Neurally Mediated Negative Inotropic Effect Impairs Myocardial Function During Cholinergic Coronary Vasoconstriction in Pigs
Background Myocardial dysfunction elicited during cholinergic coronary vasospasm is generally ascribed to myocardial ischemia. However, when pericoronary nerves extend to the ventricles, efferent vagal discharges that elicit coronary vasoconstriction may, at the same time, depress myocardial contractility by releasing acetylcholine into the myocardium. We analyzed whether a neurally mediated effect contributes to impairment of ventricular function during cholinergic coronary vasoconstriction.
Methods and Results Left ventricular (LV) function, ECG mapping, and coronary angiography were analyzed during pericoronary application of methacholine (MCh) to the left anterior descending coronary artery in 53 chloralose-anesthetized open-chest pigs. MCh induced local coronary vasoconstriction and depressed coronary blood flow (13.3±3.7 to 6.0±4.6 mL/min [54%]; ANOVA, P<.001), systolic LV pressure (105±10 to 91±10 mm Hg [13%], P<.001), peak LV (+)dP/dt (2890±524 to 2270±447 mm Hg/s [21%], P<.001), peak LV (−)dP/dt (1446±484 to 1048±276 mm Hg/s [27%], P<.001), and regional systolic segment shortening (0.32±0.09% to 0.18±0.14% [43%], P=.02) and caused local ST-segment elevation (0.9±0.8 to 8.4±3.9 mV, P<.001). These changes were not preceded by heart rate variations, were reproducible, and were inhibited by atropine. MCh during perfusion of nitroglycerin (2 μg·kg−1·min−1 IV) continued to depress LV pressure (9%, P=.002), LV (+)dP/dt (16%, P<.01), LV (−)dP/dt (20%, P<.01), and segment shortening (18%, P=.03) even though coronary blood flow drop was markedly attenuated (7% versus 54%; ANOVA, P<.01). Disruption of pericoronary nerves with phenol attenuated MCh-induced LV pressure fall (P<.05). Histological data show that cholinergic and adrenergic pericoronary nerves extend into the ventricular myocardium.
Conclusions A neurally mediated effect, in addition to ischemia, impairs LV function during cholinergic coronary constriction in a model with pericoronary nerves extending into the ventricles.
The parasympathetic nervous system is thought to play a role in the genesis of coronary vasoconstriction in humans and in other species.1 2 3 This effect is exerted via local neural release of acetylcholine from the cholinergic pericoronary nerves that innervate the adventitia of the epicardial coronary arteries. Extraluminally liberated acetylcholine can constrict the coronary artery by acting on muscarinic receptors of smooth muscle cells, as seen in isolated rings of human and porcine coronary arteries.4 5 6 7
In species with cholinergic pericoronary nerves extending to the ventricles, efferent vagal discharges that elicit adventitial neural release of acetylcholine might, at the same time, induce release of this ester into the myocardial interstitium. Because acetylcholine depresses myocardial inotropism,8 9 it is possible that the impairment of left ventricular (LV) function during cholinergic coronary spasm10 is due not only to myocardial hypoperfusion but also to a vagally mediated negative inotropic effect.
In this study, we analyzed whether a neurally mediated mechanism is in part responsible for the LV dysfunction induced by cholinergic coronary constriction in pigs.
Fifty-three adult pigs of either sex weighing 25 to 30 kg underwent a midsternotomy under general anesthesia with metomidate (4 mg/kg IV) followed by α-chloralose (100 mg/kg IV). Pulmonary ventilation was maintained with a pressure respirator. The pericardium was opened, and its free margins were sutured to the borders of the sternotomy to cradle the heart. The left anterior descending coronary artery (LAD) was dissected 20 to 25 mm below its origin for later application of methacholine. A second dissection 10 to 15 mm distal to the former was made to place a blood flow probe.
