Functional Anatomy of the Cardiac Nerves in the Baboon' WALTER C. RANDALL, J. ANDREW ARMOUR, DAVID C. RANDALL AND ORVILLE A. SMITH Departments of Physiology, Loyola University, Chicago, and t h e University of Washington, Seattle, and t h e Regional Primate Research Center, University of W a s h i n g t o n , Seattle ABSTRACT In 17 anesthetized baboons, the autonomic innervation of the heart was carefully exposed and electrically stimulated to determine the course of fibers having direct inotropic and/or chronotropic actions. The superior cervical and nodose ganglia are intimately associated by means of short but large interconnections, and the sympathetic and parasympathetic trunks descend with the carotid artery in a common epineural sheath. The middle cervical ganglion is invariably well defined and completely separated from the vagus trunk in the upper portion of the thoracic cage. Direct nervous connections between the sympathetic and vagal trunks are frequent at all levels within the thorax. Both systems also send small nerves into the phrenic and recurrent laryngeal nerves. Separate inferior cervical and first thoracic ganglia were not found, but rather, a large and well defined stellate ganglion extending across the heads of the first and second ribs. The stellates are connected to the middle cervical ganglia by means of both dorsal and ventral ansae subclavia of varying size. Although fine nerves arising from the upper thoracic trunk were located, they appeared to have no direct inotropic or chronotropic actions. The major sympathetic and parasympathetic nerves converge upon richly interconnected dorsal and ventral cardiopulmonary plexuses and several minor (superior vena cava, left atrial, pulmonary veins) plexuses. Both ipsilateral and contralateral control of cardiac function is possible through these pathways. The baboon cardiac innervation thus appears to resemble that of man in some respects and the dog in others. The nerve supply to the canine heart has been investigated in considerable detail including emphasis upon both anatomic and functional systems. Similarly, the cardiac innervation in man is well described anatomically, but due to obvious difficulties in carrying out experimental procedures in man, relatively few functional responses to nerve stimulation have been reported. Recent interest in the question of cardiac reinnervation following surgical transplantation reiterates the importance of more precise knowledge of this system. Direct carry-over of information from the experimental animal (dog, cat, rabbit) to primates may not be permissable, and therefore, it was decided to experimentally pursue several pertinent questions concerning the innervation of the heart in nonhuman primates. ANAT. REC.,170: 183-198. Diligent search of the literature failed to reveal significant information on the gross innervation of the nonhuman primate heart. A description by Kuntz ('33) in the autonomic nervous system of the Rhesus monkey proved to be grossly inadequate. Sketches by Perman ('24) are so diagrammatic as to be of little use. Regiele ('26) described in considerable detail his dissections in a single baboon with sketchy confirmation from a second specimen. These workers did not, of course, include functional data in their descriptions. Thus, it became apparent that while a few important physiological, surgical and behavioral studies have been performed in Received June 24, '70. Accepted Oct. 14, '70. 1 Supported by NIH grant HE 08682 to Loyola University and grant FR 00166 to the Regional Primate Research Center, Seattle. 183 184 W. RANDALL, J. ARMOUR, D. RANDALL AND 0. SMITH this group of animals, few if any structurefunction relationships have been reported. A very real need exists for both anatomical description and functional testing of the autonomic cardiac nerves in the nonhuman primate, and consequently the present experiments were undertaken on the baboon. METHODS Acute experiments were carried out in seventeen open-chest baboons (Papio anubis and Papio cynocephalus) anesthetized with Serynlan (0.2 to 1.0 mg/kg) and chloralose (20 to 60 mg/kg) in propylene glycol. Modified Walton-Brodie strain gauge arches were stitched to the epicardial surfaces of all four cardiac chambers, frequently with two gauges on both left and right ventricles. These gauges permit essentially isometric recording of contractile force from an approximately 1 cm segment of myocardium. Artificial respiration was supplied by means of a Bird Mark 7 respirator. Arterial blood pressure (and/or in traventricular pressure) was recorded by Statham P23Db pressure transducers with all recordings on a model R Beckman Dynograph. All visible nerves were carefully dissected in the thoracic region from the clavicle to the level of approximately T6, including, of course, the cardiac nerves. Each