Reflex Control of Sympathetic Activity During Ventricular Tachycardia in Dogs
Primary Role of Arterial Baroreflexes
Background The determinants of hemodynamic outcome during ventricular tachycardia (VT) are not well understood. In the present study, we addressed the relative contributions of arterial and cardiopulmonary baroreflexes to the sympathetic and arterial pressure responses to VT or ventricular pacing (VP) in dogs with inducible VT.
Methods and Results Responses of renal sympathetic nerve activity (RSNA), pulmonary capillary wedge pressure (PCWP), and mean arterial pressure (MAP) to induced VT or VP (220 to 280 beats per minute) were determined in 12 dogs with a healed anteroapical infarction and inducible VT and in 8 control dogs. The responses were determined with all reflexes intact, after selective denervation of either arterial or cardiopulmonary baroreflexes, and after combined denervation. Differences between intact and denervated conditions were used to assess the relative effects of each baroreflex. In the infarct group, responses during VT were comparable to those during VP. RSNA and PCWP increased significantly (P<.01), whereas MAP decreased significantly (P<.001) during VT or VP with baroreflexes intact in both groups. The increase in RSNA and the recovery of MAP during sustained VP were greater in the infarct group (P<.05); in addition, the increase in PCWP was greater in the infarct group (P<.05). Arterial baroreflex denervation abolished the increased RSNA and recovery of MAP during VP in both groups. After cardiopulmonary baroreflex denervation, the increase in RSNA was augmented in both groups (control group more than infarct group), but recovery of MAP was increased further only in the control group.
Conclusions These results suggest that arterial baroreflex-mediated sympathoexcitation plays an important role in determining the hemodynamic outcome during VT, whereas cardiopulmonary baroreflexes play only a modest modulatory role.
Sudden cardiac death is a major cause of mortality in modern society. Ventricular tachyarrhythmias often preface sudden cardiac death, but it remains unclear why some episodes of ventricular tachycardia are well tolerated by patients and often self-terminate, whereas others deteriorate to severe hypotension and culminate in ventricular fibrillation. Activation of the sympathetic nervous system normally is an important mechanism to maintain arterial pressure homeostasis. Previously, Smith et al1 showed that ventricular tachycardia or rapid ventricular pacing provokes increases in sympathetic nerve activity and that preservation of arterial pressure during the tachyarrhythmia was related directly to the magnitude of sympathoexcitation.
The mechanisms responsible for regulation of sympathetic nerve activity during ventricular arrhythmias are not well defined. Ventricular tachycardia provokes directionally opposite changes in stimuli to the two primary baroregulatory regions: increased cardiac filling pressures, which stimulate cardiopulmonary baroreceptors, and decreased arterial pressure, which unloads arterial baroreceptors.2 Halliwill et al3 recently reported that in anesthetized healthy dogs, rapid ventricular pacing (214 bpm) leads to decreases in sympathetic nerve activity that were reversed by denervation of vagal cardiopulmonary baroreceptors. These data suggested that in healthy dogs, cardiopulmonary baroreceptors impart the greatest control of sympathetic nerve activity during simulated ventricular tachycardia. Because ventricular tachycardia generally occurs predominantly in individuals with underlying myocardial disease with substrate for the arrhythmia, it is unclear whether these findings in healthy dogs are relevant to the clinical events in patients. This question of clinical relevance is particularly important because the canine sympathetic neural responses to simulated ventricular tachycardia were directionally opposite the responses observed in patients with inducible tachycardia.1 3
Heart disease can affect reflex control mechanisms4 and may result in impaired sympathetic modulation by either arterial or cardiopulmonary baroreceptors. Cardiopulmonary baroreceptor control of sympathetic nerve activity and vascular resistance is impaired in patients and animals with left ventricular dysfunction, hypertension, and congestive heart failure and after myocardial infarction.5 6 7 8 9 This impairment occurs early during development of left ventricular dysfunction when there is no impairment of arterial baroreflex function.9 Moreover, arterial baroreflex control of sympathetic activity is preserved in animal models of left ventricular dysfunction and congestive heart failure.7 9 10 11 We sought to determine the relative roles of arterial and cardiopulmonary baroreceptors in sympathetic neural and arterial pressure regulation during ventricular tachycardia or rapid ventricular pacing in healthy dogs and in dogs with healed myocardial infarction with inducible ventricular tachycardia. We hypothesized that in dogs with myocardial infarction, significant sympathoexcitation occurs during ventricular pacing or tachycardia and that this response is mediated predominantly by arterial baroreflexes.
