus nontransmural myocardial in-farction on inducibility of ventricular arrhythmias during sympa-

Ž .thetic stimulation in dogs, J. Am. Coll. Cardiol. 2 1988 414–421.w x14 B.R. Kliks, M.J. Burgess, J.A. Abildskov, Influence of sympathetic

tone on ventricular fibrillation threshold during experimental coro-Ž .nary occlusion, Am. J. Cardiol. 36 1975 45–49.

w x15 F. Lombardi, P. Della Bella, R. Casati, A. Malliani, Effects ofintracoronary administration of bradykinin on the impulse activity ofafferent sympathetic unmyelinated fibers with left ventricular end-

Ž .ings in the cat, Circ. Res. 48 1981 69–75.w x16 F. Lombardi, C.P. Patton, P. Della Bella, M. Pagini, A. Malliani,

Cardiovascular and sympathetic responses reflexly elicited throughthe excitation with bradykinin of sympathetic and vagal cardiac

Ž .sensory endings in the cat, Cardiovasc. Res. 16 1982 57–65.w x17 F. Lombardi, C. Casalone, P. Della Bella, G. Malfatto, M. Pagani,

A. Malliani, Global versus regional myocardial ischaemia: Differ-ences in cardiovascular and sympathetic responses in cats, Cardio-

Ž .vasc. Res. 18 1984 14–23.w x18 A. Malliani, P. Schwartz, A. Zanchetti, A sympathetic reflex elicited

Ž . Ž .by experimental coronary occlusion, Am. J. Physiol. 217 3 1969703–709.

w x19 A. Malliani, D.F. Peterson, V.S. Bishop, A.M. Brown, Spinal sym-Ž .pathetic cardio–cardiac reflexes, Circ. Res. 30 1972 158–166.

w x20 A. Malliani, Cardiovascular sympathetic afferent fibers, Rev. Phys-Ž .iol. Biochem. Pharmacol. 94 1982 11–74.

w x21 A.J. Minisi, M. Dibner-Dunlap, M.D. Thames, Vagal cardiopul-monary baroreflex activation during phenylephrine infusion, Am. J.

Ž .Physiol. 257 1989 R1147–R1153.w x22 A.J. Minisi, D. Brands, M.D. Thames, An experimental model for

Ž .the creation of transmural ischemia in dogs, FASEB J. 4 1990A420.

w x23 A.J. Minisi, M.D. Thames, Activation of cardiac sympathetic affer-ents during coronary occlusion: Evidence for reflex activation of thesympathetic nervous system during transmural myocardial ischemia

Ž .in the dog, Circulation 84 1991 357–367.w x24 B.H. Neeley, G.R. Hageman, Differential cardiac sympathetic activ-

Žity during acute myocardial ischemia, Am. J. Physiol. 258 Heart. Ž .Circ. Physiol. 27 1990 H1534–H1541.

w x25 K. Nishi, M. Sakanashi, F. Takenaka, Activation of afferent cardiacsympathetic nerve fibers of the cat by pain producing substances and

Ž .noxious heat, Pfluegers Arch. 372 1977 53–61.w x26 M. Pagani, P. Pizzinelli, R. Furlan, S. Guzzetti, O. Rimoldi, G.

Sandrone, A. Malliani, Analysis of the pressor sympathetic reflexproduced by intracoronary injections of bradykinin in conscious

Ž .dogs, Circ. Res. 56 1985 175–183.w x27 W.C. Randall, Sympathetic control of the heart, in: W.C. Randall

Ž .Ed. , Neural Regulation of the Heart, Oxford University Press, NewYork, 1977.

w x28 K.A. Reimann, L.C. Weaver, Contrasting reflexes evoked by chemi-Ž .cal activation of cardiac afferent nerves, Am. J. Physiol. 239 1980

H316–H325.w x29 O. Rimoldi, S. Pierini, A. Ferari, S. Cerutti, M. Pagani, A. Malliani,

Analysis of short-term oscillations of R–R and arterial pressure inŽ . Ž .conscious dogs, Am. J. Physiol. 258 Heart Circ. Physiol. 27 1990

H967–976.w x30 J.J. Salata, D.P. Zipes, Autonomic nervous system control of heart

rate and atrioventricular nodal conduction, in: I.H. Zucker, J.P.Ž .Gilmore Eds. , Reflex Control of the Circulation, CRC Press, Boca

Raton, Florida, 1991, pp. 67–101.w x31 P.J. Schwartz, H.L. Stone, A.M. Brown, Effects of unilateral stellate

ganglion blockade on the arrhythmias associated with coronaryŽ .occlusion, Am. Heart. J. 92 1976 589–599.

w x32 J. Staszewska-Barczak, S.H. Fereira, J.R. Vane, An excitatory noci-ceptive cardiac reflex elicited by bradykinin and potentiated by

Ž .prostaglandins and myocardial ischaemia, Cardiovasc. Res. 10 19673114–3327.

w x33 J. Staszewska-Barczak, G.J. Dusting, Sympathetic cardiovascularreflex initiated by bradykinin-induced stimulation of cardiac pain

Ž .receptors in the dog, Clin. Exp. Pharmacol. Physiol. 4 1977443–452.

w x34 M.D. Thames, B.D. Miller, F.M. Abboud, Baroreflex regulation ofrenal nerve activity during volume expansion, Am. J. Physiol. 243Ž .1982 H810–H814.

w x35 P. Thoren, Role of cardiac vagal c-fibers in cardiovascular control,Ž .Rev. Physiol. Biochem. Pharmacol. 86 1979 1–94.

w x36 Y. Uchida, S. Murao, Bradykinin-induced excitation of afferentŽ .cardiac sympathetic nerve fibers, Jpn. Heart J. 15 1974 84–91.

w x37 R.L. Verrier, B. Lown, Sympathetic–parasympathetic interactionand ventricular electrical stability, in: P.J. Schwartz, A.M. Brown,

Ž .A. Malliani, A. Zanchetti Eds. , Neural Mechanisms in CardiacArrhythmias, Raven Press, New York, 1978.

w x38 R.G. Victor, P. Thoren, D.A. Morgan, A.L. Mark, Differentialcontrol of adrenal and renal sympathetic nerve activity during

Ž .hemorrhagic hypotension in rats, Circ. Res. 64 1989 686–694.w x39 L.C. Weaver, L.M. Danos, R.S. Oehl, R.L. Meckler, Contrasting

reflex influences of cardiac afferent nerves during coronary occlu-Ž . Ž .sion, Am. J. Physiol. 240 Heart Circ. Physiol. 9 1981 L:H620–

H629.w x40 L.C. Weaver, H.K. Fry, R.L. Meckler, R.S. Oehl, Multisegmental

spinal sympathetic reflexes originating from the heart, Am. J. Phys-Ž . Ž .iol. 245 Regulatory Integrative Comp. Physiol. 14 1983 R345–

R352.w x41 R.D. Wurster, Spinal sympathetic control of the heart, in: W.C.

Ž .Randall Ed. , Neural Regulation of the Heart, Oxford UniversityPress, New York, 1977, pp. 211–246.