LV pressure was sampled with percutaneous catheters introduced through the right femoral artery. In 12 pigs in which we studied regional myocardial function by sonomicrometry, the LV pressure was measured with precalibrated Millar SPC-350 pressure transducers (Millar Instruments) and recorded in a thermal-array Nihon Kohden RTA 1200 polygraph. In these animals, the LV dP/dt was derived from the LV pressure signal with a derivative computer, Nihon Kohden ED 601G. In the remaining cases, the LV pressure and LV dP/dt were measured with saline-filled pigtail 7F catheters by use of a Hewlett-Packard 7754B polygraph and a 8814A derivative computer.
Arterial blood gases were measured at regular intervals and were kept within normal limits by adjustment of the respirator characteristics. Intravenous isotonic saline was administered to compensate for blood losses. Pigs were handled in accordance with the position of the American Heart Association on research animal use. This study was approved by the Ethics Committee of our institution.
Cholinergic coronary stimulation
A 5-mm2 piece of gauze soaked in a 1% solution of methacholine chloride (Sigma-Aldrich) was applied for 15 seconds to the adventitia of the proximal segment of the LAD. In a subset of pigs, we also tested the effects of 0.5% methacholine. Care was taken to avoid spreading methacholine over the epicardium because, as seen in dogs,11 epicardial cholinergic afferents can be chemically stimulated, and this leads to a reflex decrease of heart rate and blood pressure.
Epicardial ECG mapping
DC electrograms over the anterior LV surface were recorded in 35 of the 53 pigs with 32 electrodes sutured to a rubber membrane arranged in three parallel columns with a 5-mm interelectrode distance.12 The membrane was aligned parallel to the LAD and sutured to the epicardium with a Prolene 5-0 snare. Electrodes covered an area extending above and below the site of methacholine application. The signals were recorded with a 32-channel differential amplifier system, and samples of 2 seconds' duration were digitized at 500 Hz and stored in a computer. Selected analog signals were recorded with a seven-channel Elema ink-jet polygraph. Electrodes were thin cotton threads soaked with isotonic saline solution connected to the amplifiers through a silver-silver chloride interface. A 0-mV potential reference was obtained at the mediastinal fat. ST-segment data were expressed as total TQ+ST-segment displacement because this corresponds to the ST-segment shift recorded by the conventional ECG.
Coronary blood flow
Changes in coronary blood flow (systolic, diastolic, and mean) were measured in all pigs by an electromagnetic system (Skalar MDL 1401). Probes of 1.5- or 2.0-mm ID were placed around the LAD 5 to 10 mm distal to methacholine application. Care was taken to prevent extrinsic coronary compression by the flowmeter, and the system was calibrated before each methacholine application.
Regional myocardial function
In 12 of the 53 pigs, two pairs of hemispherical polystyrene divergent lenses molded onto 1- to 2-mm, 6-MHz piezoelectric crystals were inserted into the inner third of the myocardium to measure regional systolic shortening with a sonomicrometer (Triton Technology Inc). The crystals of each pair were positioned 1 to 2 cm apart parallel to the short axis of the LV. One pair was inserted at the LV apex, and a control pair was placed at the lateral region perfused by the left circumflex coronary artery. End-diastolic length (EDL) and end-systolic length (ESL) were identified on the recordings, and systolic segment (SS) shortening was calculated from the equation SS=(EDL−ESL)/EDL. End-diastolic measurements were taken at the point at which positive dP/dt begins to rise, and the end-systolic dimensions were taken 20 ms before the nadir of the negative dP/dt.13 Maximal length of the crystal signals was also measured. To correct for possible variations in the distance between each pair of crystals, we normalized to a 10-mm initial segment length.13
In 6 of the 53 pigs, we obtained angiograms of the left coronary trunk at baseline, during methacholine-induced ST-segment elevation, and after ST-segment recovery. A SIAS SM9 HF x-ray image intensifier was positioned to obtain a left anterior oblique projection of the LAD. An Amplatz 5F catheter was introduced into the left coronary ostium, and the angiographic dye, a nonionic contrast medium (iopamidol 76%), was selectively injected. Angiograms were stored on videotape with a Sony videocassette recorder (SVO-9500MDP).