Studies were performed in 20 adult mongrel dogs weighing 18 to 26 kg. Twelve dogs were studied 7 to 12 days after an anterior myocardial infarction was produced. Eight dogs served as controls; four underwent a sham operation (thoracotomy) for the myocardial infarction, and four were studied acutely without previous thoracotomy. Acute experimental studies were performed in which responses were recorded of RSNA, arterial pressure, and ventricular filling pressure to ventricular tachycardia or rapid ventricular pacing.
Model of Ventricular Tachycardia
Induction of an anterior myocardial infarction was performed under sterile conditions using a Harris two-stage occlusion of the anterior descending coronary artery similar to that used by Garan et al.12 Anesthesia was initiated with pentobarbital (5 mg/kg) and maintained at stage III anesthesia with isoflurane (1% to 2%). Dogs were intubated and mechanically ventilated throughout the surgery. A left lateral thoracotomy exposed the anterior wall and apex of the left ventricle. All visible epicardial collateral branches from the left circumflex or posterior descending coronary arteries were ligated. The left anterior descending coronary artery was dissected just distal to the first diagonal branch. The Harris two-stage occlusion involved a 20-minute partial occlusion produced by tying a ligature around the vessel and a 19-gauge blunt-end needle and then removing the needle. After 20 minutes, a second ligature was secured to produce a complete occlusion. After an additional 20 to 30 minutes, sterile wire electrodes (Teflon-coated stainless steel) were implanted along the peri-infarct zone and tunneled within polyethylene tubing to the subcutaneous tissue at the back of the neck. The thoracotomy was closed, and the dogs were allowed to recover for 7 to 12 days.
At the time of the acute study, the ends of the wires were harvested from the back of the neck and used to induce ventricular tachycardia or pace the ventricle. Standard electrophysiological testing was used to determine ventricular refractory periods and perform programmed stimulation for the induction of ventricular tachycardia using a programmable stimulator (Bloom Associates DTU-101).
To control for the effects of thoracotomy on reflex control, sham surgery was performed in four control dogs. This involved thoracotomy as described without dissection of myocardium or occlusion of epicardial vessels. Four additional control dogs were studied without thoracotomy.
Setup for Acute Experimental Study
The dogs were anesthetized with thiamylal (20 to 30 mg/kg IV) and α-chloralose (40 to 80 mg/kg). Supplemental chloralose (10 to 30 mg/kg) was given hourly. Dogs were intubated and mechanically ventilated with a mixture of oxygen and room air. Arterial blood gas values were obtained routinely, and appropriate adjustments of ventilation rate, volume, and gas content were made to maintain Po2 at >80 mm Hg, Pco2 at 30 to 40 mm Hg, and arterial pH at 7.35 to 7.45.
A femoral artery was dissected, and a cannula was inserted for measurement of arterial pressure. Surface electrodes were placed to record the ECG. An external jugular vein was exposed, and a Swan-Ganz catheter was inserted and advanced into the pulmonary artery for the measurement of PAP and PCWP. The pacing wires were harvested from the subcutaneous tissue of the neck and connected to a programmable stimulator. In the control dogs and in two dogs with myocardial infarction, a quadripolar pacing catheter was advanced from the jugular vein into the right ventricular apex to perform ventricular pacing.
The neck was dissected so as to identify the aortic depressor nerves bilaterally and to prepare the carotid sinuses for vascular isolation. The left cervical aortic nerve was identified, with a dissecting microscope, where it emerges from the vagosympathetic trunk caudal to the superior laryngeal nerve and nodose ganglion. The identity of the nerve was confirmed by the pulse-synchronous discharge pattern obtained from a bipolar recording electrode. A fine silk thread was looped around the nerve for later identification and retrieval. Likewise, a thread was looped around the vagal trunk (excluding the aortic depressor nerve). The left carotid sinus was prepared for isolation by ligating all vascular branches except the common carotid trunk and the external carotid artery. Reversible occluders were placed around these two vessels. A cannula was inserted into the carotid sinus region through the lingual artery and connected to a pressure transducer for assessment of carotid sinus pressure. The aortic nerve and carotid sinus dissections were repeated on the right side. Preservation of carotid sinus baroreflex function was confirmed by the slowing in heart rate produced when the carotid sinus pressure was increased.
After preparation of the neck for selective barodenervation or isolation, the dog was placed in the right lateral decubitus position and an incision was made in the left flank. With a retroperitoneal approach, the renal vessels and nerves were dissected free from the surrounding connective tissue. A branch of the renal sympathetic nerve was cut distally, and the nerve sheath was removed. The nerve was covered with mineral oil and placed on a bipolar platinum electrode for recording of efferent sympathetic nerve traffic. Nerve signals were amplified and filtered (high frequency, 1 to 3 kHz; low frequency, 30 to 100 Hz). The amplified signals were analyzed with a spike counter (Nerve Traffic Analyzer, model 706C, University of Iowa) that counted and integrated all nerve spikes above a predetermined voltage threshold (just above the noise level).
