Search

Cardiac nerve sprouting and sympathetic hyperinnervation after myocardial infarction (MI) both contribute to arrhythmogenesis and sudden death. However, the mechanisms responsible for nerve sprouting after MI are unclear. The expression of nerve growth factor (NGF), growth associated protein 43 (GAP43), and other nerve markers were studied at the infarcted site, the noninfarcted left ventricle free wall (LVFW), and the left stellate ganglion (LSG) at several time points (30 minutes to 1 month) after MI. Transcardiac (difference between coronary sinus and aorta) NGF levels were also assayed. Acute MI resulted in the immediate elevation of the transcardiac NGF concentration within 3.5 hours after MI, followed by the upregulation of cardiac NGF and GAP43 expression, which was earlier and more pronounced at the infarcted site than the noninfarcted LVFW. However, cardiac nerve sprouting and sympathetic hyperinnervation were more pronounced in the noninfarcted than the infarcted LVFW site and peaked at 1 week after MI. The NGF and GAP43 protein levels significantly increased in the LSG from 3 days ( P

Abstract —The possible involvement of the local angiotensin system in ganglionic functions was investigated in the canine cardiac sympathetic ganglia. Positive chronotropic responses to preganglionic stellate stimulation at high frequencies, after intravenous administration of pentolinium plus atropine, were inhibited by the nonpeptide angiotensin AT 1 receptor antagonist forasartan or the angiotensin I–converting enzyme inhibitor captopril, whereas the rate increases elicited by the postganglionic stellate stimulation and norepinephrine given intravenously failed to be inhibited by these antagonists. The levels of endogenous immunoreactive angiotensin II, as determined by radioimmunoassay in the incubation medium of the stellate and inferior cervical ganglia, were increased after the high-frequency preganglionic stimulation of the isolated ganglia. The increment of the peptide was also antagonized by the pretreatment with captopril but not by a chymase inhibitor, chymostatin. The expression of angiotensinogen mRNA was observed in the stellate ganglion, adrenal, liver, and lung but not in the ovary and spleen. The expression of the mRNA in the stellate and inferior cervical ganglia increased after high-frequency preganglionic stimulation of the in vivo dogs for a period of 1 hour. These results indicate that an intrinsic angiotensin I–converting enzyme–dependent angiotensin system exists in the cardiac sympathetic ganglia, which is activated by high-frequency preganglionic stimulation.

Abstract —The factors that contribute to the occurrence of sudden cardiac death (SCD) in patients with chronic myocardial infarction (MI) are not entirely clear. The present study tests the hypothesis that augmented sympathetic nerve regeneration (nerve sprouting) increases the probability of ventricular tachycardia (VT), ventricular fibrillation (VF), and SCD in chronic MI. In dogs with MI and complete atrioventricular (AV) block, we induced cardiac sympathetic nerve sprouting by infusing nerve growth factor (NGF) to the left stellate ganglion (experimental group, n=9). Another 6 dogs with MI and complete AV block but without NGF infusion served as controls (n=6). Immunocytochemical staining revealed a greater magnitude of sympathetic nerve sprouting in the experimental group than in the control group. After MI, all dogs showed spontaneous VT that persisted for 5.8±2.0 days (phase 1 VT). Spontaneous VT reappeared 13.1±6.0 days after surgery (phase 2 VT). The frequency of phase 2 VT was 10-fold higher in the experimental group (2.0±2.0/d) than in the control group (0.2±0.2/d, P

Recent work has shown that alterations in the dynamic atrioventricular (AV) nodal response to changes in heart rate can significantly modify AV nodal function. The present study was designed to evaluate the nature and potential importance of sympathetic regulation of the rate-dependent properties of the AV node. Selective stimulation protocols and mathematical formulations were used to independently quantify AV nodal recovery, facilitation, and fatigue in 12 morphine-chloralose-anesthetized dogs. Vagal effects were prevented by bilateral vagal transection and intravenous atropine, and the sinus node was crushed to allow a broader range of pacing cycle lengths. In seven dogs with sympathetic nerves intact, beta-adrenergic receptor blockade increased the recovery time constant (tau rec) for the conduction of premature test beats from 47 +/- 2 (mean +/- SEM) msec (control) to 62 +/- 1 msec (p less than 0.001), whereas isoproterenol decreased tau rec to 38 +/- 1 msec (p less than 0.001). In addition, beta-blockade increased the maximum amount of rate-dependent AV nodal fatigue from 7 +/- 1 msec (at a cycle length of 198 +/- 9 msec [control]) to 17 +/- 2 msec (p less than 0.001). In five dogs with decentralized stellate ganglia, tau rec was decreased from 71 +/- 3 msec (control) to 57 +/- 4 msec and 48 +/- 2 msec (p less than 0.001 for each) by left stellate ganglion stimulation at 5 and 10 Hz, respectively. Maximum fatigue was similarly reduced from 16 +/- 1 msec (control) to 12 +/- 2 msec (p = NS) and 8 +/- 1 msec (p less than 0.01), respectively. Stellate ganglion stimulation, isoproterenol, and beta-blockade did not alter AV nodal facilitation. A mathematical model incorporating quantitative indexes of AV nodal function accurately accounted for tachycardia-dependent increases in the atrial-His activation interval, which were enhanced by beta-adrenergic receptor blockade and reduced by isoproterenol. Furthermore, this model showed that beta-adrenergic effects were increased by increasing heart rate, with the majority of the rate-dependent action being due to changes in the time course of AV nodal recovery. We conclude that beta-adrenergic receptor stimulation alters functional properties that govern the AV nodal response to changes in heart rate. These changes in functional properties alter the ability of the AV node to conduct impulses during tachycardia and, as such, could play a major role in the ability of sympathetic stimulation to promote and beta-adrenergic receptor blockade to prevent the occurrence of AV nodal reentrant arrhythmias.

See related article, pages 76–83 Although the concept of remodeling is well established with respect to heart muscle, the importance of a restructuring of cardiac innervation, or “rewiring,” after myocardial infarction (MI) has only recently received due attention. Pioneering contributions in this regard have emerged from the laboratories of Zipes1 and Chen.2–6 Recently, Cao, Chen, and coworkers2–5 provided evidence implicating nerve sprouting in ventricular arrhythmogenesis and potentially sudden cardiac death (SCD). These investigators reported a significant correlation between increased sympathetic nerve density as reflected in immunocytochemical markers with history of arrhythmias including ventricular tachycardia and SCD in native hearts of human transplant recipients with severe heart failure.3 Their observations suggested an association between postinjury sympathetic nerve density and susceptibility to life-threatening ventricular arrhythmias in these patients. In a canine post-MI model, they demonstrated that induction of nerve sprouting by infusion of nerve growth factor (NGF) into the left stellate ganglion (LSG) resulted in increased incidence of ventricular tachycardia and fibrillation.4 Significantly, the predisposition to arrhythmias was again linked to immunocytochemical evidence of a heterogeneous pattern of sympathetic reinnervation. In a similar conscious canine model, the …