Protocol and Study Population
Hemodynamic and ECG variables were recorded at baseline and continuously after application of methacholine until drug effects recovered. In addition, samples of 2 seconds' duration of the 32 epicardial signals were stored in a computer every 5 seconds during the first 5 minutes of methacholine exposure and every 60 seconds thereafter.
The study included the following subsets of experiments.
Series 1 (n=12)
These pigs were used to assess whether a nonischemic mechanism impairs LV function during methacholine-induced coronary constriction. We compared the hemodynamic effects of methacholine alone with those elicited when coronary constriction is prevented by simultaneous administration of nitroglycerin (2 μg·kg−1·min−1 IV). Perfusion of nitroglycerin was started 60 minutes before methacholine to achieve a steady hemodynamic state.
Series 2 (n=8)
In this group, we compared the drop in LV pressure induced during the first 60 seconds of methacholine with that elicited during a similar period of LAD ligature at the level of previous methacholine exposure.
Series 3 (n=7)
This series served to assess whether interruption of pericoronary neural transmission with phenol influences the hemodynamic depression elicited by methacholine. Phenol (carbolic acid, 88%) was applied 30 minutes before methacholine at the adventitia of the LAD in a region just below the zone of methacholine exposure. We chose this time interval because other authors11 have verified interruption of neural transmission 30 minutes after application of phenol on canine epicardium. Two weeks after pericoronary application of phenol, we found apical myocardial sympathetic and cholinergic denervation in swine.14
Series 4 (n=14)
These pigs underwent two applications of 1% methacholine spaced 30 minutes apart to assess the reproducibility of the drug effects.
Series 5 (n=6)
This group was used to analyze the effects of atropine (0.04 mg/kg IV) administered 2 minutes before methacholine.
Series 6 (n=6)
In this group, we assessed the coronary vascular effects of methacholine by coronary angiography.
Hemodynamic and ECG changes elicited after each treatment were evaluated by ANOVA for repeated measures with commercially available software (SYSTAT Inc). Because changes in hemodynamic parameters evolved rapidly after application of methacholine, we evaluated the early changes with samples taken at baseline and 15, 30, 45, and 60 seconds after methacholine, whereas the late changes were assessed at baseline and 2, 3, 4, and 5 minutes after drug exposure. A significant probability value (P<.05) indicates that both the early and late changes of that variable are significant. If the probability values of these two phase changes are different, we report the one with lower significance. Data are expressed as mean±SD.
Anatomic analysis was performed to verify that pericoronary nerves extend into the myocardium in pigs. This would be demonstrated if pericoronary application of phenol destroyed the pericoronary nerves and, at the same time, caused denervation of the distal myocardium. Twelve additional pigs underwent an aseptic left lateral thoracotomy at the level of the fifth intercostal space under intravenous anesthesia with thiopental sodium (30 mg/kg IV) and mechanical ventilation. A portion of the rib was removed, and the pericardium was opened. In 6 of the 12 pigs, we dissected the LAD at its origin to apply phenol around the vessel, avoiding spreading this drug over the epicardium. In the remaining 6 pigs, the LAD was not dissected and phenol was not applied. The thorax was closed, and pleural air was aspirated. The animals were allowed to recover and received analgesia and antibiotic therapy. Two weeks later, the hearts were removed under barbiturate anesthesia, and we obtained transverse block sections of the LAD and right coronary artery as well as transmural myocardial samples of the anteroseptal and left lateral walls. All samples were immediately frozen with liquid nitrogen, and we studied the adrenergic and cholinergic innervation by adrenergic histofluorescence15 and acetylcholinesterase reaction,16 respectively. Specimens processed for adrenergic histofluorescence were cut into 20- to 30-μm sections in a cryostat at −20°C and were immersed in glyoxylic acid (glyoxylic acid [Sigma], 2 g; glucose, 5.4 g; sodium phosphate, 5.5 g; distilled water, 100 mL; 10N sodium hydroxide to titrate the solution to pH 7.4; and distilled water to reach a final volume of 150 mL) for 15 seconds. Thereafter, preparations were consecutively incubated at 37°C for 45 minutes and at 85°C for 10 minutes. They were immediately observed in a Leitz fluorescent microscope with a K 490 filter.