In all dogs, arterial pressure, PCWP, and RSNA were recorded continuously during a baseline period and during ventricular pacing for 1 minute at rates of 220, 250, and 280 bpm. The order of pacing rates was randomized. Programmed electrical stimulation was performed in the dogs with myocardial infarction, and data were collected in dogs in which ventricular tachycardia was inducible.
The dogs were assigned randomly to one of two groups based on the order of baroreceptor denervation. In group 1, after the control data were collected (all baroreceptors intact), a “functional” SAD was performed to denervate the arterial baroreceptors. This involved sectioning both aortic depressor nerves, vascularly isolating the carotid sinuses, and pressurizing the carotid sinuses with a static pressure near the MAP. We describe this as a functional denervation because the carotid sinus afferent nerves remained intact and were exposed to a constant pressure stimulus and thus continued to provide neural input to the brain stem. The lack of a change in heart rate or sympathetic nerve activity in response to systemic bolus infusion of nitroglycerin (100 μg) was taken to indicate complete functional denervation. After a stabilization period, the responses to ventricular pacing and tachycardia were repeated as described above. Last, the vagal cardiopulmonary baroreceptors were denervated by VGX. After a stabilization period, the responses to ventricular pacing and tachycardia were recorded. In group 2, the responses to ventricular pacing and tachycardia were first obtained in the intact control state, then after selective cardiopulmonary denervation, and finally after combined cardiopulmonary and functional SAD.
Systolic, diastolic, and mean arterial pressures; PCWP; and raw and integrated RSNAs were recorded continuously with a strip-chart recorder (Gould Instruments). All signals were analyzed with a digitizer tablet (model 2210, Numonics) and a customized computer program. Average values of arterial pressures, PCWP, and RSNA were obtained over a 1-minute baseline period before each episode of ventricular pacing or during a 20-second period before the programmed stimulation train that induced ventricular tachycardia. During ventricular pacing or tachycardia, data were analyzed over 10 seconds at the nadir of the initial fall in arterial pressure (≈10 seconds after pacing or tachycardia onset) and over 20 seconds during sustained pacing or tachycardia (≈60 seconds after pacing or tachycardia onset). Responses of RSNA are reported as percent change from baseline to the nadir or steady state period of pacing.
All data sets were tested for normality (Kolmogorov-Smirnov). A two-way ANOVA was used to compare baseline MAP and PCWP among denervation conditions and between groups. Because baseline RSNA was not normally distributed across denervation conditions, a Kruskal-Wallis ANOVA on ranks was used to compare baseline RSNA values across denervation conditions. A two-way ANOVA with repeated measures design was used to determine effects of pacing rate and time during the control state in the 11 dogs with myocardial infarction. The time effect assessed differences between baseline, onset of ventricular pacing or ventricular tachycardia (≈10 seconds), and sustained ventricular pacing or tachycardia (≈60 seconds). The specific effects of functional arterial baroreceptor denervation or cardiopulmonary baroreceptor denervation were determined in each study group as described in Table 1⇓. The effect of arterial baroreceptor denervation was assessed as the difference in responses between control condition and SAD condition in group 1 dogs and as the difference between VGX condition and VGX+SAD condition in group 2 dogs. The effect of cardiopulmonary baroreceptor denervation was assessed as the difference in responses between SAD condition and SAD+VGX condition in group 1 dogs and as the difference between control condition and VGX condition in group 2 dogs. A new variable was created for the arterial baroreceptor and cardiopulmonary baroreceptor denervation effects for each dog. A two-way ANOVA was used to determine differences between groups and across rates for each denervation effect. Because a group effect was not found for the effect of either denervation on sympathetic activity or arterial pressure, the data and variance values for all 11 dogs were pooled. A two-way ANOVA was used to determine significance for denervation effects among different pacing rates. For all statistical analyses, significance was defined at an α level of .05.
After the dogs with myocardial infarction were euthanatized, the hearts were excised and preserved in 10% formalin for later evaluation of the myocardial infarction. Myocardial infarction was confirmed by gross inspection and histological evidence of fibrosis. Infarctions were located in the anteroapical region in all dogs.