In two groups of dogs we studied the effect of right and/or left stellectomy on myocardial reactive hyperemia (RH) and on coronary blood flow distribution. In the first group of 14 conscious dogs, the percent repayment of flow debt produced by a 10-second occlusion of the left circumflex coronary artery was recorded with a Doppler ultrasonic flow probe and a hydraulic vascular occluder. The dogs were studied under control conditions, after right stellectomy and after left stellectomy and after administration of propranolol and phentolamine. Right stellectomy did not affect RH. RH was significantly increased by left stellectomy from 476 ± 71% to 622 ± 86% (+31%) at the spontaneous heart rate and from 407 ± 51% to 577 ± 106% (+42%) during pacing. Propranolol significantly reduced RH from 447 ± 25% to 390 ± 27% (-13%) at the spontaneous heart rate and from 456 ± 25% to 311 ± 24% (-32%) during pacing. Phentolamine significantly increased RH from 419 ± 63% to 517 ± 71% (+23%). Propranolol was effective after bilateral stellectomy, whereas phentolamine was not effective after left stellectomy. In the second group of 14 anesthetized dogs with constant heart rate (15 (im) microspheres were injected twice into the left atrium. The first injection provided a control measurement; in nine dogs the second injection was made after left stellectomy. Left stellectomy significntly increased the left ventricular endocardial to epicardial ratio from 1.7 ± 0.03 to 1.23 ± 0.04. We conclude that the sympathetic nervous system has a tonic influence on coronary circulation and that left stellectomy increases the ability of the coronary bed to dilate and improves the endocardial perfusion.

Experiments were performed to determine the influence of cardiac sympathetic efferent nerve stimulation on the discharge rate of atrial type B stretch receptors in the anesthetized, open-chest dog. In all experiments, the left stellate ganglion was stimulated following volume expansion. To determine the effects of stellate stimulation, the responses of atrial receptors were observed during the withdrawal of blood in steps following intravascular volume expansion. Stimulation of the stellate ganglion decreased receptor discharge and left atrial pressure in all experiments. A change from a control left atrial pressure of 2.5 cm H 2 O (in peak left atrial v -wave pressure) resulted in a mean increase in receptor discharge of 6.6 ± 1.4 spikes/cardiac cycle during stellate stimulation; however, a change from a control pressure of 0.0 cm H 2 O resulted in a mean decrease in receptor discharge of 1.5 ± 1.5 spikes/cardiac cycle. The curves relating the change in atrial receptor discharge to the change in left atrial pressure during hemorrhage and the curve for these parameters during stellate stimulation were not significantly different from each other. Apparently, the decreased discharge during sympathetic stimulation was the result of a decline in left atrial pressure rather than a result of any direct effect on the receptor per se.

Acute volume alterations were produced in eight anesthetized dogs to determine if the contractility of the left ventricle is partially volume-dependent. Pressures were measured in the left ventricle with a micromanometer and regional ventricular function was measured with sonomicrometers implanted in the midwall of the anterior, lateral, and posterior left ventricle. Regional end-systolic pressure-length relations (ESPLR) were determined with transient venae cavae occlusions. The ESPLR data were fitted to a quadratic equation, then end-systolic lengths were compared at matched end-systolic pressures at multiple time periods after the acute load alterations. Bilateral vagotomy, carotid sinus denervation, and stellate ganglion denervation were performed to prevent reflex alterations in contractility. The heart was paced at a constant rate. In seven animals, dextran was infused intravenously over 77 +/- 23 seconds (+/- SD) to increase left ventricular end-diastolic pressure from 5 +/- 2 to 13 +/- 3 mm Hg and peak pressures from 113 +/- 11 to 147 +/- 18 mm Hg. The end-diastolic lengths increased 14 +/- 6% in the anterior, 9 +/- 2% in the lateral, and 12 +/- 4% in the posterior segments. The ESPLR was measured immediately (77 +/- 23 seconds), early (3 +/- 1 minutes), and late (10 +/- 2 minutes) after initiation of the volume load. In all three regions, there was a significant time-dependent leftward shift in the ESPLR. From 1 to 10 minutes after the volume load, the regional end-systolic lengths decreased by a mean of 4-6% when compared at a matched end-systolic pressure of 96 +/- 10 mm Hg, and decreased by 7-9% when compared at a matched pressure of 139 +/- 20 mm Hg. The end-systolic lengths immediately after the volume load were not significantly different from the control value (compared at a matched end-systolic pressure) but were significantly shorter 10 minutes after the volume load. The leftward shift of the ESPLR represented a true increase in contractility and not merely a recovery from a transient myocardial depression. A similar leftward shift in the ESPLR occurred in four animals after release of a partial venae cavae occlusion, producing an acute volume load without acute hemodilution. Acute volume loads in four animals treated with propranolol also produced a leftward shift in the ESPLR, ruling out the possibility that a time-dependent increase in circulating catecholamines was responsible for the alterations in contractility. In five animals, the ESPLR shifted to the right after a transient (1-2 minutes) decrease in venous return.(ABSTRACT TRUNCATED AT 400 WORDS)

We studied the effects of cardiac sympathetic stimulation and ablation on characteristics of the anodal-strength interval curve in 10 pentobarbital-anesthetized open-chest dogs. We gave special attention to the form and particularly the depth of early dips of the anodal strength-interval curve. When the central connections of the cardiac sympathetic nerves were intact, left stellate stimulation decreased threshold for excitation during the dip only slightly (from 883 +/- 442 muA to 774 +/- 326 muA P less than 0.1). However, after left stellectomy and upper thoracic ganglionectomy, stimulation of the cut ends the ansa subclavia decreased the dip threshold from an average of 952 +/- 321 muA to 707 +/- 210 muA, A less than 0.01. Dip thresholds in strength-interval curves determined 1-1.5 hours after ganglionectomy average 1164 +/- 296 muA compared to an average dip threshold of 791 +/- 269 muA in strength-interval curves measured immediately prior to stellectomy. This increase is statistically significant (P less than 0.005). In addition to the effects of sympathetic tone on disparity of refractory period and ventricular fibrillation threshold, the effects of sympathetic tone on the form of the strength-interval curve may be another factor influencing vulnerability to arrhythmias and particularly in patients with abnormal autonomic tone.

Studies of the stellate ganglion and middle cervical ganglion indicate that sympathetic efferent nerve activity can be modified by peripheral excitatory inputs and that these neural connections may function as pathways for a peripheral reflex at the level of the thoracic sympathetic ganglia. This excitatory synaptic input could have a soma in either the central or the peripheral nervous system. A study was designed to determine whether chronic decentralization (3 weeks) of the stellate ganglion in cats would 1) abolish sympathetic cardiac afferent nerve activity recorded at the stellate cardiac nerve and 2) abolish local thoracic reflexes that are generated by stimulation of peripheral nerves. The ansae subclaviae, T3 and T4 rami, and stellate ganglion were also examined by electron microscopy for the extent of Wallerian degeneration. Afferent cardiac activation of the axon collaterals arising from cell bodies located in the dorsal root ganglia was abolished due to degeneration. However, sympathetic afferent nerve activity from the left ventricular receptors was still present and was recorded from the stellate cardiac nerve in all cats. Cardiac receptors were sensitive to mechanical distortion, increases in the left ventricular pressure, and epicardial application of veratrine hydrochloride. These data imply that 1) cardiovascular afferent input to the stellate ganglion persists following chronic decentralization and 2) the sensory neurons are located in the peripheral sympathetic nervous system. Thus, we find that regulation of the heart occurs in part via thoracic ganglia, independently of the central nervous system.