Acetylcholinesterase activity was analyzed by the “direct-coloring” copper ferrocyanide method.16 Samples were preincubated for 15 minutes in ice-cold 0.1 mol/L phosphate buffer (pH 7.0) in the presence of 10% formaldehyde and incubated at 37°C for 1.5 to 3.5 hours with acetylthiocholine iodide as the substrate. Sympathetic and parasympathetic fibers were studied consecutively in the same preparation17 and in similar microscopic fields. In each field, myocardial fiber density relative to samples from normal tissue was qualitatively determined at ×100 magnification by two separate observers.
As depicted in Figs 1 and 2⇓⇓, application of methacholine to the adventitia in 12 pigs (series 1) induced a progressive drop in coronary blood flow at the LAD associated with a parallel loss in apical systolic segment shortening and with a biphasic fall in LV pressure and in LV dP/dt. The first LV pressure fall occurred 30 seconds after application of methacholine, and the second developed 90 to 120 seconds thereafter. Maximal changes were attained 30 to 80 seconds after exposure to methacholine. As shown in the Table⇓, mean coronary blood flow fell by 54% (ANOVA, P<.001), systolic LV pressure by 13% (P<.001), peak LV (+)dP/dt by 21% (P<.001), peak LV (−)dP/dt by 27% (P<.001), and systolic segment shortening by 43% (P=.002) in the apical regions and by 11% in neighboring basal sites. The second fall in LV pressure and LV dP/dt coincided with the maximal decline in coronary blood flow and with the occurrence of holosystolic expansion in the apical ultrasonic signals (Fig 3⇓). The hemodynamic changes induced by methacholine were not preceded by heart rate variations, and only a slight (6%) increase in heart rate occurred 30 seconds after exposure to the drug (P=.01). Coronary blood flow recovered 3 to 4 minutes after methacholine, giving rise to a hyperemic reaction (Figs 2⇓, left, and 3). Five minutes after methacholine, systolic segment shortening recovered by 100%, LV pressure by 93%, and LV dP/dt by 94%. Methacholine applied during intravenous perfusion of nitroglycerin in the same pigs of series 1 (Fig 2⇓, right, and Table) continued to elicit a drop in LV pressure (9%, P=.002), a fall in peak LV (+)dP/dt (16%, P<.01) and peak LV (−)dP/dt (20%, P<.01), and a loss in apical systolic segment shortening (18%, P=.03), although the drop in coronary blood flow was almost prevented (7% versus 54%; ANOVA, P<.01). In contrast to the greater loss in systolic segment shortening during methacholine alone than during methacholine with nitroglycerin (43% versus 18%, P<.01), differences in LV pressure (13% versus 9%; ANOVA, P<.05), LV (+)dP/dt (21% versus 16%, P<.01), and LV (−)dP/dt (27% versus 20%, P<.01) were less marked.
Data from series 2 (Fig 4⇓, left) show that the drop in LV pressure during the first 60 seconds of methacholine is more marked (P<.001) than that elicited during a similar period of acute LAD ligature at the level of previous methacholine application in the same pigs.
Acute disruption of pericoronary nerves below the site of methacholine exposure (series 3) attenuated the drop in LV pressure induced by methacholine (Fig 4⇑, right, P<.05). The protective effect of phenol is not caused by a concurrent attenuation of coronary vasoconstriction, since the drop in coronary blood flow induced by methacholine without phenol treatment (10.8 to 5.7 mL/min) was comparable to that induced after phenol application (10.2 to 4.0 mL/min).