nerve on both right and left sides was stimulated by means of bipolar electrodes and either a Grass model 5 square wave voltage generator or a Nuclear-Chicago model 7151 constant current stimulator. When using the voltage generator the stimulating voltages were monitored on a cathode ray oscilloscope and maintained at the stated level throughout the period of stimulation. Immediately following stimulation, an identifying ligature was loosely looped around the nerve at the point of stimulation for later anatomical identification. In addition, the vagosympathetic trunk in the neck was carefully dissected free and the sympathetic trunk separated from the vagus, so that each trunk could be isolated and electrically stimulated. Thus, electrodes were successively applied to the cervical vagosympathetic trunk as well as to its individual components, to the thoracic vagus, to each of the small nerves taking origin from the middle cervical ganglion, dorsal and ventral ansae, caudal pole of the stellate ganglion, and to the interganglionic segments from T3 through T6, the latter both before and after transection of the trunk above and below each ganglion. RESULTS The efferent innervation of the baboon heart consists of fibers coursing in the vagus and in the cervical-thoracic sympathetic nerves. Contrary to conventional concepts, these divisions of the autonomic nervous system are neither anatomically nor functionally isolated from each other, since direct and frequently profuse interconnections occur at all levels including the superior cervical, middle cervical and stellate ganglia. Figures 1 and 2 present detailed descriptions of the typical anatomic distribution of autonomic nerves to the baboon heart. Variable interconnections exist between right and left sides including branchings from both sympathetic and vagus trunks. There exists a concerted flow of nerves from the lateral position of the stellate and middle cervical ganglia, medially and caudally, to enter the large plexuses situated over the tracheal bifurcation, the pulmonary artery, and aortic arch. These plexuses are intimately interconnected and are extended distally to both atria, large central veins and arteries to provide innervation for both ventricles. Thus, nearly every portion of the heart is provided with innervation from both right and left sympathetic and parasympathetic trunks. In addition, there generally exist major pathways directly from the right sympathetics to the superior vena cava and right atrium as well as from the left middle cervical ganglion (and sometimes the stellate) to the posterior surface of the left atrium and ventricle. The cervical vagosympathetic trunks The superior cervical ganglion is situated medial to the nodose ganglion on either side at the level of C2-C3. The two ganglia are tightly interconnected by means of a varying number of short but large nerves, and receive connecting rami from, or branching to, the pharynx, esophagus, laryngeal and hypoglossal nerves, the CARDIAC NERVES IN THE BABOON 185 Fig. 1 Ventral view of the primary cardiac nerves on the left side, baboon 8. Dashed lines indicate the nerves course dorsal to the overlying structure. plexuses on the external and common caro- clature in existing descriptions of this systid artery, and the upper three or four tem in higher mammals including man, cervical nerves. The large cervical sym- we have elected to employ those names pathetic trunk enclosed within its sheath, most commonly accepted for the primary extends from the caudal pole of the su- structures (stellate, middle cervical ganperior cervical ganglion and joins with the glion, vagus, phrenic, etc.) and appropriate vagus approximately 1.5 cm distal to the anatomically descriptive names for the caudal pole of the nodose ganglion. Thus, nerves themselves. In this way it is hoped while the cervical sympathetic and vagus that we may convey a three dimensional are each invested within their own sheaths, picture of the nervous projections onto the both are wrapped within a common sheath multidimensional surfaces of the heart. as they descend caudally with the carotid Our nomenclature thus substitutes the artery. They are readily separated at any term dorsal cardiopulmonary plexus for the level between the superior and middle conventional “pretracheal” or “deep” cardiac plexus and the term ventral cardiocervical ganglia. Figure 1 illustrates the origin of cardiac pulmonary plexus for “aortic” or “supernerves on the left while figure 2 shows a ficial’’ plexus. Our objective is to emphasize comparable display on the right, both ac- the facts, recognized by Mizeres (’63) and curately reflecting the anatomic origin and McKibben and Getty (’69), that there is course of the major cardiac nerves. In view no real separation between deep and superof the remarkable differences in