We studied the effects of 5- and 6-hydroxydopamine on adrenergic neurotransmission, fluorescence histochemistry, and nerve terminal ultrastructure in the canine pulmonary vascular bed. Fluorescence histochemistry on stretched preparations and sections of intrapulmonary artery and vein demonstrated that these vessels are well supplied with adrenergic nerves electron microscopy revealed adrenergic terminals in the adventitia and outer third of the media in the artery, but only in the adventitia in the vein. Adrenergic terminals in artery and vein contained many small and a few large dense-core vesicles. At least 20% of the terminals in the artery contained many small agranular vesicles and a few large opaque vesicles; this suggests that they were of the cholinergic type; Such terminals were not found in intrapulmonary veins. Under conditions of controlled blood flow, stimulation of the sympathetic nerves to the lung and intralobar injection of norepinephrine increased pressure in the perfused lobar artery and small intrapulmonary vein in a stimulus-related manner. The rise in pressure in the lobar artery and vein in response to nerve stimulation was blocked after administration of either 5- and 6-hydroxydopamine; Neither agent modified the response of the pulmonary vascular bed to norepinephrine; In contrast, the rise in pressure in the lobar artery and vein in response to both norepinephrine and to nerve stimulation was blocked by phenoxybenzamine, an alpha-receptor blocking agent. The attenuated neurogenic vasoconstrictor response in dogs treated with 5- and 6-hydroxydopamine was associated with a marked decrease in intensity of fluorescence of the abundant adrenergic innervation in both intrapulmonary artery and vein, and with the appearance of an osmiophilic material in dense-core vesicles of adrenergic terminals in artery and vein. We believe that these data suggest that 5- and 6-hydroxydopamine interfere with adrenergic transmission in intrapulmonary vessels by depleting norepinephrine from adrenergic terminals. Furthermore, we conclude from hemodynamic, histochemical, and ultrastructural studies that vasomotor tone in the pulmonary vascular bed can be regulated by the sympathetic nervous system.

Recent evidence has suggested that cardiac factors may play a role in the evolution of arterial hypertension. To test the possibility that an increase in cardiac-performance can lead to a sustained increase in systemic blood pressure, we electrically stimulated the left stellate ganglion of six conscious dogs continuously for a 7-day period and monitored cardiac output and arterial blood pressure. In all six dogs, stimulation elicited an abrupt rise in systemic blood pre-sure that was entirely due to rise in cardiac output that lasted at least 6 hours. After 1 day of continuous stimulation, cardiac output retured to control values, but blood pressure remained elevated. After 7 days of stimulation, blood pressure was increased by an average of 25 mm Hg and peripheral resistance by 35 plus and minus 4%. Measurements of blood volume, plasma renin activity, circulating catecholamines (three of the six dogs), and sodium balance showed that none of these factors could explain the development pf this sustained hypertension. Pharmacologic blockade with phenoxybenzamine prevented in large part the rise in blood pressure in short-term stellate ganglion stimulations, whereas propranolol had very little effect on the pressor response, although it nearly abolished the increase in cardiac output. The data indicate that continuous stimulation of the stellate ganglion in conscious dogs leads to substained rises in both blood pressure and peripheral resistance; these changes are apparently mediated by increased activity of the sympathetic nervous system.

The activity of type A right atrial vagal receptors was recorded from the right cervical vagus in cats anesthetized with sodium pentobarbital, immobilized with gallamine, and with their chests open. Nerve impulses initiated by receptor activation were recorded simultaneously with instantaneous right atrial pressure and dimensional changes under various hemodynamic conditions. Atrial volume changes induced by infusion of saline, bleeding, and occlusion of the inferior vena cava did not alter consistently the systolic activity of the receptors. Electrical stimulation of the right stellate ganglion significantly increased the frequency of discharge during systole, whereas electrical stimulation of the left thoracic vagus significantly reduced the frequency of discharge. These inotropic interventions produced similar effects when the heart was paced at a fixed rate. Pacing the right atrial appendage increased the systolic discharge of the receptors only when at high rates the atrium contracted against closed atrioventricular valves. To investigate the influence of tonic efferent sympathetic activity on spontaneous receptor discharge, three receptors were studied before and after bilateral surgical stellectomy, and in cats with their chest closed three receptors were studied before and after infusion of propranolol. Both of these interventions markedly reduced the systolic activity. In addition to having effects on systolic activity, injection of saline, vagal stimulation, and sympathetic "denervation" always activated the receptors during filling. Our results indicate that: (1) the systolic discharge of type A receptors is a function of the active tension developed by atrial muscle during contraction; and (2) the pattern of discharge of the receptors during the atrial cycle depends on both the degree of atrial distention and the state and extent of contraction.

Acetylstrophanthidin perfused directly into the sinus node artery of the dog decreases the response of the sinus node to adrenergic stimulation. When sinus arrest followed per-fusion of high concentrations of glycosides, either sympathetic stimulation or intranodal epinephrine transiently restored sinus rhythm.

The effects of electrical stimulation of the carotid sinus on sinus rate and atrioventricular (A-V) conduction were studied before and after alternate interruption of the vagi and sympathetic nerves to the heart in dogs. In group I, carotid sinus stimulation caused a similar absolute decrease in sinus rate before and after vagotomy, although after vagotomy the response was more latent. Subsequent administration of propranolol blocked the effects of carotid sinus stimulation on sinus rate. In group II (dogs with intact vagi), the effects of carotid sinus stimulation and bilateral occlusion of the common carotid artery on sinus rate were abolished by bilateral stellatectomy and upper thoracic ganglionectomy. In group III, the effects of carotid sinus stimulation on A-V conduction during atrial pacing at a fixed rate were not significantly altered by vagotomy and were blocked by subsequent administration of propranolol. In group IV, carotid sinus stimulation had no effect on A-V conduction after sympathetic denervation. It is concluded that the effects of the carotid sinus reflex on the heart are mediated primarily by the sympathetic nervous system efferents. In addition, evidence suggesting that sympathetic tone is a major determinant of vagal tone has been presented.