Eleven of 12 dogs with healed anteroapical myocardial infarction completed all rates of ventricular pacing during each baroreceptor denervation condition. The other dog died when induced ventricular tachycardia deteriorated to ventricular fibrillation, from which resuscitation was not achieved. Sustained ventricular tachycardia was inducible in 8 dogs with either two or three extrastimuli. However, data were not collected during ventricular tachycardia in 3 of the 8 dogs as the induced ventricular tachycardia rates were more than 350 bpm and were not tolerated hemodynamically. Specifically, MAP fell precipitously to less than 40 mm Hg and cardioversion was required within 30 seconds of induction of the rhythm in these dogs. In Fig 1⇓, sample tracings from 1 dog illustrate the responses of RSNA, PCWP, arterial pressure, and ECG during ventricular pacing and ventricular tachycardia at similar rates with all baroreceptors intact.
Response to Ventricular Tachycardia in Dogs With Myocardial Infarction
Hemodynamically tolerated ventricular tachycardia was induced in five dogs. The mean ventricular tachycardia rate was 240±12 bpm (range, 199 to 276 bpm). The responses during ventricular tachycardia were compared with the ventricular pacing rate closest to the tachycardia rate and are shown in Fig 2⇓. In animals with intact baroreceptors, the responses did not differ between ventricular tachycardia and ventricular pacing.
The responses to ventricular pacing in the 11 dogs with myocardial infarction with baroreceptors intact (control condition) are shown in Fig 3⇓. Significant decreases in MAP and increases in PCWP and RSNA occurred at the onset (10 seconds) of pacing regardless of pacing rate. As pacing was sustained for 1 minute, MAP and PCWP increased significantly from the nadir observed at the onset of pacing, whereas RSNA did not increase further (P>.21). No significant differences in the responses of MAP were found across pacing rates (P=.47). However, a rate effect was observed for RSNA (P=.02) and for PCWP (P=.05), with significant differences occurring between 220 and 280 bpm.
Baseline arterial pressure, PCWP, and RSNA during each baroreceptor denervation condition are summarized in Table 2⇓. No differences between infarct and control dogs were observed (P>.05). Sympathetic nerve activity after combined SAD and VGX was elevated above the intact state (P<.05) but did not differ between any other denervation conditions.
The responses to ventricular pacing and the effects of arterial and cardiopulmonary denervation did not differ between control dogs with or without thoracotomy. The responses to ventricular pacing at 250 bpm and the denervation effects for the two control groups are summarized in Table 3⇓. Because no differences between the two control groups were found, the data from all control dogs were pooled for comparisons with the dogs with myocardial infarction. The responses to ventricular pacing with reflexes intact are summarized for the control and myocardial infarction groups (Table 4⇓). The increase in RSNA tended to be greater in the infarction group at all pacing rates (main effect, P<.05), with a significant difference noted at 280 bpm. This augmented sympathoexcitatory response was accompanied by a modest but significant recovery of MAP (increase in MAP from onset of pacing to sustained pacing) during sustained ventricular pacing. The increase in PCWP during pacing was greater in the infarction group (P<.05).
Baroreceptor Denervation Effects
The effects of baroreceptor denervation in the two groups of dogs with myocardial infarction are summarized in Figs 4⇓ and 5⇓. In group 1 (Fig 4⇓), arterial baroreceptor denervation eliminated the sympathoexcitatory response and recovery of arterial pressure (difference between 10 and 60 seconds) during sustained ventricular pacing. This contrasted with group 2 (Fig 5⇓), in which cardiopulmonary baroreceptor denervation resulted in slightly augmented sympathoexcitation but with no improvement in the arterial pressure response. The changes in PCWP were comparable during all conditions. In both groups, combined arterial baroreceptor and cardiopulmonary baroreceptor denervation resulted in sustained decreases in arterial pressure during pacing with no significant change in sympathetic nerve activity.
The specific effects of either denervation were further assessed by the change in response from the condition with the specific receptors intact to the denervated condition. Details of the comparisons for either arterial baroreceptor or cardiopulmonary baroreceptor effects are summarized in Table 1⇑. The arterial baroreceptor effect was estimated as the difference between control condition and SAD condition in group 1 and as the difference between VGX alone and combined VGX plus SAD in group 2. The cardiopulmonary baroreceptor effect was estimated as the difference between control condition and VGX condition in group 2 and as the difference between SAD alone and combined SAD plus VGX in group 1. The effects of selective arterial baroreceptor denervation or cardiopulmonary baroreceptor denervation during pacing for the myocardial infarction and control groups are summarized in Figs 6⇓ and 7⇓. Positive values indicate augmented responses after denervation during pacing or ventricular tachycardia, whereas negative values indicate attenuated responses. Arterial baroreceptor denervation abolished the sympathoexcitatory response to ventricular pacing in all dogs, resulting in a net negative value for the arterial baroreceptor effect (Fig 6⇓; range, −19% to −34%). This was accompanied by a significant reduction in the MAP recovery during sustained ventricular pacing (range, −11 to −20 mm Hg). No differences were observed between infarct and control groups (P>.20). In the infarct group, cardiopulmonary baroreceptor denervation resulted in a modest augmentation in the sympathoexcitatory responses to ventricular pacing, ranging from 4% to 15% improvement (Fig 7⇓). This was accompanied by an insignificant increase in the MAP response (P>.10). Cardiopulmonary baroreceptor denervation in the control group resulted in a greater increase in RSNA and significant MAP recovery (range, 5 to 9 mm Hg) during sustained pacing (P<.05). For both denervation effects, there was a significant rate effect on RSNA responses (P<.05), with the greatest effects at the fastest pacing rate. There was no significant rate effect on responses of MAP or PCWP.
