L system innervating the human heart. Electrical stimulation

Transcription

1 Hu.man Cardiac Nerve Stimulation David A. Murphy, MD, and J. Andrew Armour, MD, PhD Departments of Surgery, Physiology, and Biophysics, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada Cardiovascular responses were elicited in 12 patients undergoing cardiac operations when cardiopulmonary neural elements between the aortic root and pulmonary artery or in the right atrial ganglionated plexus were stimulated. Heart rate and left ventricular intramyocardial sy!stolic pressure were augmented when cardiopulmonary nerves between the aorta and pulmonary artery were stimulated in 11 of the 12 patients. Right ventricular intramyocardial systolic pressure was augmented in 7 of these 1 1 patients. Cardiodepressor responses were elicited when the right atrial ganglionated plexus (9 patients) or a cardiopulmonary nerve (2 patients) was stimulated. These results demonstrate that electrical stimulation of the human extrinsic and intrinsic cardiac nervous systems can alter cardiodynamics, different responses being elicited when different neural structures are stimulated. These data are in accord with those obtained from canine experiments and suggest that the human extrinsic and intrinsic cardiac nervous system contains functionally similar neural elements to those found in other mammals. (Ann Thoruc Surg 2992;54:502-6) ittle is known about the efferent autonomic nervous L system innervating the human heart. Electrical stimulation of cardiopulmonary nerves located between the aortic root and pulmonary artery at the base of the heart induces negative or positive chronotropic changes, depending on the nerve stimulated, in humans undergoing aorta-coronary artery bypass grafting for coronary artery revascularization [l]. Although responses elicited in humans are similar to those elicited when homologous canine cardiopulmonary nerves are stimulated, including similar variations in responses, the magnitude of responses elicited in humans were considerably less than those elicited in dogs [2]. Efferent sympathetic and parasympathetic neurons can exert considerable influences on canine Ventricular dynamics [2-71. As ventricular dynamics were not monitored in the previous human investigation [l], the relative lack of responses elicited could have been due, in part, to the lack of assessment of ventricular inotropic changes. Furthermore, the discrepancy in responses elicited when homologous nerves are stimulated in humans versus dogs might have been due to the fact that the investigations in humans, but not animals, were performed after bypass and coronary artery revascularization with its attendant use of hypothermia and cardioplegia. Such procedures obtund the function of the efferent autonomic nervous system regulating the mammalian heart [8]. The purpose of the present study was to determine the effects of (electrical stimulation of the intrathoracic efferent autonomic nervous system regulating the human heart in patients with valvular or congenital heart conditions before instituting cardiopulmonary bypass. Thus variables that coulcl result in blunting of cardiac responses elicited when the intrathoracic autonomic nerves are stimulated, - Accepted for publication Jan 14, 1992 Address reprint requests to Dr Murphy, Victoria General Hospital, Suite 3067, Dickson Center, Halifax, Nova Scotia, B3H 2Y9, Canada. namely, cardioplegia and hypothermia [8], were avoided. In addition to heart rate and systemic and pulmonary vascular pressures, indices of right and left ventricular contraction were also monitored in these studies to determine if the human intrapericardial autonomic nervous system can modify ventricular contractility. Furthermore, because stimulation of loci in the right atrial ganglionated plexus in animal models can activate efferent parasympathetic neural elements and thereby regularize supraventricular tachycardias [9], the effects of right atrial neuronal stimulation was also studied. Material and Methods Surgical Preparation All experiments were conducted according to the guidelines for clinical research involving experiments at the Victoria General Hospital, Halifax, Nova Scotia, every patient having signed a consent form for the protocol. Twelve patients were entered in the study. Patients studied were having cardiac operations for aortic valve stenosis (4 patients), aortic valve insufficiency (2 patients), mitral valve stenosis (1 patient), mitral valve insufficiency (1 patient), atrial septa1 defect (3 patients), and subaortic stenosis (1 