Myocardial contractile force was recorded at three to five positions on the canine heart while progressively more distal portions of the cardiac sympathetic nerves were electrically stimulated. When the electrodes were applied directly to the stellate ganglion or to the ansa subclavia, positive inotropic responses were generally observed on all of the test regions of the heart. However, when more distal branches of the cardiac nerves were stimulated, individual test regions were successively deleted from the general response. By carefully removing narrow strips of epicardium (epicardial denervation) from different portions of the heart, similar deletions in response were observed. These observations indicate that the peripheral autonomic innervation of the heart is distributed to circumscribed regions of cardiac musculature. Contractile segments of a given region of the heart may be organized in series and each component of the chain influences the contraction of neighboring elements, normally innervated segments mechanically interacting with noninnervated regions. Individual units of cardiac muscle may be markedly influenced through local intervention by the sympathetic nerve supply.

The effect of sympathetic stimulation on cerebral blood flow was investigated in dogs anesthetized with chloralose. A preparation has been developed for the moment-to-moment measurement of cerebral venous outflow with an electromagnetic flow transducer. The brain's arterial supply was left undisturbed. The sympathetic innervation of the cerebral vessels was stimulated at the stellate ganglion (3-9 v, 3 msec, and 1, 3, 6, 10, and 15 Hz for 60 or 90 seconds). Stimulation at 15 Hz resulted in an average decrease in cerebral blood flow of 79.7%. During stimulation the arterial oxygen tension decreased from 93.2 to 84.9 mm Hg, the arterial carbon dioxide tension increased from 32.9 to 34.6 mm Hg, and arterial pH fell from 7.392 to 7.378. These changes in blood gas variables all opposed the observed vasoconstriction. Interactions between intracranial pressure and sympathetic cerebral vasoconstriction were evaluated by measuring cerebrospinal fluid pressure and cerebral venous outflow pressure. Stimulation of the left sympathetic stellate ganglion produced a 64% decrease in cerebral blood flow and an 8 mm Hg increase in intracranial pressure. Infusion of saline into the cisterna magna, raising intracranial pressure to 47 mm Hg, produced a 3% decrease in cerebral blood flow. Opening the cerebrospinal fluid space and thus fixing intracranial pressure at zero (atmospheric pressure) did not alter the cerebral blood flow response to sympathetic stimulation. It was concluded that stimulation of the sympathetic stellate ganglion resulted in cerebral vasoconstriction which was independent of changes in arterial P CO 2 , Po 2 , and pH and was also independent of changes in cerebrospinal fluid pressure.

The effects of stellate ganglion stimulation on idioventricular automaticity were studied in anesthetized dogs. An idioventricular rhythm was obtained in one series of animals by ligating the His bundle and in the other by stimulating the vagus. On isolation of the left stellate ganglion, the idioventricular rate fell by 5 beats/min which is compatible with a rate of "tonic" discharge of about 3 impulses/sec. On stimulation of the stellate ganglia at increasingly higher frequencies, the idioventricular rate accelerated progressively and the frequency-response curve was sigmoid. When the stimulation frequency was increased above 10 to 15 pulses/sec, there was no further enhancement of ventricular automaticity. The ventricular rates attained during sympathetic stimulation rarely exceeded 70/min. The interval between the initiation of sympathetic stimulation and the maximal ventricular acceleration was shorter at higher stimulation frequencies. The effects of right and left stellate stimulation on the rate of idioventricular pacemakers were comparable. Reflex vagal inhibition had little influence on the acceleratory action of the sympathetic nerves on the idioventricular pacemakers since in dogs with atrioventricular block the results before and after vagotomy were similar. Furthermore, the idioventricular rhythm during vagally induced block increased on sympathetic stimulation to values comparable to those obtained in dogs with surgically induced block.

The functional peripheral cardiac sympathetic pathways of the dog were delineated in isovolumetric left ventricle preparations. In unpaced hearts, supramaximal stimuli at 2 cycles/sec to the right stellate ganglion caused the heart rate to increase 2.8 times more than stimuli applied to the left side. In paced hearts, such stimuli evoked rises in left ventricular systolic pressure which were 2.3 times greater when applied to the left than to the right stellate ganglion. Tonic cardiac sympathetic impulses appear to funnel through the stellate ganglia. Decentralization of the stellate ganglia abolished the ventricular responses to stimulation of the carotid baroreceptors and cephalic ischemia. Approximately two thirds of the tonic and baroreceptor reflex influences to the left ventricle entered the stellate ganglia from lower segments of the thoracic paravertebral chain, and one third entered from the communicating rami of the stellate ganglia. The ratio of the effects of complete decentralization of the left to those of decentralization of the right stellate ganglion was 1.6, both with respect to eliminating sympathetic tone and abolishing the baroreceptor reflex.

Cardiac and aortic pressures were recorded after stellate ganglionectomy and vagotomy. Acute heart block was produced by injecting the atrioventricular node, and atrial and ventricular systoles were controlled electronically to occur independently or in any desired relationship. Angiocardiograms recorded on video tape after injections of 4 ml 69% Renovist into the left ventricle were analyzed with a videodensitometer able to detect small refluxes of contrast medium into the left atrium and correlate them with phases of the cardiac cycle. When ventricular driving was temporarily suspended but atrial driving continued, pressure records indicated mitral valve closure after each atrial systole, but reflux of contrast medium into the atrium occurred after each systole not followed by a normally sequenced ventricular systole. Driving with a 2: 1 atrioventricular stimulation resulted in reflux, with the alternate atrial contraction dissociated from ventricular systole. Thus, the mitral valve was not effectively closed by atrial systoles that were not followed by normally sequenced ventricular systoles.

The effects of small doses of sodium pentobarbital (5 mg/kg) on the cardiovascular responses to stimulation of the medulla, hypothalamus, stellate ganglia and sciatic nerve, and to bilateral occlusion of the carotid arteries were studied. It was concluded that: 1) this drug depressed cardiovascular responses to all of these procedures except stimulation of the stellate ganglia; 2) the effects of the anesthetic in this dose are referable to some part(s) of the central nervous system; 3) cardiovascular responses to hypothalamic stimulation were depressed more than those elicited by stimulation of the dorsal medullary reticular formation.

At constant heart rates, efferent stimulation of the vagus nerve and of the left stellate ganglion revealed the following: 1. Vagal stimulation exerts a profound depressant effect on the strength of the atrial contraction and can thereby influence ventricular filling and ventricular stroke work; it elevates mean atrial pressure at any given level of ventricular stroke work. This occurs under experimental conditions wherein the vagal stimulation used does not produce an alteration in the performance characteristics of the ventricle. The effects of vagal stimulation are blocked by atropine. 2. Stellate ganglion stimulation or norepinephrine infusion augments the strength of atrial contraction and thus the atrial contribution to ventricular filling. The augmented atrial contraction takes place in a shorter period of time. 3. Stellate ganglion stimulation or norepinephrine infusion increases the external stroke work and power produced by the ventricle from any given mean atrial pressure and from any given end-diastolic pressure or fiber length.4. There is a family of curves representing the relation between end-diastolic fiber length and stroke work, as well as a family of curves representing the relation between mean atrial or end-diastolic pressure and stroke work.5. When taken together with the well-known sympathetic and parasympathetic effects on heart rate, the above data are believed to comprise a reasonably comprehensive description of the means available to the central nervous system for directly inducing acute changes in the activity of the heart.