nomen- ficial divisions nor separation from the pul- 186 W. RANDALL, J. ARMOUR, D . RANDALL A N D 0. SMITH Fig. 2 Lateral view of the primary cardiac nerves on the right, baboon 8. monary plexus. Rather, there is demonstrable input to a common plexus from both right and left sides and from both sympathetic and parasympathetic divisions with rich interconnections at every level. Further, it is important to recognize that visual separation of the cardiac and pulmonary innervations is impossible at the plexus level, and interruptions here invoke alterations in both cardiac and pulmonary functions. Finally, it should be noted that many large nerve trunks maintain continuity through the plexus with direct projections onto different surfaces of the heart including both atrial and coronary plexuses. Figures 3 and 4 illustrate the responses to electrical stimulation of the cervical vagus and sympathetic trunks, as well as the combined vagosympathetic complex, in a single animal. In figure 3A, the isolated sympathetic portion of the right trunk was stimulated with positive inotropic and chronotropic alterations developing in all test segments. Heart rate accelerated from control levels of 162/minute to a maximum of 198/minute while arterial blood pressure increased from 65/46 to 80/50 mm Hg. Augmentation in contractile force was most pronounced on the right heart, particularly in the right atrium (RA) and right ventricular sinus (RVS) which encompasses the inflow tract of the right ventricle. Augmentation was not prominent on the left heart but was discernible on both apex (LVA) and base (LVB). Alterations in rate of increase in contractile force were also indicated (dF/dt) as measured from the ascending slopes of the fast traces in each myocardial segment. Panel B shows bradycardia (HR 162 to 84/minute) and decreased arterial blood pressure (65/40 to 40/20 mm Hg), together with suppression in contractile force and dF/dt, on some segments during stimulation of the vagal portion of the H 10 scc H 200mscc RT. CERV. VAGUS B - RT. VAGO. SYMP. C Fig. 3 Responses to electrical stimulation (10 cps, 5.0 msec, 4.0 v ) of the caudal pole of the superior cervical ganglion (panel A ) , caudal pole of nodose ganglion (panel B ) , and of the combined right vagosympathetic trunk 1.5 cm distal to the junction of the cervical sympathetic and vagal trunks (panel C ) , baboon 16. The period of stimulation is indicated by the signal marker i n each panel. mmHg BP&wwwdUuL-0 RT. CERV. SYMP. A MAR19170 - 4 188 W. RANDALL, J ARMOUR, D. RANDALL AND 0. SMITH . 0 h ? K a I E CARDIAC NERVES IN THE BABOON vagosympathetic trunk. Right atrial contractile force showed the greatest amount of inhibition with relatively little suppression in force on RVS and LVA. Note the progressive recovery of heart rate and contractile force following cessation of stimulation, unaccompanied by post-stimulation tachycardia or overshoot in contractile force. Panel C presents responses to identical electrical stimulation of the combined vagosympathetic trunk 1 cm distal to the point of stimulation of each of the components illustrated in panels A and B. Although bradycardia occurred (162 to 102/ minute) it was of lesser magnitude than induced during similar stimulation of the vagus alone. Whereas depression in contractile force was less in the right atrium, i t was essentially comparable to that in panel B in the left atrium. Contractile force was inhibited only on the left ventricular base, whereas that on all other ventricular test segments was augmented. Careful scrutiny of the ventricular records reveals almost exact algebraic summation of responses to separate stimulation of the sympathetic (3A) and parasympathetic (3B) trunks. Mean arterial pressure showed little or no change although pulse pressures were distinctly increased. Post-stimulation “rebound in contractile force was prominent in the atrial traces and postvagal stimulation tachycardia ( 192/minUte) was evident in the recovery traces immediately following cessation of stimulation. Figure 4 shows responses to separate electrical stimulation of the left cervical vagosympathetic trunk in the same baboon as illustrated in figure 3. During excitation of the sympathetic portion of the trunk (3A), acceleration in heart rate (168 to 180/minute), increased arterial blood pressure (65/45 to 75/50), and augmentation in contractile force occurred. The latter characterized all myocardial test segments and was accompanied by increased dF/dt. Excitation of the vagal component (panel B) induced cardiac slowing (162 to 120/ minute) and decreased contractile force. The latter was most prominent in the atrial traces but was distinctly apparent in all chambers. Recovery in both heart rate and contractile force was gradual and did not show postvagal tachycardia or re- 189 bound. Stimulation of the