patient). All patients received benzodiazepine (midazolam hydrochloride or diazapam) and phenylpiperdine analgesia (fentanyl, sufentanil), as well as inhalation anesthesia (enflurane or isoflurane). After intubation, ventilation was initiated and maintained at an inspired oxygen fraction of 1. A lead I1 electrocardiogram was monitored, as were radial artery and pulmonary artery pressures (Swan-Ganz catheter). Miniature solid-state 5F pressure transducers (Millar Mikro-Tip model SPR 344 pressure tipped sensors; Millar Instruments, Houston, TX), with their sensor surfaces placed on the inside of angled (15 degrees for right ventricular recording and 45 degrees for left ventricular recordings) 1.5-mm-diameter needles (Fig l), were in by The Society of Thoracic Surgeons $5.00

2 Ann Thorac Surg 1992;54:502-6 MURPHY AND ARMOUR 503 Fig 1. The 45-degree angled Millar Mikro-Tip pressure transducer used to record intramyocardial pressure in the human left ventricle. serted by pushing the needle-tipped sensor into the midwall regions of the right ventricular conus as well as the midwall regions of the anterior and posterior (diaphragmatic) walls of the left ventricle. These sensors were employed to record regional intramyocardial pressures. Regional intramyocardial pressures were recorded to determine effects of the efferent autonomic nervous system on the ventricles as chamber and aortic pressures are not adequate indices of such effects [6]. All variables were recorded using an Astro-Med Inc (West Warwick, RI) model MT9500 rectilinear recorder. Cardiopulmonary nerves were identified between the root of the aorta and the pulmonary artery (Fig 2) [lo]. Before cardiopulmonary bypass was begun, each of these neural structures was isolated and stimulated in turn with 4-V, 5-ms, 10-Hz stimuli for 30 seconds using a bipolar electrode connected with a Grass SD-9 square wave stim- ulator, the output of which was monitored on a Telequipment D-54 oscilloscope. These voltages are supramaximal for stimulation of canine intrathoracic neural elements [7]. Supramaximal voltages of 4 ms duration were used to stimulate as many neural elements as possible within the ganglionated plexus. It had been determined previously that stimuli of shorter duration are not sufficient to effectively stimulate neural elements within the canine right atrial ventral ganglionated plexus [9]. Enough time occurred between stimulations to allow the recorded variables to return to basal conditions. In 10 of the patients, the center of the right atrial anterior collection of fat, which contains a ganglionated plexus, was also stimulated with the bipolar electrodes for 30 seconds. This tissue overlies the right superior and inferior pulmonary veins in the interatrial groove. Stimulations were not repeated after cardiopulmonary bypass due to the fact that the time required for repeating such stimuli might have prolonged the operation unnecessarily. Data Analyses Heart rate and systolic and diastolic arterial pressures, as well as peak systolic right and left ventricular intramyocardial pressures, were recorded for five consecutive cardiac cycles during preintervention periods as well as during peak responses elicited during stimulation. Their means _t the standard error of the mean were calculated. Because differing responses were elicited when different neural structures were stimulated, the cardiac responses elicited in each patient were grouped according to whether augmentation or depression was induced. Thus all responses in which rate and force changes were elicited were grouped according to whether the measured variables were enhanced or depressed. Data obtained before and during stimulations were compared using the Student's t test for paired data (p < 0.01). Because the placement of the Millar pressure transducer with respect to myocardial depth could not be assured to be consistent from one case to the other, percent change in intramyocardial systolic pressures was determined. Significant artifacts can occur with respect to data obtained when such a pressure transducer is introduced into the myocardium so that its sensor surface is not parallel to the endocardia1 surface of the ventricles. Therefore, care was taken to ensure that its sensor surface faced the cavity. Fig 2. Surgeon's view of the human intrapericardial cardiopulmonay nerve plexus (N) located superior to the heart between the main pulmonay artery (MPA) and the aorta (Ao). In this figure the aorta has been divided to expose this plexus. (A = ascending aorta; Ad = aortic adventitia; IV = innominate vein; RA = right atrium; RPA = right pulmona y arte y ; SVC = superior vena cava; TS = transverse sinus.) Results Cardiopulmonary Nerve Stimulation Two to five cardiopulmonary nerves were stimulated in each of the 12 patients. Heart rate was augmented in 11 of the 12 patients when at least one of the identified cardiopulmonary nerves was stimulated. Left ventricular anterior wall intramyocardial systolic pressure was augmented in each of these instances (Table 1). Left ventricular posterior wall intramyocardial pressure was augmented in four of these instances. Right ventricular intramyocardial pressure was augmented in 7 of these 11