In dogs with denervated hearts intravenous infusions of epinephrine (1.0 to 3.0 µg/kg/min) caused marked acceleration of the heart rate and a significant reduction of intraventricular conduction time. Equal doses infused into the descending thoracic aorta increased the heart rate only slightly, but exerted undiminished effects upon conduction time. Evisceration did not modify the intraventricular dromotropic effect of epinephrine so long as the hepatic arterial supply was preserved. Occlusion of the hepatic artery abolished the dromotropic effect of epinephrine without modifying its chronotropic action. The time course of the dromotropic effect and of the kalemic action of epinephrine was approximately the same. Intravenous infusions of KCl matching the hyperkalemic effect of epinephrine caused similar dromotropic actions. Isoproterenol even in high doses (up to 2.0 µg/kg/min) did not decrease intraventricular conduction time and had only a slight hyperkalemic effect. After the administration of SY28 (1.0 mg/kg) epinephrine caused neither hyperkalemia nor decreased intraventricular conduction time. It is concluded that in the heart in situ the effect of epinephrine on intraventricular conduction is the result of an indirect action mediated through the liberation of K + ions from the liver.

Changes of the electrocardiogram and of ventricular refractory period were measured following either unilateral stellate ganglion stimulation or ablation in the open chest dog preparation. Right stellate ganglionectomy or left stellate stimulation produces prolonged Q-T intervals and increased T-wave amplitude. Left stellate ganglionectomy or right stellate stimulation produces increased T-wave negativity without measurable change in the Q-T interval. The differing patterns of electrocardiographic wave form resulting from changes in sympathetic tone mediated by right and left stellate innervation could be correlated with changes in ventricular refractory period. Following right stellate ganglionectomy, refractory period prolongations were most marked over the anterior ventricular surface; left stellate ganglionectomy produced the greatest prolongation on the posterior surface. Although the right and left stellate innervations of the ventricles overlap, the left stellate influence is predominant over the posterior wall of the ventricles, while right stellate influence dominates the anterior ventricular walls. The electrocardiographic form changes observed following unilateral alteration of sympathetic tone paralleled those electrocardiographic abnormalities seen in patients with lesions of the central nervous system, suggesting a possible functional explanation for these clinical findings.

Acetylcholine is directly perfused into the canine sinus node in vivo through its cannulated nutrient artery. After atropinization, the chronotropic action from intranodal administration of acetylcholine was identical to that of control injections of Ringer's solution alone. Eserine produced only a negative chronotropic action before, and no significant action after, atropinization. Direct perfusion of the sinus node with hemicholinium (1 mg/ml) completely blocked the response to vagal stimulation but had no effect on the response to stimulation of the stellate ganglion. The evidence is interpreted as indicating that acetylcholine has only a negative chronotropic action in the canine sinus node and that it neither releases significant amounts of local nodal norepinephrine nor is necessary to the normal response of the sinus node to stellate ganglion stimulation.

The effect of sympathetic nerve stimulation on the pulmonary circulation was studied in a hemodynamically separated dog lung in which blood flow was maintained constant with a pump. Electrical stimulation of the stellate ganglia at 3, 10, and 30 cps produced a significant increase in mean lobar arterial perfusion pressure. Since lobar blood flow and left atrial pressure did not change during nerve excitation, the increase in lobar arterial pressure reflected an increase in vascular resistance across the lung. The increase in pulmonary vascular resistance was dependent on the stimulus frequency, and the time from the onset of stimulation to the attainment of the peak response was inversely related to the stimulus frequency. The response to nerve stimulation was decreased by alpha-receptor blocking agents. Lobar perfusion with a roller pump had an effect similar to perfusion with a piston pump. The response to nerve stimulation was independent of changes in rate, rhythm, and volume of respiration, changes in aortic blood pressure, and changes in airway resistance. The effects of stellate stimulation and of injected norepinephrine on vascular resistance were similar in the hemodynamically separated lobe. In essence, sympathetic nerve stimulation produced an active increase in vascular resistance in the pulmonary circulation, and the contribution of passive factors such as changes in respiration, bronchomotor tone, and bronchial circulation was minimal. Since the response was blocked by phentolamine, the increase in pulmonary vascular resistance in response to nerve stimulation was attributed to activation of alpha receptors in the pulmonary vascular bed by neurally released norepinephrine. These results demonstrate that pulmonary vascular resistance can be increased at stimulus frequencies in the physiological range of discharge for the sympathetic nervous system.

In the canine isovolumetric left ventricle preparation, stimuli were applied to the efferent end of a transected cervical vagus nerve. The changes in left ventricular systolic pressure in response to vagal stimulation were compared under control conditions and when ventricular contractility was enhanced by various kinds of inotropic stimulation-left stellate ganglion stimulation, paired pacing, calcium chloride infusions, aminophylline infusions, and acetylstrophanthidin injections. Vagal stimulation under control conditions caused mean decreases in pressure in the various groups ranging from 13.2 to 16.2% of the control level. The depressant effect of vagal stimulation was potentiated during sympathetic stimulation, in confirmation of previous findings; the mean decrease in pressure produced by vagal stimulation during concurrent sympathetic stimulation varied from 20.1 to 20.8%. Calcium and aminophylline infusions had no significant effect on the ventricular response to vagal stimulation; the percent reductions in pressure caused by vagal stimulation were 18.1 and 15.5%, respectively. However, the response to vagal stimulation was markedly attenuated during paired pacing and after acetylstrophanthidin; the percent reductions in pressure caused by vagal stimulation were 5.0 and 9.4%, respectively. The potentiation of the response to vagal stimulation during increased cardiac sympathetic activity probably represents a specific adrenergic-cholinergic interaction.

Norepinephrine extraction and spillover rates were determined in the heart and lungs of anesthetized dogs under resting conditions, during sympathetic stimulation, and during epicardial pacing. The fractional extraction of norepinephrine across the coronary and pulmonary vascular beds was measured from the venoarterial difference in tritiated norepinephrine after infusion of a tracer dose to a steady state level. Cardiac extraction averaged 0.299 +/- 0.03 and pulmonary extraction averaged 0.215 +/- 0.014; extraction was unaffected by sympathetic stimulation or pacing. Norepinephrine spillover from sympathetic nerve terminals in the heart and lungs was measured from the venoarterial difference in endogenous norepinephrine and plasma flow after correction for the extraction component. Cardiac norepinephrine spillover increased linearly with increasing frequency of sympathetic stimulation to 7.44 times resting levels at 2 Hz. During pacing, there was no change in cardiac norepinephrine spillover despite marked changes in heart rate. Norepinephrine spillover was demonstrated under resting conditions in the lung and was greater than observed in the heart. Pulmonary norepinephrine spillover increased with sympathetic stimulation to 4.15 times resting levels at 2 Hz. It is possible to separate the contributions of norepinephrine extraction and spillover to measured venoarterial differences of norepinephrine under physiological conditions in the dog.

The effect of sympathetic stimulation (stellate ganglion) on dog cerebral and cephalic blood flows was studied via a cervical or a thoracic approach to the stellate ganglion under sodium pentobarbital or chloralose anesthesia. Two different stimulation voltages (3v and 5v) of monophasic pulses were applied for 1 minute. Venous outflow was measured at the confluence of the sagittal, straight and lateral sinuses with the lateral sinuses occluded and with them patent. When the lateral sinuses were occluded, stellate ganglion stimulation resulted in a marked decrease in common carotid blood flow to 38 plus or minus 2.5% (SE) of control and dilation of the ipsilateral pupil, but cerebral blood flow did not change. Similar effects were observed with each of the anatomic approaches, anesthetics, and voltages used and in dogs with low cerebral vascular tone induced by hypercapnia. When the lateral sinuses were kept patent, sympathetic nerve stimulation decreased the venous outflow to 89 plus or minus 2.9% of control and clamping both of the external jugular veins increased venous outflow to 120 plus or minus 2.7% of control. When the lateral sinuses were kept patent and the extracranial venous pressure was increased by clamping both of the external jugular veins, the decrease in venous outflow in response to sympathetic stimulation was even larger: venous outflow was only 65 plus or minus 4.9% of control. We conclude that stimulation of the stellate ganglion has no effect on the cerebral vasculature. Sympathetic stimulation significantly decreases venous blood flow measured at the confluence of the sinuses only when communications between the intracranial and extracranial venous vasculatures are present.

SUMMARY This study was performed to determine whether acute or chronic sympathetic denervation increases or redistributes cerebral blood flow (CBF) during hypotension or during the action of vasoconstrictor stimuli (hypocapnia and hypertension). Left superior cervical and stellate ganglionectomy was performed in anesthetized dogs. Total and regional CBF were measured by using microspheres. In acute experiments, hemorrhagic hypotension produced a redistribution of CBF which tended to preserve blood flow to the brainstem and to cerebral gray matter. Hypertension and hypocapnia did not redistribute CBF. Blood flows were similar in the acutely denervated and nondenervated half of the brain during control conditions, hypotension, hypertension, and hypocapnia. Completeness of sympathetic denervation was demonstrated by large increases in blood flow to the masseter muscle on the denervated side. Similar studies were undertaken 6-7 days after sympathetic ganglionectomy, at which time cerebral vascular catecholamines were depleted on the denervated side: norepinephrine content in innervated and denervated middle cerebral arteries was 3.1 ± 0.5 and 0.1 ± 0.02 ng/g, respectively. Blood flows in the chronically denervated and nondenervated half of the brain were similar during control conditions and during interventions. The major new findings in this study are, first, that hypotension produces a redistribution of CBF which tends to preserve blood flow to brainstem and to cerebral gray matter, and second, that acute "or chronic sympathetic denervation does not alter distribution of CBF over a wide range of arterial pressure or during hypocapnia.

The left of right cardiac sympathetic nerves in open-chest, anesthetized dogs were stimulated at mean frequencies of 2 or 4 Hz. The stimuli were applied intermittently, in patterns with repetition rates or either 60/min or 15/min, to simulate the spontaneous patterns of sympathetic neural activity that occur synchronously with the cardiac or respiratory cycles, respectively. With either repetition rate, intermittent stimulation of the left sympathetic nerves was about 10-20% less effective in enhancing myocardial contractile force (CF) and about 10% less effective in increasing coronary sinus blood flow than was steady stimulation at the same mean frequency. With right-sided stimulation, there was no appreciable difference between steady and intermettent stimulation patterns with respect to the effect on heart rate. With either left-or right-sided stimulation, the rate of norepinephrine (NE) overflow into the coronary sinus blood was 20-40% less with intermittent than with steady stimulation. Cocaine administration did not materially affect this difference in NE overflow. It was concluded that the higher instantaneous frequencies that prevail during intermittent stimulation result in a reduction in the rate of NE release at the sympathetic postganglionic nerve endings in the heart.

The effect of sympathetic alpha-receptor coronary vasoconstriction on myocardial oxygen tension was studied in open- and closed-chest, chloralose-anesthetized dogs. Blood oxygen tension in the coronary sinus and blood flow in the circumflex coronary artery were continuously measured in a three-part experiment. With stimulation of the left stellate ganglion (15 Hz, 3 msec, 4-7 v, 90-second train) under vagotomy control conditions (part 1), heart rate, blood pressure, and coronary blood flow increased, but coronary sinus oxygen tension decreased from 19 mm Hg to 15 mm Hg. In part 2, beta-receptor blockade with propranolol (2.0 mg/kg. iv) in the same dogs blunted the positive inotropic and chronotropic effects of sympathetic stimulation; coronary alpha-receptor vasoconstriction was unmasked, and coronary sinus blood oxygen tension fell from 17 mm Hg to 11 mm Hg. Since increases in oxygen metabolism were blunted, it appeared that the decrease in coronary sinus oxygen tension was caused by alpha-receptor coronary artery vasoconstriction. This hypothesis was tested in part 3 by the addition of alpha-receptor blockade with Dibozane (3.0 mg/kg, iv). Sympathetic stimulation no longer resulted in a change in either coronary vascular resistance or coronary sinus oxygen tension. These results indicate that the fall in oxygen tension during beta-receptor blockade in part 2 was due to alpha-receptor coronary vasoconstriction. Thus, myocardial oxygen tension may be regulated by coronary sympathetic vasomotion as well as by myocardial oxygen metabolism and local vascular control mechanisms.

The relationship between the increase in catecholamine levels of the coronary sinus blood and the amplitude of various cardiac responses to adrenergic nerve stimulation was studied in anesthetized dogs. Plasma catecholamine levels in both coronary sinus and aortic blood were measured by a modification of the radiometric enzymatic assay for tissue catecholamines and were found to be 0.622 plus or minus 0.104 (SE) ng/ml and 0.933 plus or minus 0.116 ng/ml, respectively, under basal conditions. The catecholamine levels in coronary sinus blood increased linearly during right cardioaccelerator nerve stimulation up to a frequency of 10 Hz. At this frequency, maximum values were observed in both coronary sinus blood catecholamine levels and cardiac responses. The correlation between the response in heart rate, mean coronary blood flow, and dP/dt of left ventricular pressure and the increase in endogenous catecholamine levels of coronary sinus blood was significant, but the relationship was nonlinear. The present experimental design may prove to be a reliable means of studying the role of the sympathetic nervous system in the regulation of cardiovascular function in vivo.

Recent studies from this laboratory have indicated that sympathetic alpha-receptor-mediated coronary vasoconstriction can compete with local metabolic vasodilation to reduce the oxygen supply to the myocardium. In vitro studies from other laboratories on isolated coronary vessel strips suggest that large epicardial vessels are the dominant site of adrenergic alpha-receptor activity. In this study we used anesthetized, open-chest dogs to test the hypothesis that alpha-receptor-mediated vasoconstriction occurs predominantly in epicardial vessels, which are partially removed from the metabolic milieu in the myocardium. Adrenergic beta-receptor blockade was achieved by propranolol (3 mg/kg, iv). The circumflex coronary artery was pump-perfused at constant pressure to minimize passive changes in large vessel resistance. Pressure was measured at the tip of the perfusion cannula sealed in the circumflex artery, and in an apical branch of the circumflex artery. Large vessel resistance was calculated as the pressure gradient along the vessel segment divided by the coronary flow. Intracoronary injections of nitroglycerin were used as an independent measure of the vasomotor responsiveness of the large vessel segment. Adrenergic alpha-receptor activation was produced by intracoronary bolus injections of norepinephrine and by electrical stimulation of the left stellate ganglion. Alpha-receptor stimulation caused an increase in total coronary vascular resistance; however, the relative increase in the resistance of the large vessel was only about 60% of that seen for the entire coronary bed. These data suggest that, contrary to the proposed hypothesis, adrenergic alpha-receptor-mediated vasoconstriction in the large coronary vessels is not proportionally greater than that observed in the total coronary vascular bed.

The diameter of a major coronary artery, the ramus interventricularis ventralis (RIV), was measured in dogs with arrested hearts perfused by an extracorporeal circulation. The resting diastolic diameter was 1.78 +/- 0.07 mm (mean +/- SE) at a diastolic pressure of 74.2 +/- 3.4 mm Hg. Bilateral supramaximal stimulation of fibers leaving the cranial pole of the stellate ganglion decreased the diameter by 71.2 +/- 8.9 micrometer, i.e., 4.0 +/- 0.5% of the resting diameter. Stimulation of the left stellate ganglion contributed 59.8 +/- 5.7% of the maximum response; that of the right contributed 40.3 +/- 5.5%. Stimulation of the thoracic ganglia (T2-4) resulted in a 1.2 +/- 0.4% decrease in coronary vessel diameter. RIV failed to respond to bilateral caudal cervical ganglion stimulation. After iv administration of phentolamine, 1-2 mg/kg, no response to sympathetic stimulation could be elicited. Therefore, it appears that alpha-receptors are activated by the release of the sympathetic neurotransmitter to sympathetic stimulation and that beta-receptors are not involved in the response of RIV to sympathetic stimulation.

We performed experiments to determine whether or not extracardiac factors are involved in the increase of the glycogen phosphorylase activity after coronary artery ligation in dog hearts. A branch of the left anterior descending coronary artery (LAD) was ligated for 1.5 minutes. The glycogen phosphorylase activity was determined in the endo- and epicardial layers. LAD ligation significantly increased the glycogen phosphorylase activity in both LAD (ischemic) and circumflex (nonischemic) areas. Pretreatment with reserpine (1 mg/kg, im, 24 hours before experiments) or hexamethonium (3 mg/kg, iv) prevented the LAD ligation-induced increase in the glycogen phosphorylase activity, but adrenalectomy did not. In the heart-lung preparation and stellectomized dogs, LAD ligation did not increase the glycogen phosphorylase activity in both ischemic and nonischemic areas. Changes in the glycogen phosphorylase activity in the endocardial layers were not essentially different from those in the epicardial layers. In other dogs, metabolic intermediates and ST segment of the surface electrocardiogram were measured. LAD ligation significantly decreased the tissue level of creatine phosphate and increased that of lactate in the ischemic but not in the nonischemic area. An elevation of ST segment occurred only in the ischemic area. Thus ischemic and nonischemic areas were confirmed. The ATP level, however, did not change in any of the ischemic and nonischemic areas. It is suggested that LAD ligation increases the glycogen phosphorylase activity in both ischemic and nonischemic areas probably by an increase in the efferent cardiac sympathetic nerve activity.

The purpose of this study was to examine the effect of secobarbital vs. alpha-chloralose anesthesia, and of tonic autonomic influence, upon ischemia-induced subepicardial and subendocardial bipolar electrogram changes during acute coronary artery occlusion in the open-chest dog. We found that the degree of bipolar electrogram alterations for similar reductions of regional myocardial blood flow was less in dogs anesthetized with alpha-chloralose than in those anesthetized with secobarbital. We also noted greater electrogram change when the sympathetic system was unopposed by the vagus nerves. Finally, this study demonstrates the reproducibility of ischemia-induced changes in bipolar electrograms during serial, short term, acute coronary artery occlusions.

Reflex bradycardia was elicited in rabbits via repetitive electrical stimulation of the central end of the sectioned left aortic nerve. Supramaximal stimulation produced a 16.9 plus or minus 1.3% (SE) increase in the R-R interval when vagal and sympathetic efferent pathways were intact. Reducing the stimulation voltage allowed selective stimulation of the myelinated (A) fibers, and polarizing electrodes placed central to the stimulus site permitted A fiber blockade and selective stimulation of the unmyelinated (C) fibers. When afferent A fibers were selectively stimulated, 64% of the maximum response was obtained; selective C fiber activation elicited 63% of the maximum observed response. Selective stimulation of A or C fibers after either vagotomy or stellectomy indicated that A fiber afferents elicit heart rate responses via both vagal and sympathetic efferents, whereas C fiber afferent information is mediated predominantly via vagal efferents. This afferent-efferent specificity of the aortic baroreceptor pathways suggests baroreceptor mechanisms normally used to modulate heart rate. Small increments in blood pressure would activate low-threshold A fibers and result in reciprocal changes in vagal and sympathetic efferent activity. More substantial increases in blood pressure would activate afferent C fibers and produce additional heart rate effects via vagal efferents.

Acute and chronic cardiovascular effects of repeated injections of 6-hydroxydopamine were studied in unanesthetized dogs. The acute response to initial injections of small doses of 6-hydroxydopamine consisted of an intense sympathomimetic effect characterized by a marked increase in blood pressure and a reflex bradycardia. Three days after the administration of a total dose of 50 mg/kg of 6-hydroxydopamine, significant decreases in heart rate, blood pressure, and total peripheral resistance were observed, but cardiac output remained unchanged. The absence of a response to intravenously injected tyramine, to electrical stimulation of the right stellate ganglion, and to bilateral carotid occlusion was consistent with a state of complete peripheral sympathectomy. Along with the functional impairment of transmission at the sympathetic nerve endings, marked endogenous norepinephrine depletion was observed in the heart, carotid sinus, and spleen. In contrast, endogenous norepinephrine was increased in the stellate ganglion. The response to electrical stimulation of the distal end of the right vagus was not altered by treatment with 6-hydroxydopamine. In anesthetized dogs treated with 6-hydroxydopamine, clamping of the hilar vessels of the adrenal glands for 10 minutes caused hypotension, but the same procedure in normal dogs did not affect blood pressure, demonstrating the importance of the compensatory role of the adrenal medulla in sympathectomized animals. It is concluded that 6-hydroxydopamine can be used effectively to obtain a generalized chemical sympathectomy in unanesthetized dogs and provides a useful means for studying the role of the autonomic nervous system in the regulation of cardiovascular functions.

Myocardial contractile force in five areas of the left ventricle was measured in 14 dogs under chloralose anesthesia prior to and following segmental chemical epicardiectomy. Electrical stimulation of the stellate ganglia was performed before and after each segment of epicardium was destroyed, and records of myocardial contractile response to stimulation were obtained. Epicardiectomy resulted in ablation of contractile response in specific areas of the left ventricle. In a recent publication from our laboratory (Circulation Res. 22: 315, 1968), the predominance of an epicardial distribution of sympathetic fibers was demonstrated. As an extension of that work, this study demonstrates the projection of fibers from both left and right stellate ganglia and the innervation of both anterior and posterior surfaces of the left ventricle. The major projections of sympathetic fibers are oriented along an axis passing from base to apex and can be interrupted by epicardial destruction in the area of the cephalad portion of the anterior descending coronary artery and adjacent to the origin of the left marginal artery.

Experiments were performed on open-chest anesthetized dogs to determine the quantitative effects of autonomic nerve stimulation on pacemaker activity and conduction. The lead II electrocardiogram together with bipolar electrograms were recorded from the atria, the His bundles, and the ventricles. The vagi or the stellate ganglia were stimulated in dogs which exhibited either sinus rhythm, ectopic atrial rhythm, junctional rhythm, or ectopic ventricular rhythm. The time courses of the change in heart rate in response to vagal or stellate stimulation were characteristic for each type of rhythm. The characteristic responses of different cardiac pacemaker sites to autonomic influence were demonstrated to be important factors in the production of wandering pacemakers and in the emergence of ectopic beats. Sinus pacemaker activity was more sensitive to modification by autonomic stimulation than was atrioventricular (AV) conduction. However, subliminal autonomic effects on AV transmission were brought out during conduction of premature atrial beats, thereby demonstrating a coupling interval dependency of autonomic influences on AV conduction. The present experiments also showed how fluctuations in autonomic activity could result in Mobitz type II second-degree heart block, pseudosupernormal conduction, and the concertina effect observed in the preexcitation syndrome.

Major branches from right and left sympathetic ganglia were electrically stimulated while force of contraction was recorded from multiple areas of the right and left ventricles. Stimulation of the stellate ganglia generally elicited alterations in force of contraction from all test segments, but excitation of selected nerve trunks induced responses in highly localized regions of the heart. Ablation of narrow strips of epicardium resulted in obliteration of contractile responses in specific, highly localized regions of the heart; thus a major fraction of the sympathetic innervation of the ventricular chambers is by way of the epicardial plexus. The anterior surface of the right ventricle is supplied by projection pathways arising within the immediately subepicardial regions of the right A-V groove and, to a lesser extent, from the tissues immediately adjacent to the left anterior descending artery. The left ventricle receives minor projections from the right A-V groove with major projections from subepicardial tissues along the left anterior descending artery. In some animals there also exists a definite left ventricular supply from the region of the left A-V groove. Whereas the thoracic vagi send dense projections to the atria, and particularly to nodal tissue, they also supply both ventricles with inhibitory and augmentor fibers. Although isolated cardiac nerves may carry predominantly sympathetic or parasympathetic fibers, many show rich intermingling of these fibers in trunks distal to the caudal cervical ganglion.

While pressures were recorded separately from the four chambers of the heart in openchest dog preparations, together with a standard limb lead of the ECG, the right and left stellate ganglia were stimulated. Measurements from the beginning of the P wave to the initial rise in chamber pressures revealed that pressure in the left ventricle began to rise approximately 20 msec prior to that in the right ventricle. This interval was significantly shortened during electrical stimulation of the stellate ganglion. Although the interval between P wave and the initial pressure elevations in the right atrium (RA) and the left atrium (LA) both shortened during stellate stimulation, P-RA decreased considerably more than P-LA. The P wave, PR interval, and the QRS complex all shortened markedly, indicating increased conduction velocity over the atria, the nodal conducting tissue, and the ventricles during stellate stimulation. These events occurred with or without alterations in heart rate and were correlated with large increases in the rate of change of ventricular pressures. Simple algebraic summation of individual muscle segment contractions may achieve significantly greater intraventricular pressure as the contractions become more synchronous. The total pressure developed is further augmented when each individual unit contracts with greater force and greater velocity. The recruitment of additional functional units would result in further summation and thus in still greater augmentation in total force and pressure. Alterations in pacemaker site are common during stellate stimulation.

During prolonged electrical stimulation of the stellate ganglion in the dog, blood pressure, and more specifically systolic pressure, is significantly elevated. The amount of augmentation in pulse pressure is related to the frequency of stimulation, as is the duration of the sustained pressor response. Postmortem examination of the hearts revealed a distinctive pattern of subendocardial hemorrhages distributed along the long axis of the papillary muscles and the trabeculae carneae. They also frequently marked the endocardium at the base of the mitral valve. These lesions appeared in the left ventricle in 29 of 33 animals in which the left stellate ganglion was stimulated at a frequency of three to ten per second. They were found in only five instances in the right ventricles of these animals. The incidence of lesions in the ventricular wall was low during stimulation of the right stellate ganglion, although the mitral valve became involved in about 50 per cent of the animals. The similarity of the lesions to those induced during hemorrhagic hypotension, elevated intracranial pressures, and infusion of large quantities of the catecholamines suggests the possibility of a similar etiology. A satisfactory explanation of the lesion, however, is not yet available.

The influence of cardiac sympathetic nerve stimulation on certain aspects of ventricular excitation was examined in dogs at constant heart rate. Bipolar electrodes were used to record electrical activity from atrium, His bundle, right Purkinje papillary muscle junction, basal interventricular septum and the epicardial surface of the ventricles. Stimulation of the left stellate ganglion reduced markedly the A-V nodal delay, but produced little or no change of conduction velocity in the Purkinje system or in the pattern of epicardial depolarization. Conduction in muscle was enhanced slightly by stellate stimulation and appears to account for the observed decrease of 2 to 3 msec in ventricular activation time. These data indicate that cardiac sympathetic nerve stimulation does not result in major changes in the sequence of ventricular excitation.

The sympathetic control of left circumflex coronary blood flow was studied by stellate ganglion and hypothalamic stimulation. Flow was measured acutely in dogs under chloralose or pentobarbital anesthesia by an electromagnetic flowmeter. Coronary vasoconstriction was observed when the left stellate ganglion was stimulated after blocking beta receptors with propranolol or INPEA. The possibility of coronary sympathetic cholinergic vasodilation of the type found in skeletal muscle was examined in three types of experiments in which vasodilation occurred in skeletal muscle: (1) peripheral sympathetic nerve stimulation after chronic catecholamine depletion with reserpine, (2) peripheral sympathetic nerve stimulation after acute block of alpha and beta receptors, (3) hypothalamic stimulation to give a defense reaction after acute block of alpha and beta receptors. Within the limits of these experiments, skeletal muscle cholinergic vasodilation was demonstrated but coronary sympathetic cholinergic vasodilation was not, even as part of the defense reaction elicited by hypothalamic stimulation.