Two applications of methacholine spaced 30 minutes apart (series 4) induced a comparable decline in coronary blood flow (13.2±4.8 to 4.5±3.4 mL/min after the first test and 12.9±5.8 to 3.8±3.4 mL/min after the second test) and a similar LV pressure drop (122±18 to 99±17 mm Hg after the first application of methacholine and 120±18 to 96±14 mm Hg after the second application). In 6 pigs in which we compared the effects of 1% and 0.5% methacholine, the higher dose induced a greater drop in coronary blood flow (64% versus 36%) but a comparable fall in LV pressure (14% versus 11%) and in LV dP/dt (22% versus 18%).
Muscarinic receptor blockade with atropine (series 5) inhibited the hemodynamic effects of methacholine (Fig 5⇓).
Immediately (10 to 20 seconds) after application of methacholine to the LAD, epicardial electrodes located distal to the site of methacholine exposure showed marked changes in amplitude and polarity of the T wave (Fig 6⇓). Repolarization abnormalities occurred before coronary blood flow dropped more than 12%. ST-segment elevation began 40 to 60 seconds after methacholine (Fig 6⇓) and was maximal in apical electrodes (0.9±0.8 mV at baseline to 8.4±3.9 mV, P<.001). Atropine prevented all ECG changes (Fig 5⇑).
Coronary angiography performed at baseline and during methacholine-induced ST-segment elevation (series 6) revealed constriction of the coronary artery at the site of methacholine application. In 3 of the 6 pigs, we observed a total interruption of coronary flow (Fig 7⇓). Angiograms taken when the ST segment returned to normal levels depicted disappearance of coronary constriction.
The proximal segments of the LAD and right coronary artery in the 6 control pigs showed a variable number of cholinergic (Fig 8⇓, top) and adrenergic plexuses that coursed parallel to the vessel and innervated its adventitial region without penetrating beyond the adventitial-medial border. In contrast, all 6 pigs subjected 2 weeks previously to an application of phenol at the origin of the LAD exhibited a marked loss of cholinergic (Fig 8⇓, bottom) and adrenergic fibers in coronary segments apical to phenol application. In contrast, the nontreated right coronary artery showed a normal innervation. Arterial samples at the site of phenol application showed a nonspecific perivascular inflammatory reaction with preservation of the arterial wall structures. Samples from the myocardium supplied by the phenol-treated LAD showed transmural sympathetic and parasympathetic denervation. In contrast, this region was abundantly innervated in control pigs (Fig 8⇓, top), indicating that autonomic pericoronary nerves extend into the ventricles in swine.
Present data support the concept that a neurally mediated negative inotropic effect, in addition to myocardial hypoperfusion, impairs LV function during cholinergic coronary vasoconstriction. In this model with pericoronary nerves extending to the ventricles, application of methacholine to the adventitia elicited local coronary vasoconstriction and, at the same time, depressed myocardial function via the cholinergic pericoronary nerves.
The drop in LV pressure and LV dP/dt and the loss in regional systolic shortening observed after application of methacholine should be ascribed primarily to myocardial ischemia, because these mechanical alterations were accompanied by significant coronary hypoperfusion and by typical features of acute myocardial ischemia, such as holosystolic expansion in the ultrasonic signals13 and local ST-segment elevation. However, a nonischemic mechanism should also be mentioned to explain the depressed LV function during cholinergic coronary vasoconstriction, because when coronary hypoperfusion induced by methacholine was prevented by simultaneous intravenous perfusion of nitroglycerin in a subset of pigs, they continued to show a significant, although less marked, drop in LV pressure and LV dP/dt as well as a reduction in regional systolic segment shortening. Moreover, in agreement with the concept of a nonischemic LV dysfunction, we observed that the fall in LV pressure 60 seconds after methacholine administration is greater than that induced 60 seconds after acute LAD ligature at the level of exposure to methacholine in the same pigs.