Arterial baroreceptor and cardiopulmonary baroreceptor denervation effects also were compared between ventricular tachycardia and ventricular pacing. The change in MAP was significantly reduced by arterial baroreceptor denervation (P<.001) but did not differ between ventricular tachycardia and pacing (−17±4 versus −21±4 mm Hg, respectively). Similarly, the change in RSNA was significantly reduced by arterial baroreceptor denervation (P<.01) but did not differ between ventricular tachycardia and pacing (−27±8% versus −24±6%, respectively). The change in MAP was not affected by cardiopulmonary baroreceptor denervation (5±4 versus 3±6 mm Hg for tachycardia and pacing, respectively). The change in RSNA was augmented after cardiopulmonary baroreceptor denervation (P<.05) but did not differ between ventricular tachycardia and pacing (14±6% versus 10±5%, respectively).
Ventricular tachycardia or ventricular pacing provokes rapid decreases in arterial pressure and increases in cardiac filling pressures in humans with inducible ventricular tachycardia.1 2 When tachycardia or pacing is sustained, arterial pressure recovers toward baseline. The extent of this recovery of arterial pressure is dependent on the magnitude of sympathoexcitation.1 In the present study, we demonstrated that in animals with recent myocardial infarction and inducible ventricular tachycardia, arterial baroreflexes play the principal role in determining the sympathoexcitatory response. Cardiopulmonary baroreflexes limit this sympathoexcitatory response; however, this modulation does not affect arterial pressure recovery importantly. In contrast, control dogs with or without previous thoracotomy demonstrated limited sympathoexcitation that was modulated by a balance of arterial and cardiopulmonary baroreceptor inputs and resulted in limited recovery of arterial pressure. These data demonstrate that compared with cardiopulmonary baroreflexes, arterial baroreflexes play the dominant role in controlling sympathetic nerve activity and supporting arterial pressure during ventricular tachycardia in dogs with previous myocardial infarction and inducible ventricular tachycardia.
Hemodynamic Regulation During Ventricular Dysrhythmias
At the onset of ventricular tachycardia or pacing in patients with inducible ventricular tachycardia, arterial pressure decreases precipitously and usually is accompanied by increases in sympathetic nerve activity.1 As the dysrhythmia is sustained for >1 minute, sympathetic nerve activity remains elevated, although it may decrease from the peak level achieved at the onset of the dysrhythmia and is usually accompanied by some degree of recovery of arterial pressure.1 We demonstrate similar qualitative changes in arterial pressure and RSNA in dogs with a healed anteroapical infarction, although the magnitudes of sympathoexcitation and arterial pressure recovery were less than those reported in humans.1 The difference in sympathetic responses and arterial pressure recovery between these studies may be due in part to the effects of anesthesia in the present study. Anesthesia is known to blunt baroreflex gain.13 We chose to use α-chloralose because these attenuating effects are thought to be minimal14 15 ; however, modest attenuation of baroreflex function also occurs with α-chloralose.16 In addition, baseline RSNA would be expected to be much higher in the anesthetized, surgically traumatized animals of the present study; thus, percentage changes would be expected to be less. Differences between this and the study in humans1 are also attributable to differences in ventricular rate. Average rates of ventricular tachycardia in the study in humans were 189±8 bpm (versus 240±12 bpm in the present study) and ventricular pacing was performed at rates ranging from 100 to 150 bpm (versus 220 to 280 bpm in the present study). The more rapid rates of ventricular tachycardia and pacing in this study result in very brief ventricular filling times. This shortened filling time results in significant reduction of end-diastolic dimensions17 18 and probably contributes to impaired stroke volume and cardiac output. This is supported indirectly by the data of Smith et al,1 in which the extent of arterial pressure recovery during pacing at 150 bpm was greater than that during pacing at 182 bpm (30 versus 24 mm Hg), despite much greater sympathoexcitation at the faster pacing rate (126% versus 69%). We found a similar rate effect in which arterial pressure responses to ventricular pacing at rates between 220 and 280 bpm did not differ, whereas sympathoexcitation was progressively greater with increasing ventricular rates. In dogs in which ventricular tachycardia was induced at rates of more than 330 bpm, arterial pressure did not recover and often continued to decrease as the dysrhythmia was sustained. These data taken together with those from humans1 suggest that the sympathoexcitatory response increases with increasing rates but that the functional effectiveness of this response (net support of arterial pressure) becomes progressively less at faster rates until at very fast rates reflex-mediated increases in vascular resistance are ineffective in increasing arterial pressure back toward baseline.