combined vagosympathetic trunk (panel C ) elicited changes in cardiac dynamics showing characteristics of both of the previous procedures (A and B). The induced bradycardia was less intense (164 to 150/minute), the decline in arterial blood pressure was minimal, and inhibition in contractile force was confined primarily to the right atrium where i t was much less intense. Recordings from several ventricular segments actually showed increased contractile force. There appeared a modest postvagal tachycardia (to 175/minute) and clear “overshoot” in contractile force of the atrial segments. Comparison of the indivdual myocardial segment responses in figures 3 and 4 reveals qualitative as well as quantitative difference in responses. Note, for example, the greater magnitude of right atrial response during excitation of the right vagosympathetic components. While the right stimulations excited relatively marked augmentation in contractile force of both right ventricular segments, left side stimulation elicited only minor changes. The primary distribution of augmentor fibers to the left atrium is by way of the left sympathetics. Thus, in spite of great intermingling of nerves from right and left, there is evidence for localized distribution of both components of the autonomic innervation of the primate heart. Electrical excitation of the vagus trunk in the thorax as compared with vagosympathetic stimulation in the neck further revealed a different composition of the nerves in these two locations. Figure 5 illustrates such stimulations before and after atropine. Panels A and B demonstrate typical responses to electrical excitation of the right thoracic (A) and cervical (B) vagosympathetic trunks, while panels C and D represent comparable stimulations on the left. In each instance the thoracic vagus exerted greater negative inotropic and chronotropic influences than did the cervical levels of the vagosympathetic trunks. The per cent changes in arterial blood pressures were correspondingly less during stimulation of the cervical trunks. After atropine, identical stimulation elicited slight to moderate positive inotropic and chronotropic responses, indicating the W. RANDALL, J. ARMOUR, D. RANDALL A N D 0. SMITH In > U CARDIAC NERVES IN THE BABOON 191 often be traced across to the large nerves and the middle cervical ganglion on the opposite side (figs. 1, 2, 6). Distinct and sometimes large nerves connect directly between the ventral ansa or the stellate ganglion and the vagus nerve (stellate vagal nerve). These were often in addition to smaller connections on the right between the inferior loop of the recurrent nerve and the ventral ansa. Rich intermingling of vagus and sympathetic nerves occurred caudal to the middle cervical ganglion. The cardiopulmonary plexus arises from multiple branches from both sympathetic and parasympathetic trunks (fig. 6). From two to five nerves coursing Middle cervical g a n g l i m toward the heart from the left middle The middle cervical ganglion is remark- cervical and stellate ganglia send major ably variable in size and shape, occasion- projections to the region of the tracheal ally consisting of several small ganglionic bifurcation, the root of the pulmonary swellings connected in series. The superior artery, and to the concavity of the aortic poles of the ganglion receives several small arch. The left recurrent nerve arises as a nerves, described by Riegele as coming major branch of the vagus at the level of from the third through the sixth cervical the aorta which it encircles, its branches nerves. There are also a variable number intermingling with sympathetic nerves of small nerves which interconnect with from the middle cervical ganglion, and the recurrent, the phrenic and the vagus courses rostrally. Many nerves pass from nerves. In many instances fine nerves may the aortic arch to the ventral surfaces of be traced from the ganglion in both direc- the pulmonary artery and along its right tions along the subclavian artery. Com- and left primary divisions, as well as to the pletely encircling the subclavian artery the left atrium and pulmonary veins. Extendorsal and ventral ansae interconnect the sions of this plexus invest the coronary middle cervical and the stellate ganglia. In arteries to innervate both right and left some animals the ventral ansa is larger ventricles. than the dorsal but in a majority, the dorsal A majority of the larger nerves from the ansa is more massive. Small nerves branch right sympathetics and vagus converge on off both ansae and pass either superiorly the region between the aorta and pulto join with rami of the cervical nerves or monary artery and contribute to the dense enter plexuses around the subclavian and plexus dorsal to the aortic arch and overinnominant arteries. Fiber connections can lying the bifurcation of