Although the nature of the nonischemic mechanism leading to LV dysfunction during application of methacholine is unknown, our study suggests that this drug altered LV function via a neurally mediated effect, because acute disruption of pericoronary nerves with phenol attenuated the drop in LV pressure induced by methacholine. This neurally mediated LV dysfunction is probably exerted via the cholinergic pericoronary nerves, in view of the fact that cholinergic receptor blockade with atropine inhibits LV dysfunction induced by methacholine. It may be reasoned that stimulation of cholinergic pericoronary nerves by methacholine would lead to neural release of acetylcholine into the myocardium and this, in turn, would elicit LV dysfunction, because acetylcholine is known to depress myocardial contractility.8 9 18 In fact, the decline in peak LV (+)dP/dt and in peak LV (−)dP/dt induced by methacholine in nitroglycerin-treated pigs is in agreement with the concept of a vagally mediated action, because vagal stimulation depresses both LV contraction (dP/dtmax) and LV relaxation (dP/dtmin) in dogs submitted to fixed ventricular pacing and constant arterial pressure.19 Compared with the drop in LV pressure and LV dP/dt, the loss in systolic segment shortening was markedly greater during methacholine alone than during methacholine in the presence of nitroglycerin, suggesting that acute myocardial hypoperfusion affects regional function to a greater extent than LV pressure and LV dP/dt. The less marked loss in segment shortening during methacholine in the presence of nitroglycerin than during methacholine alone cannot be attributed to an inappropriate position of the ultrasonic crystals, because both tests were done in the same pigs with no reinsertion of the crystals. Conversely, the possibility that nitroglycerin blunted the loss in segment shortening during an increased vagal drive is speculative, because the effect of vagal stimulation on systolic segment shortening measured by sonomicrometry in nonischemic hearts is not well known.
It is unlikely that the neurally mediated negative inotropic effect of methacholine resulted from stimulation of afferent cholinergic fibers, since stimulation of cholinergic epicardial afferents in dogs11 and in 5 pigs (unpublished data) elicited hypotension preceded by slowing of heart rate and, in 3 pigs, salivary gland hypersecretion. In the present study, we never observed slowing of heart rate or salivary gland hypersecretion during application of methacholine to the adventitia.
Regional T-wave changes elicited immediately after methacholine can be explained on the basis of intramyocardial release of acetylcholine, since this ester is known to alter transmembrane action potential repolarization.20 21
Considerations Concerning the Model
Coronary angiography confirmed that application of methacholine to the adventitia causes coronary vasoconstriction at the site of drug exposure. The vasoconstrictor effect of cholinergic agents in swine has been recognized previously,7 22 23 24 25 and because of its inhibition by atropine, coronary vasoconstriction is thought to be mediated by muscarinic receptors of medial smooth muscle cells.24 25 Although nonadrenergic noncholinergic pericoronary nerve fibers are thought to regulate the basal coronary blood flow in rats,26 it is unlikely that these fibers mediate the vasoconstrictor effect of methacholine because porcine potassium-precontracted coronary artery rings vasodilate under exposure to capsaicin, a peptidergic neurotoxin of nonadrenergic noncholinergic fibers.27
The present model allows us to assess in vivo coronary artery reactivity at preselected vascular sites and thus may be relevant, since the coronary tree does not react uniformly to different pharmacological agents28 29 30 because of its heterogeneous receptor distribution.31
Extraluminal methacholine did not cause bradycardia or hypotension preceding the coronary blood flow changes, suggesting that coronary vasoconstriction was not initiated by a reflex sympathetic nerve stimulation in this model. Moreover, the extraluminal route avoided direct stimulation of endothelial receptors that may promote vasodilation linked to the production of endothelium-derived relaxing factor.32 33 The coronary vascular effects of cholinergic agents depend on the number of medial constrictor receptors, which vary among animal species,3 and on the ability of the ester to diffuse through the media and contact endothelial relaxing receptors.34 Acetylcholine elicits coronary dilation in dogs,5 35 36 constriction in pigs,7 22 23 37 38 and dose-dependent constriction in baboons.39 Conversely, although the integrity of the endothelium determines the response to systemic acetylcholine40 and to sympathetic stimulation41 in humans, the endothelial tissue plays no role in cholinergic spasm in pigs.7 Nevertheless, we purposely avoided direct manipulation of the endothelium in our model.
The ability of phenol to disrupt pericoronary neural fibers was confirmed 2 weeks after drug exposure in 6 pigs. In canine models, disruption of neural transmission was achieved 30 minutes after application of phenol11 or 5 minutes after infiltration with lidocaine.42
This study provides new insight into the relevance of pericoronary nerves in clinical pathophysiology. In addition to the recognized role of vagal afferent pericoronary fibers in mediating reflex hypotension and bradycardia during coronary artery occlusion,42 our data suggest that vagal efferents in the pericoronary nerves mediate a negative inotropic effect during cholinergic coronary vasospasm. This mechanism, in addition to ischemia, would be responsible for the early drop in LV pressure and LV dP/dt recorded in patients with vasospastic angina.10 Hypothetically, stretching of pericoronary nerves during coronary angioplasty could also elicit efferent cholinergic discharges, which could be in part responsible for the rapid drop in LV dP/dt observed after catheter balloon inflation.43
Extrapolation of our model to clinical pathophysiology is based on three circumstances: (1) Extraluminal application of methacholine mimics the accumulation in the adventitia of acetylcholine released from the vagal pericoronary nerves; (2) extraluminal methacholine induces coronary vasoconstriction in swine and in humans4 5 6 ; and (3) pericoronary nerves extend toward the myocardium in both species.44 45
This work was supported by grants 92/0734 and 93/0450 from Fondo de Investigaciones de la Seguridad Social, Madrid, Spain. We are grateful for the technical support provided by Dr Inocencio Anivarro during coronary angiography, Dr Ramo´n Bosch during anatomic examination, and Dr Josep Rode´s during some of the experiments. We also appreciate the statistical advice given by Dr Lluis Armadans.
- Received January 11, 1996.
- Revision received February 21, 1996.
- Accepted March 4, 1996.
- Copyright © 1996 by American Heart Association
Yasue H, Horio Y, Nakamura N, Fujii H, Imoto N, Sonoda R, Kugiyama K, Obata K, Morikami Y, Kimura T. Induction of coronary artery spasm by acetylcholine in patients with variant angina: possible role of the parasympathetic nervous system in the pathogenesis of coronary artery spasm. Circulation. 1986;5:955-963.
Kalsner S. Cholinergic constriction in the general circulation and its role in coronary artery spasm. Circ Res. 1989;65:237-257.
De Geest H, Levy MN, Zieske H, Lipman I. Depression of ventricular contractility by stimulation of the vagus nerves. Circ Res. 1965;17:222-235.
Randall WC, Wechsler JB, Pace JB, Szentivanyi M. Alterations in myocardial contractility during stimulation of the cardiac nerves. Am J Physiol. 1968;214:1205-1212.
Guazzi M, Polese C, Fiorentini F, Magrini F, Bartorelli C. Left ventricular performance and related hemodynamic changes in Prinzmetal's variant angina pectoris. Br Heart J. 1971;33:84-94.
Barber MJ, Mueller TM, Davies BG, Zipes DP. Phenol topically applied to canine left ventricular epicardium interrupts sympathetic but not vagal afferents. Circ Res. 1984;55:532-544.
Cinca J, Bardaji´ A, Carren˜o A, Mont L, Bosch R, Soldevilla A, Tapias A, Soler-Soler J. ST segment elevation at the surface of a healed transmural myocardial infarction in pigs: conditions for passive transmission from the ischemic peri-infarction zone. Circulation. 1995;91:1552-1559.
Theroux P, Ross J, Franklin D, Kemper WS, Sasayama S. Regional myocardial function in the conscious dog during acute coronary occlusion and responses to morphine, propranolol, nitroglycerin, and lidocaine. Circulation. 1976;53:302-314.
Cinca J, Bardaji´ A, Figueras J, Salas-Caudevilla A, Serrano A, Rius J. Effects of regional denervation on epicardial DC electrograms during coronary occlusion in pigs. Am J Physiol. 1987;253:H138-H146.
De la Torre JC, Spurgeon JW. A methodological approach to rapid and sensitive monoamine histofluorescence using a modified glyoxylic acid technique: the SPG method. Histochemistry. 1976;28:81-91.
Karnovsky MJ, Roots LA. A ‘direct-coloring’ thiocholine method for cholinesterases. J Histochem Cytochem. 1964;12:219-221.
Martin PJ, Levy JR, Wexberg S, Levy MN. Phasic effects of repetitive vagal stimulation on atrial contraction. Circ Res. 1983;52:657-663.
Henning RJ, Cheng J, Levy MN. Vagal stimulation decreases rate of left ventricular relaxation. Am J Physiol. 1989;256:H428-H433.
Gadsby DC, Wit AL, Cranefield PF. The effects of acetylcholine on the electrical activity of canine cardiac Purkinje fibers. Circ Res. 1978;43:29-35.
Beny JL, Brunet PC, Huggel H. Effect of mechanical stimulation, substance P and vasoactive intestinal polypeptide on the electrical and mechanical activities of circular smooth muscles from pig coronary arteries contracted with acetylcholine: role of endothelium. Pharmacology. 1986;33:61-68.
Yamada S, Yamazawa T, Nakayama K. Direct binding and functional studies on muscarinic cholinoceptors in porcine coronary artery. J Pharmacol Exp Ther. 1990;252:765-769.
Yaoita H, Sato E, Kawaguchi M, Saito T, Maehara K, Maruyama Y. Nonadrenergic noncholinergic nerves regulate basal coronary flow via release of capsaicin-sensitive neuropeptides in the rat heart. Circ Res. 1994;75:780-788.
Brazenor RM, Angus JA. Ergonovine contracts isolated canine coronary arteries by a serotonergic mechanism: no role for alpha adrenoceptors. J Pharmacol Exp Ther. 1981;218:530-536.
Broten TP, Miyashiro JK, Moncada S, Feigl EO. Role of endothelium-derived relaxing factor in parasympathetic coronary vasodilation. Am J Physiol. 1992;31:H1579-H1584.
Feigl EO. Parasympathetic control of coronary blood flow in dogs. Circ Res. 1969;25:509-519.
Van Winkle DM, Feigl EO. Acetylcholine causes coronary vasodilation in dogs and baboons. Circ Res. 1989;65:1580-1593.
Cowan CL, McKenzie JE. Cholinergic regulation of resting coronary blood flow in domestic swine. Am J Physiol. 1990;28:H109-H115.
Sakai K, Akima M, Shiraki Y, Hoshimo E. Effects of a new antianginal agent, nicorandil, on the cardiovascular system of the miniature pig: with special reference to coronary vasospasm. J Pharmacol Exp Ther. 1983;27:220-228.
Knight DR, Shen YT, Young MA, Vatner SF. Acetylcholine-induced coronary vasoconstriction and vasodilation in tranquilized baboons. Circ Res. 1991;69:706-713.
Suehiro S, Ninomiya I. Pericoronary nerve mediates inhibitory sympathetic response to coronary occlusion. Am J Physiol. 1982;243:H170-H174.
Serruys PW, Wins W, van den Brand M, Meij S, Slager C, Schuurbiers LCH, Hugenholtz PG, Brower RW. Left ventricular performance, regional blood flow, wall motion, and lactate metabolism during transluminal angioplasty. Circulation. 1984;70:25-36.