3 504 MURPHY AND ARMOUR Ann Thorac Surg 1992;54:5024 Table 1, Augmentor or Depressor Cardiovascular Responses Elicited by Stimulation of lntrapericardial Cardiopulmonary Nerves in Humansa Intervention ~~ Systemic Pressure HR RVC IMP LV IMP Systolic Diastolic (beatdmin) (% change) (% change) (mm Hg) (mm Hg) Augmentor responses (n = 11) Prein tervention 79 f ? Stimulate 94 f 4 209? ? p Value NS NS NS NS Depressor responses (n = 10) Preintervention t Stimulate 79 f ? k 4 p Value NS NS NS NS a Augmentor or depressor cardiovascular responses were induced in humans by stimulation of intrapericardial cardiopulmonary nerves (12 patients studied), or the right atrial ventral ganglionated plexus (10 patients studied) were stimulated. Tachycardia or bradycardia responses being elicited depending on the neural structure stimulated. Tachycardia was elicited when intrapericardial cardiopulmonary nerves were stimulated. Augmentation of right and left ventricular intramyocardial and systolic systemic vascular pressures accompanied these chnages. Depressor responses were elicited when loci in the right atrial ventral ganglionated plexi were stimulated in 9 of 10 patients and when cardiopulmonary nerves were stimulated in 2 of 12 patients. When bradycardia was induced, intramyocardial systolic pressures became reduced. Values are presented as means k standard errors of the mean. HR = heart rate; LV IMP = left ventricular ventral intramyocardial systolic pressure; intramyocardial systolic pressure. NS = not significant; RVC IMP = right ventricular conus patients (Fig 3). Systemic vascular systolic pressure was augmented minimally overall, increasing in only 5 of these patients. Pulmonary artery pressure was unaffected overall, becoming elevated in only 3 cases (see Fig 3). Thus considerable variability of results were achieved when different intrapericardial nerves between the aortic root and pulmonary artery were stimulated. Heart rate and ventricular systolic intramyocardial pressures were reduced in 2 of the 12 patients when other cardiopulmonary nerves were stimulated. In 1 patient no cardiovascular responses were detected when the cardiopulmonary nerves that were visualized between the aortic root and pulmonary artery were stimulated. Data derived from this patient were not included in the final data analysis. Great variability exists with respect to the anatomic location of mediastinal cardiopulmonary nerves in humans [2]. As such nerves are frequently located behind the major vessels and isolating them to accomplish stimulation could jeopardize the operation, the function of such nerves was not investigated. Atrial Ganglionated Plexus Stimulation When the right atrial ventral ganglionated plexus was stimulated in 10 patients, bradycardia was induced in 4 of the patients. Thus no significant heart rate changes were induced overall (see Table 1, depressor responses). Right ventricular intramyocardial systolic pressure was reduced in 5 patients. Left ventricular intramyocardial systolic pressure was reduced in 9 of the 10 patients, the amount of suppression being significant overall. Systemic and pulmonary vascular pressures were unaffected overall. Fig 3. Card'iovascuh' responses elicited by electrical stimulation bar at bottom) of a mediastinal cardio- EKG pulmonary nerve in a human before undergoing a mitral valve operation. Stimulation induced tachycardia as well as augmentation of right ventricular infundzbuluni systolic pressure. Aortic pressure was AP In'[ augmented minimally. (AP = aortic pressure; EKG (mmhg) 0 = a lead I1 electrocardiogram; IMP = right ventricular infundibirlar intramyocardial pressure; PAP = pulmonary artery pressure.) PAP "1 (mmhg) 0