the trachea. It is Fig. 5 Responses to electrical stimulation (10 comparable to the pretracheal or deep cardiac plexus described in the dog and man. cps, 2 msec, 0.3 m a ) of the vagosympathetic It also receives multiple branches from the nerves in the cervical and thoracic regions before (upper) and after (lower) atropine (1.0 mg/kg). vagosympathetic system on the left, and Onset and duration of stimulation is indicated is richly interconnected with the plexus by signal marker at top of channel 1. Strain situated in the arch of the aorta. There gauge arches were applied to the right atrium ( R A ) , right ventricular conus (RVC), right ven- are prominent extensions to the right tricular sinus (RVS), and left ventricular base atrium, superior vena cava, right pulmon(LVB). Systemic arterial blood pressure is shown ary artery and veins. In order to preserve in channel 5 . All records were made following appropriate dorso-ventral relationships bilateral cervical vagotomy and decentralization of the upper thoracic sympathetic trunk. Stimula- without obfuscation from the plethora of tions were applied to the right thoracic vagus existing interconnecting nerves, only a few ( A and E ) , peripheral end of the right cervical of the major contributions from the right vagosympathetic trunk (B and F ) , left thoracic vagus (C and G), and left cervical vagosympa- and left cardiac nerves are depicted in figure 6 . thetic trunk (D and H). presence of sympathetic fibers in these nervous structures conventionally thought to be exclusively parasympathetic. The increases in heart rate varied from 20 beats/ minute in E, 15/minute in F and G , and lO/minute in H. With exception of the right thoracic vagus (panel E), all responses were characterized by augmentation in contractile force on both right and left atrial and ventricular surfaces. Thus, the presence of both sympathetic and parasympathetic fibers in the vagosympathetic trunks, both superior and inferior to the middle cervical ganglion is clearly indicated. 192 W. RANDALL, J. ARMOUR, D. RANDALL A N D 0. SMITH Fig. 6 Sketch of the cardiopulmonary plexus from left lateral view. In order to reveal important dorso-ventral relationships of the plexus, as well as the rich interconnections between these levels, only a portion of the major cardiac nerves descending from the right and left vagosympathetics are shown. Figure 7 illustrates the considerable variability in anatomical structure of the middle cervical ganglion and its ansal connections with the stellate. In none of the seventeen specimens dissected could separate inferior cervical and first thoracic ganglia be differentiated, all showing distinct and large cone-shaped stellate ganglia on both sides. In some animals the ventral ansa was large and represented the primary connection between the stellate and middle cervical ganglion (fig. 1). In many animals the dorsal ansa was the more massive and often fused directly with the middle cervical ganglion (fig. 7). The ventral ansa was occasionally very tiny and gave rise to no subsidiary nerves. In a few instances, distinct ganglionic swellings were small or absent and the ansae themselves were enlarged and gave rise to cardiac nerves. On the left, the vagus descended medial to the middle cervical ganglion and in most specimens made direct neural interconnection with it. On the right, the vagus generally crossed directly over the subclavian artery at the level of the junction between the ansa subclavia and the middle cervical ganglion. Here again numerous discrete, small nerves connected the sympathetic and parasympathetic systems. In no instance was the middle cervical ganglion wrapped within a common epineurium with the vagus as is generally true in the dog. The vagus continued distally and gave rise to the recurrent nerve which in turn contributed large branches to the cardiac plexuses. The middle cervical ganglion on the right was sometimes discrete and globular (fig. 2), but in other animals consisted of a series of small ganglionic swellings fused together and gave rise to nerves which passed both superiorly and inferiorly along the subclavian and innominate arteries. Fewer nerves were sent directly to the cardiac plexus than from the left, but rich communication with the vagus and recurrent nerves provide sympathetic pathways to the heart. Riegele noted numerous small 193 CARDIAC NERVES IN THE BABOON RECURRENT MIDDLE CERVKP s1'ELLATE ' 8 . RtGHT SIDE LLATE LEFT SIDE RECURRENT Fig. 7 Illustrating anatomic variations in form and structure of the stellate and middle cervical ganglia in the baboon (all different from figs. 1, 2). Note differences in ansal connections, the origins of nerves, and the variable patterns of interconnection between sympathetic and parasympathetics. connections from this ganglion (or its ansae) with the C3-C7 