Role of Baroreceptors
Single ectopic ventricular beats are accompanied by transient increases in sympathetic nerve activity in humans.19 20 21 The magnitude of sympathoexcitation appears to be inversely related to the transient decrease in arterial pressure. This implies that arterial baroreceptors play a primary role in this response. This conclusion is supported by the findings of Herre and Thames,22 who demonstrated in dogs that the sympathoexcitatory response to single or double ectopic beats was prevented by SAD. During sustained ventricular tachycardia, Smith et al23 reported that the sympathoexcitatory response was directly related to estimated arterial baroreflex gain. This sympathoexcitatory response is accompanied by forearm vasoconstriction that is blocked by the α-adrenergic antagonist phentolamine.24 Feldman et al18 found that ventricular pacing at 240 bpm in healthy dogs was accompanied by significant recovery of arterial pressure that was prevented by α-adrenergic blockade (terazosin) or ganglionic blockade (hexamethonium). All of these results point to a baroreflex-mediated sympathoexcitation that contributes importantly to the support of arterial pressure.
We used selective baroreceptor denervation to determine the relative influence of arterial and vagal cardiopulmonary baroreflexes in regulating sympathetic nerve activity and arterial pressure during ventricular tachycardia. In dogs with a healed myocardial infarction and inducible ventricular tachycardia, significant sympathoexcitation occurred during ventricular tachycardia or pacing and was eliminated by arterial baroreceptor denervation. This effect had significant hemodynamic consequences as the arterial pressure recovery during sustained tachycardia or pacing was abolished by arterial baroreceptor denervation. Vagal cardiopulmonary denervation led to greater increases in sympathetic nerve activity, although it had minimal effects on arterial pressure recovery. These data suggest that in the presence of myocardial infarction, the arterial baroreflexes play the primary role in regulating sympathetic nerve activity and that cardiopulmonary baroreflexes play a less important role.
Halliwill et al3 reported that RSNA is decreased at the onset of ventricular pacing and remains decreased when pacing is sustained for 1 minute. They concluded that in healthy dogs, cardiopulmonary baroreceptors play a dominant role in modulating sympathetic nerve activity. We found that healthy control dogs with or without healed thoracotomy had very modest increases in sympathetic nerve activity and recovery of arterial pressure. The relative roles of arterial and cardiopulmonary baroreceptors on control of sympathetic nerve activity were approximately balanced (see Figs 6⇑ and 7⇑). Thus, these data support the conclusion that cardiopulmonary baroreceptors play an important role in modulating sympathetic responses during ventricular pacing, but the magnitude of this role was somewhat less than was reported by Halliwill et al.3 Several factors may account for these differences. First, the thoracotomy or dissection of the pericardium could have caused partial disruption of vagal cardiopulmonary afferent fibers. This could attenuate the modulating influences of cardiopulmonary baroreceptors. We studied two groups of control dogs with and without healed thoracotomy. The sympathetic and arterial pressure responses to graded ventricular pacing were similar in both control groups; therefore, it is unlikely that the thoracotomy adversely affected cardiopulmonary baroreceptor function. Second, we used ventricular pacing rates ranging from 220 to 280 bpm compared with the pacing rate of 214 bpm used by Halliwill et al. The sympathoexcitation and arterial pressure recovery were generally similar at all three rates. Thus, it is unlikely that a slightly slower pacing rate would alter the net reflex and hemodynamic responses. Third, we used a functional SAD in which the carotid sinus was isolated and pressurized. This results in a functional denervation with a constant afferent input to brain-stem cardiovascular centers. Complete SAD is often accompanied by high sympathetic activity and labile arterial pressure. Moreover, Halliwill et al studied dogs with an open-chest protocol, whereas our studies were conducted in closed-chest animals. An open-chest preparation presents greater stress during the experimental procedure and potentially higher baseline sympathetic nerve activity. Elevated baseline sympathetic nerve activity favors inhibitory effects (versus excitatory effects) and may account for the more dominant cardiopulmonary baroreflex influence during rapid ventricular pacing in control dogs suggested by Halliwill et al. Although the specific reason or reasons for the differences between the responses of the healthy control dogs in the two respective studies remain unclear, it is apparent from our study that the responses in healthy dogs are different than those of dogs with a healed myocardial infarction. These differences are discussed below.
Effects of Healed Myocardial Infarction
In the intact state, the sympathoexcitatory response to ventricular pacing was greater in dogs with healed myocardial infarction than in healthy control dogs. Likewise, the recovery of arterial pressure during sustained pacing was greater in the infarct dogs. These differences were attributable to greater sympathoexcitatory effects of arterial baroreflexes than the sympathoinhibitory effects of cardiopulmonary baroreflexes and suggest that cardiopulmonary baroreflex control of sympathetic activity was impaired in the myocardial infarction group. This is supported by two findings. First, the cardiopulmonary denervation “effect” was less in the myocardial infarction group, showing that denervation of cardiopulmonary baroreceptors has less influence on the net sympathoexcitatory response to ventricular tachycardia or pacing. Second, the magnitude of sympathoexcitation in the intact state was greater in the myocardial infarction group despite a greater increase in filling pressure (stimulus for cardiopulmonary baroreceptors) and no difference in the change in arterial pressure (stimulus for arterial baroreceptors).
Impairment of cardiopulmonary baroreceptor function occurs early in the development of pacing-induced left ventricular dysfunction in dogs9 and soon after myocardial infarction in humans.5 8 In contrast, arterial baroreceptor control of sympathetic activity is preserved well into the development of left ventricular disease, including congestive heart failure.7 10 11 The differences between our dogs with myocardial infarction and control dogs strongly suggest that cardiopulmonary baroreceptor modulation of sympathetic nerve activity during ventricular tachycardia or pacing is impaired after an anterior wall infarction. Two mechanisms may account for this effect. First, myocardial infarction can damage or disrupt either the mechanoreceptors or afferent nerves that traverse the infarct zone.25 Minisi and Thames26 demonstrated that posterior wall infarction significantly impairs sympathoinhibitory responses to volume infusion, whereas anterior wall infarction has much less of an effect on these cardiopulmonary baroreceptor-mediated responses. The infarction in our dogs may have produced a modest denervation of these receptors and contributed to the partial impairment. Second, the myocardial infarction can alter ventricular function and potentially alter the stimulus to mechanoreceptors within the walls. This impairment may be exacerbated during nonsinus rhythms such as ventricular tachycardia and apical ventricular pacing. Although there is no direct evidence that altered ventricular mechanics adversely affect cardiopulmonary baroreceptor function, left ventricular dysfunction due to myocardial infarction or chronic rapid pacing is accompanied by significant impairment of cardiopulmonary baroreceptor function.9 It follows that altered ventricular mechanics may contribute to the impaired responses to increased filling pressures that occur during ventricular tachycardia.
The clinical outcome of ventricular arrhythmias is variable, ranging from benign to fatal. The mechanisms responsible for this variability are numerous, yet not fully understood. Ventricular rate plays an important role in the hemodynamic outcome of ventricular tachycardia.1 27 28 The impact of rate on arterial pressure is greatest at very fast rates, yet rate is not the only determinant of hemodynamic outcome. Our study further supports a growing body of evidence that sympathoexcitation plays a pivotal role in the hemodynamic outcome of ventricular tachycardia.1 3 18 24 This study is the first to clearly demonstrate the importance of arterial baroreceptor-mediated sympathoexcitation in supporting arterial pressure recovery during sustained ventricular tachycardia. The arterial pressure recovery in these dogs was modest compared with that observed in humans1 and is probably attributable to hemodynamic limitations imposed by the rapid rates used in this study. Nevertheless, even modest support of arterial pressure to prevent hemodynamic collapse is probably beneficial. Moreover, it is likely that at faster rates, mean arterial pressure falls below the effective range of coronary autoregulation so that coronary perfusion becomes pressure dependent. Under these conditions, even modest increases in arterial pressure may be critical for maintaining coronary perfusion and thereby preventing ischemia and deterioration of the arrhythmia to polymorphic ventricular tachycardia and ventricular fibrillation.
Although our data point to an impairment in cardiopulmonary baroreflex function that may be beneficial in the net sympathoexcitatory response to ventricular tachycardia, we did not directly assess cardiopulmonary baroreflex gain under control conditions. As noted, several studies have demonstrated that cardiopulmonary baroreceptor control of vascular resistance is impaired in humans or dogs with left ventricular dysfunction, including after myocardial infarction.5 8 9 Therefore, it is likely that these dogs demonstrated some degree of cardiopulmonary baroreflex impairment that contributed to the altered sympathetic neural and arterial pressure responses to ventricular tachycardia. Because the role of pressure versus chamber volume in activating cardiopulmonary baroreceptors is unclear, altered chamber compliance in the infarct group could affect the net stimulus to these receptors during ventricular tachycardia or pacing. Although this could affect our assessment of cardiopulmonary baroreflex function, it does not alter the principal conclusion of these data that arterial baroreceptors play the predominant role in determining sympathetic neural and hemodynamic responses.
In conclusion, we demonstrated that during ventricular tachycardia or pacing, dogs with healed anteroapical myocardial infarction exhibit significant sympathoexcitation that contributes to recovery of arterial pressure as the dysrhythmia is sustained. This sympathoexcitatory response is mediated primarily by deactivation of arterial baroreceptors and is modulated by activation of cardiopulmonary baroreceptors. The magnitudes of sympathoexcitation and arterial pressure recovery are greater than those observed in healthy dogs. We speculate that these differences are due primarily to impaired cardiopulmonary baroreflex control of sympathetic nerve activity after myocardial infarction.
Selected Abbreviations and Acronyms
|bpm||=||beats per minute|
|MAP||=||mean arterial pressure|
|PAP||=||pulmonary artery pressure|
|PCWP||=||pulmonary capillary wedge pressure|
|RSNA||=||renal sympathetic nerve activity|
This study was supported in part by grants from the American Heart Association, Northeast Ohio Affiliate, and from the National Heart, Lung, and Blood Institute (HL-30506) and by funds from the Department of Veterans Affairs Medical Research Service. We thank Jan Malycky, LATG; Gayle Rising, LATG; and Will Austin, BA, for their technical assistance.
- Received July 31, 1995.
- Revision received October 2, 1995.
- Accepted October 4, 1995.
- Copyright © 1996 by American Heart Association
Smith ML, Thames MD. Cardiac receptors: discharge characteristics and reflex effects. In: Armour JA, Ardell JL, eds. Neurocardiology. Oxford, UK: Oxford University Press; 1994:19-52.
Thames MD, Johnson LN. Impaired cardiopulmonary baroreflex control of renal nerves in renal hypertension. Circ Res. 1985;57:741-747.
Ferguson DW, Abboud FM, Mark AL. Selective impairment of baroreflex-mediated vasoconstrictor responses in patients with ventricular dysfunction. Circulation. 1984;69:451-460.
Kinugawa T, Dibner-Dunlap ME, Thames MD. Cardiopulmonary baroreflex control of renal sympathetic nerve activity is blunted in left ventricular dysfunction without heart failure. Circulation. 1992;86(suppl I):I-778. Abstract.
Dibner-Dunlap ME, Thames MD. Baroreflex control of renal sympathetic nerve activity is preserved in heart failure despite reduced arterial baroreceptor sensitivity. Circ Res. 1989;65:1526-1535.
Garan H, Fallon JT, Ruskin JN. Sustained ventricular tachycardia in recent canine myocardial infarction. Circulation. 1980;62:980-987.
Price HL. General anesthetic and circulatory homeostasis. Physiol Rev. 1960;40:187-218.
Talman WL, Snyder D, Reis DJ. Chronic lability of arterial pressure produced by destruction of A2 catecholaminergic neurons in rat brainstem. Circ Res. 1980;46:842-853.
Lima JAC, Weiss JL, Guzman PA, Weisfeldt ML, Reid PR, Traill TA. Incomplete filling and incoordinate contraction as mechanisms of hypotension during ventricular tachycardia in man. Circulation. 1983;68:928-938.
Welch WJ, Smith ML, Rea RF, Bauernfeind RA, Eckberg DL. Enhancement of sympathetic nerve activity by single premature ventricular beats in humans. J Am Coll Cardiol. 1989;130:69-75.
Smith ML, Carlson MD, Thames MD. Reflex control of the heart and circulation: implications for cardiovascular electrophysiology. J Cardiovasc Electrophysiol. 1992;2:441-449.
Barber MJ, Mueller TM, Davies BG, Gill RM, Zipes DP. Interruption of sympathetic and vagal-mediated afferent responses by transmural myocardial infarction. Circulation. 1985;72:623-631.
Minisi AJ, Thames MD. Effect of chronic myocardial infarction on vagal cardiopulmonary baroreflex. Circ Res. 1989;65:396-405.