4 Ann Thorac Surg 1992; MURPHY AND ARMOUR 505 No augmentor responses were elicited when the right atrial ganglionated plexus was stimulated. Comment The present investigations were undertaken to determine whether cardiopulmonary nerves located at the base of the heart and neural elements in the anterior right atrial ganglionated plexus are capable of modulating the human heart and, if so, in what manner. Previously it has been demonstrated that electrical stimuli delivered to these neural elements can elicit heart rate and systemic vascular changes without causing untoward effects [l]. However, because ventricular inotropism was not monitored in that study, it was not possible to determine whether such stimulations induced cardiac inotropic effects as occurs when the homologues of these nerves are stimulated in dogs. Data derived from the present experiments indicate that not only can tachycardia be induced when dorsal mediastinal cardiopulmonary nerves are stimulated, but ventricular inotropism can be augmented (see Fig 3), as occurs in dogs [2]. Electrical stimulation of canine efferent sympathetic neurons that innervate the heart augments ventricular inotropism as indicated by a substantial augmentation in right and left ventricular intramyocardial pressure [4]. Such augmentation of canine ventricular inotropism frequently is accompanied by minor or no changes in systemic vascular pressure due to the fact that total peripheral vascular resistance is not altered significantly by such an intervention [6]. Similar results were obtained in this study where minimal changes were induced in aortic pressure despite the fact that left ventricular intramyocardial systolic pressure increased (see Table 1). When intrapericardial cardiopulmonary nerves located at the base of the heart were stimulated electrically in the present series of experiments, tachycardia was induced in 11 of the 12 humans studied. Most of these nerves are located deep in the space between the ascending aorta and the main pulmonary artery [lo], making them relatively inaccessible and thus sometimes difficult to identify in the operating room. The lack of effect found in 1 patient presumably was due to the relative difficulty in identifying cardiopulmonary nerves with sizeable numbers of axons in that patient, rather than the fact that no functional cardiopulmonary nerves were present in that patient. Furthermore, when a cardiopulmonary nerve was stimulated usually one region of the ventricles was augmented more than the others, as has been reported to occur in dogs [2]. Thus it appears that the sympathetic efferent axons in the major cardiopulmonary nerves between the aorta and pulmonary artery primarily innervate the anterior wall of the left ventricle and to a lesser extent the posterior wall of the left ventricle and the right ventricle. When neural elements in canine right atrial ganglionated plexi are stimulated electrically, afferent as well as efferent parasympathetic and sympathetic neural elements that modulate the heart are activated [9, 11-15]. Parasympathetic effects usually predominate over the sympathetic ones [9]. When the homologous neuronal plexus was stimulated in humans, bradycardia (4 of 10 patients) and negative ventricular inotropic responses (9 of 10 patients) were elicited. These data imply that parasympathetic neural elements do exist in this plexus. These results are presumed to be due to direct or indirect activation of efferent postganglionic parasympathetic neurons because in animal experiments such effects are blocked with atropine. In humans, atrial tachydysrhythmias were induced in some instances when loci in the right atrial ganglionated plexus were stimulated. This occurred particularly when loci near the edge of the fat were stimulated presumably due to current spread to adjacent atrial tissue as occurs when homologous structures are stimulated in dogs [ll]. Parasympathetic responses were also elicited (2 of 12 patients) when intrapericardial cardiopulmonary nerves between the aorta and pulmonary artery were stimulated. The reliability of using sensor-tip pressure transducers to record regional intramyocardial pressure is an important issue. Indices of intramyocardial pressure have proven of value in determining the functional innervation of ventricles by the efferent autonomic nervous system [4, 6, 7, 111. Efferent autonomic axons in a cardiopulmonary nerve [7] or a locus of the atrial plexus [3, 11, 141 usually innervate specific ventricular regions rather than all ventricular tissue. When these neural elements are stimulated left ventricular cavity systolic pressure frequently is minimally affected or not affected at all, even though force of contraction can be increased in relatively large regions of the ventricular wall [6, 7, 111. Thus animal studies have demonstrated that