nerves. Electrical stimulation of the stellate ganglion or the ansa subclavia often caused marked bradycardia and arrhythmia accompanied by augmentation in cardiac contractile force. Although such arrhythmias are sometimes encountered during stimulation of the left stellate in the dog, they are less common during excitation of the right sympathetics in this species. Another functional variation in responses to stellate stimulation in the two species (baboon and dog) is related to heart rate. In the dog, the right sympathetics generally (not invariably) cause greater acceleration in heart rate with lesser influence on systemic arterial pressure. Stimulation of the left stellate frequently exerts little or no influence on heart rate but elicits large changes in ventricular pressure and systemic arterial pressure. Such dichotomy in response was less evident in the baboon, and both heart rate and contractile force changes were generally associated with both right and left sympathetics. The thoracic vagus The left vagus often gives rise to one or two large cardiac nerves just superior to the middle cervical ganglion. These may have rich interconnections with the posterior ansa, the middle cervical ganglion, and with the cardiac nerves originating from the ganglion. There are also branches coursing to the subclavian and left common 194 W. RANDALL, J. ARMOUR, D. RANDALL A N D 0. SMITH carotid arteries and to the trachea. Inferiorly these nerves divide and send multiple branches into the cardiopulmonary plexus. There are also numerous fine neural connections with the phrenic nerve, the pericardium, and pulmonary veins. The right vagus is anatomically very close to the middle cervical ganglion, and gives off a large recurrent nerve which loops around the subclavian artery and ascends parallel to the cervical vagosympathetic trunk. At its point of origin from the vagus it gives rise to a large independent nerve which quickly divides and repeatedly interconnects with both sympathetic and vagal components along its course to the cardiopulmonary plexus. This nerve we have called the recurrent cardiac nerve (figs. 2, 7). Its electrical stimulation elicited either or both sympathetic and parasympathetic effects, the responses being functionally differentiated by the use of atropine to block the cholinergic responses. As the right thoracic vagus descends it crosses the vena cava, passes dorsally along the pericardium over the right atrium and gives off many branches to it and to the main branches of the pulmonary arteries and veins. Stellate and upper thoracic sympathetic ganglia The stellate ganglion was consistently found along the anterior aspect and extending from the head of the first to the second rib. It measured 1.5 to 2 cm in length, 5 to 7 mm in width at its superior end and tapering to 1 or 2 mm at its caudal pole. It typically gave rise to rami to the sixth, seventh, and eighth cervical nerves and to the first and second thoracic nerves with those from T3 sometimes entering at its caudal pole. Sender rami sometimes passed from the rostral pole to join the subclavian and vertebral arterial plexuses. In only a few specimens did a distinct nerve arise from the medial aspect of the rostral pole to course directly to the cardiac plexus (inferior or stellate cardiac nerve). Caudally, the thoracic sympathetic trunk consisted of segmental ganglia, each connected with its associated intercostal nerve by both white and gray rami. From the medial side of the ganglia (and frequently from the interganglionic segment) arose multiple small fibers passing to the thoracic viscera, including pulmonary veins, pericardium and cardiopulmonary plexus. Figures 8 and 9 illustrate an investigation of the functional role of small nerve fibers which may be traced from the upper thoracic sympathetic trunk directly toward the heart. Panel 8A shows slight augmentation in force of contraction on the ventricular surfaces, together with a marked elevation in arterial blood pressure elicited by electrical stimulation of the left sympathetic trunk at the interganglionic segment between T5 and T6. Right atrial contractile force was simultaneously suppressed with partial A-V block. The left thoracic trunk was then surgically sectioned immediately above the T5 ganglion and stimulation repeated at the T5-T6 segment with marked attenuation in response (8B). The thoracic trunk was next transected below the T6 ganglion, followed by stimulation at T5-T6 with abolition of all cardiac response (8C). We conclude from these results that some fibers destined for the heart ascended in the sympathetic trunk at this level, but that a minor fraction of the response in panel A was dependent upon electrical excitation of the splanchnic innervation. These were obliterated by transection at T6. The complete absence of response during stimulation of the isolated ganglion at T5 argues strongly against direct transthoracic pathways from this ganglion having any inotropic, dromotropic, or chronotropic action on the heart. The suppression in atrial contractile force was presumably related to the operation of baroreceptor reflexes. Stimulation at the T3-T4 segment next revealed an excellent cardiac response with participation of all test segments and elevation in arterial blood pressure (8D). Transection of the trunk at T3 followed by repeated stimulation of the isolated T4 ganglion again failed to induce any appreciable cardiac response (8E). The electrodes were then placed on the caudal pole of the left stellate ganglion at the vertebral level of T2 with resultant augmented contractile force (8F) and elevation in blood pressure, again with atrial depression and A-V block. A somewhat lesser response was elicited by stimulation of the ventral ansa ( G ) , as compared 195 CARDIAC NERVES IN THE BABOON A B C D RVS Fig. 8 Responses to electrical stimulation (10 cps, 2 msec, 0.3 ma) at the interganglionic segments of the upper thoracic sympathetic trunk on the left, baboon 4. Panel A stimulation interganglionic segment T5-T6; panel B, repeat stimulation at T5-T6 after surgical section above T5 ganglion; panel C, repeat stimulation at T5-T6 after section below T6 ganglion; panel D, stimulation at T3-T4 segment; panel E, repeat stimulation after section at T3; panel F, stimulation caudal pole left stellate ganglion; panel G, stimulation ventral ansa; H, stimulation dorsal ansa. with that elicited from excitation of the dorsal ansa (H). The virtual absence of cardiomotor response to electrical stimulation of the right thoracic sympathetic trunk at the T5-T6 interganglionic segment in the same animal (shown in fig. 8) demonstrates a distinct variation in level of outflow with respect to fibers ascending to the heart as well as those descending to the splanchnic system (fig. 9A). Placement of the electrodes at the T3-T4 segment however, induced excellent overall cardiac response (9B) which was abolished after transection of the trunk at T3 (9C). Again, there was little or no visible evidence for direct cardiomotor influences of nerves arising from this portion of the thoracic trunk. Excitation of 196 W. RANDALL, J. ARMOUR, D. RANDALL AND 0. SMITH I- a RVS II LVB Lo Fig. 9 Responses to electrical stimulation ( a s in fig. 8 ) upper thoracic sympathetic trunk on the right, baboon 4. Panel A, stimulation interganglionic segment T5-T6; panel B, stimulation a t T3-T4 segment; panel C, stimulation T3-T4 after surgical section a t T3; panel D, stimulation ventral ansa; panel E, stimulation dorsal ansa; panel F, stimulation caudal pole right stellate ganglion. the ventral (9D) and dorsal (9E) ansae elicited positive inotropic and chronotropic responses. Stimulation of the caudal pole of the right stellate ganglion induced similar alterations in heart rate as well as in contractile force (9F). DISCUSSION As Mizeres ( ' 6 3 ) found in describing the cardiac nerves in man, no constant pattern of nerve distribution from the superior, middle and stellate ganglia was observed in the baboon. Except for the main cervical sympathetic trunk enveloped with the vagus within the carotid sheath, no separate cardiac nerve descends from the superior cervical ganglion. Interconnections between the sympathetic and parasympathetic trunks exist between the nodose and superior cervical ganglia, at the level of CARDIAC NERVES IN THE BABOON 197 the middle cervical ganglia, and profusely tility is invariably suppressed during vagal at all levels distally into the cardiopul- stimulation. However, in panels B and C monary plexus themselves. Thus, anatomi- of figures 3 and 4, atrial contractile force cal substrate within the cervical vagosym- was dramatically reduced while ventricular pathetic trunk for both positive and contractile force either increased or denegative inotropic and chronotropic regu- creased, depending upon the predomilation of cardiac activity is provided. There- nance of sympathetic or parasympathetic fore, careful separation and testing of in- fibers. Predominance of right sided reguladividual nerve pathways must be assured tion of heart rate and left sided control before either “purely” sympathetic or para- of contractile force as observed in the dog sympathetic control may be assumed. by Randall and Rohse (’56) was not promiThe convenient tendency to interpret nent in the baboon. Connections between results of cervical vagosympathetic stimuright and left thoracic cardiac rami would lation only in terms of parasympathetic resuggest a significant difference from husponses has been discussed elsewhere by Randall, Pace, Wechsler and Kim (’69). man anatomy if Mizeres’ (’63) observaThe interpretation of “postvagal tachy- tions are correct but illustrate similarity cardia” as resulting from excitation of with human anatomy if one accepts the cholinergic parasympathetic fibers leading anatomical descriptions of Ellison and to liberation of catecholamines by Copen, Williams (’69). In none of the animals was there disCirillo and Vassale (’68), as well as the cholinergic-link hypothesis of Burn and tinction between inferior cervical and first Rand (’59) appear to be untenable. Figures thoracic ganglion. That is, a clearly de3 and 4 reveal the sharp, differential re- fined, large stellate ganglion was invarisponses to separate electrical excitation of ably located at the head of the first and both sympathetic and parasympathetic second ribs. This is in contrast to the obcomponents of the vagosympathetic trunk. servation in man that a fusion of inferior Eserine fails to unmask any adrenergic re- cervical and first thoracic ganglia occurs sponse to “purely” vagus stimulation. Post- in 37.7% (Becker and Grunt, ’57), 75vagal tachycardia may result from elec- 80% (Mitchell, ’53) or 88% (Ellison and trical stimulation of the combined cervical Williams, ’69) of cases. In this respect, vagosympathetic trunk, but does not follow then, the baboon more closely resembles excitation of the parasympathetic com- the dog. Distinct ventral and dorsal ansae ponent alone. Thus in studies involving subclavia were found on both sides of all autonomic innervation of the heart, sym- animals, connecting the stellate and midpathetic responses must be separated, both dle cervical ganglia. Considerable varianatomically and functionally, from para- ability was found, however, in their relasympathetic responses before consideration tive size. Corresponding variations in of the release of norepinephrine by acetyl- functional control were exerted by the two choline is warranted. A straight-forward ansae, the larger structure generally elicitinterpretation of direct responses to cholin- ing the more profound cardiac responses ergic or adrenergic nerves satisfies all ex- when electrically stimulated. The upper perimental results in both the dog (Ran- four or five thoracic sympathetic ganglia dall, Pace, Wechsler and Kim, ’69) and the gave rise to small filaments that ran independently and directly toward the heart, baboon. A further interesting question regarding and occasionally (as reported by Riegele) the direct action of parasympathetic fibers entered the cardiac plexuses. It is doubtful upon the ventricular musculature would however, that these fibers contribute sigseem to be resolved by the present ex- nificantly to either inotropic or chronoperiments. It is sometimes felt that the tropic regulation of the heart, since in no direct inhibitory influence of vagal fibers experiment could functional responses be on ventricular contractile force can be ex- elicited from isolated thoracic ganglionic plained by the accompanying reduction in segments (figs. 8, 9). This is not to say, atrial transference of blood to the ven- however, that they could not serve as aftricular chambers, since atrial contrac- ferent pathways from the heart, or that 198 W. RANDALL, J. ARMOUR, D. RANDALL AND 0. SMITH they may not exert efferent effects on metabolic and/or vasomotor functions. LITERATURE CITED Becker, F. R., and J. A. Grunt 1957 The cervical sympathetic ganglia. Anat. Rec., 127: 1-14. Burn, J. H., and M. J. Rand 1959 Sympathetic postganglionic mechanisms. Nature, 184: 163. Copen, D. L., D. R. Cirillo and M. Vassale 1968 Tachycardia following vagal stimulation. Am. J. Physiol., 215: 696-703. Ellison, J. P., and T. H. Williams 1969 Sympathetic nerve pathways to the human heart and their variations. Am. J. Anat., 124: 149162. Kuntz, A. 1933 The autonomic nervous system. In: The Anatomy of the Rhesus Monkey (Macaca mulatta). Chapt. 17. C. G. Hartman, ed. Williams and Wilkins. McKibben, J. S.,and R. Getty 1969 A study of the cardiac innervation in domestic animals: cattle. Anat. Rec., 165: 141-152. Mitchell, G . A. G. 1953 Anatomy of the autonomic nervous system. E. and s. Livingstone, Edinburgh. Mizeres, N. J. 1955 The anatomy of the autonomic nervous system in the dog Am. J. Anat., 96: 285-318. 1963 The cardiac plexus in man. Am. J. Anat., 112: 141-151. Perman, E., 1924 Anatomische Untersuchungen iiber die Herzenerven bei den hoheren Saugetieren und beim Menshen, Ztschr. f. Anat. u. Entwickl., 71: 382-457. Randall, W. C., J. B. Pace, J. S. Wechsler and K. S. Kim 1969 Cardiac responses to separate stimulation of sympathetic and parasympathetic components of the vagosympathetic trunk i n the dog, Cardiologia, 54: 104-118. Randall, W. C., and W. G. Rohse 1956 The augmentor action of the sympathetic cardiac nerves. Circulation Res., 4 : 4 7 0 4 7 5 . Riegele, L. 1926 uber die Innervation der Hals und Brustorgane bei einigen Affen, Ztschr. f . Anat. u. Entwickl., 80: 777-858.