indices of regional contractile force, as determined using intramyocardial pressure sensors, are important when elucidating the functional innervation of the ventricles by the efferent autonomic nervous system and cannot be studied reliably by monitoring ventricular chamber pressure changes alone [4, 61. In the present experiments this was confirmed when augmentor responses were elicited as left ventricular intramyocardial pressure was increased much more than systolic arterial pressure (see Table 1). Furthermore, ventricular augmentor responses were elicited with greater consistency than heart rate changes, presumably because efferent sympathetic axons that innervate the sinoatrial node course primarily in nerves other than those that were studied [lo]. This is in contrast to the finding that stimulation of the right or left upper thoracic sympathetic trunks in humans induces tachycardia [ 161, presumably because sympathetic efferent preganglionic axons that modulate the sinoatrial node are located in these structures. It is concluded that electrical excitation of human intrapericardial neural elements can modify cardiac rate and force. Due to the relative inaccessibility of and variability of responses elicited by human cardiopulmonary nerves, it is difficult to envisage routinely using this method to modulate cardiovascular function. On the other hand, neural elements in the right interatrial groove are more readily accessible and may prove easier to manipulate. Because electrical stimulation of canine right atrial neural elements can be employed in controlling supraventricular

5 506 MURPHY AND ARMOUR Ann Thorac Surg 1992;54:502-6 tachycardias in dogs [9] and because stimulation of the human homologue of these neural elements can also induce bradycardia, manipulation of the intrinsic cardiac nervous system in humans may be of therapeutic interest in the future. This work was supported by the Medical Research Council of Canada (MT-10122) as well as the Nova Scotia and New Brunswick Heart Foundations. The technical assistance of Cheryl Forbes, RN, and Richard Livinston is gratefully acknowledged. References 1. Murphy DA, Johnstone DE, Armour JA. Preliminary observations on the effects of stimulation of cardiac nerves in man. Can J Physiol Pharmacol 1985;63: Brandys JC, Randall WC, Armour JA. Functional anatomy of the canine mediastinal cardiac nerves located at the base of the heart. Can J Physiol Pharmacol 1986;64: Ardell JL, Randall WC. Selective vagal innervation of sinoatrial and atrioventricular nodes in canine heart. Am J Physiol 1986;251:H Armoiir JA, Randall WC. Canine left ventricular intramyocardial pressure. Am J Physiol 1971;220:183>9. 5. Blomquist TM, Priola DV, Romero AM. Source of intrinsic innervation of canine ventricles: a functional study. Am J Physiol 1987;252:H63W. 6. Butler C, Wong AYK, Armour JA. Systolic pressure gradients between the wall of the left ventricle, the left ventricle chamber and the aorta during positive inotropic states implication for left ventricular efficiency. Can J Physiol Pharmacol 1988;66: 87>9. 7. Randall WC, Armour JA, Geis WP, Lippincott DB. Regional cardiac distribution of sympathetic nerves. Fed Proc 1972;31: Murphy DA, Armour JA. Influences of cardiopulmonary bypass, temperature, cardioplegia and topical hypothermia on cardiac innervation. J Thorac Cardiovasc Surg (in press). 9. Ali IM, Butler CK, Armour JA, Murphy DA. Modification of supraventricular tachyarrhythmias by stimulating atrial neurons. Ann Thorac Surg 1990;50: Janes RD, Brandys JC, Hopkins DA, Johnstone DE, Murphy DA, Armour JA. Anatomy of human extrinsic cardiac nerves and ganglia. Am J Cardiol 1986; Butler CK, Smith FM, Cardinal R, Murphy DA, Hopkins DA, Armour JA. Cardiac responses to electrical stimulation of discrete loci in canine atrial or ventricular ganglionated plexi. Am J Physiol 1990;259:H Lazzara R, Scherlag BJ, Robinson HJ, Samet P. Selective in situ parasympathetic control of the canine sinoatrial and atrioventricular nodes. Circ Res 1973;32: Randall WC, Wurster RD, Duff M, OToole MF, Wehrmacher W. Surgical interruption of postganglionic innervation of the sinoatrial nodal region. J Thorac Cardiovasc Surg 1991;lOl: Bluemel KM, Wurster RD, Randall WC, Duff MJ, OToole MF. Parasympathetic postganglionic pathways to the sinoatrial node. Am J Physiol 1990;259:H Wallick DW, Martin PJ. Separate parasympathetic control of heart rate and atrioventricular conduction in dogs. Am J Physiol 1990;259:H Randall WC, McNally H. Augmentor action of the sympathetic cardiac nerves in man. J Appl Physiol 1960;15: