Search

“ If the heart trembles, has little power and sinks, the disease is advancing … and death is near … .” —The Papyrus Ebers (circa 3500 BCE) Numerous concepts of cardiac electrophysiology have been advanced, enthroned, and then laid to rest. Other theories have withstood the test of time despite the restless energy of inquisitive doubt of future generations. Until recently, one such concept has been the foundation of the mechanisms of fibrillation, which was imprinted in the very name to emphasize its fibrillar or myogenic nature. It was apparently known to ancient Egyptians and Chinese that an irregular heartbeat is associated with death. However, scientifically rigorous description of a causal relationship was presented only in the middle of the 19th century. Erichsen described in 1842 that coronary artery ligation led to “tumultuous,” “tremulous,” and “irregular” behavior of the ventricles.1 First documentation of the onset of ventricular fibrillation (VF) during electrical stimulation was recorded in 1849 using Ludwig’s “kymographion” by his associate Hoffa.2 Interestingly, at the time, Hoffa was assigned to investigate autonomic nervous system effects on cardiac activity, which had been discovered a year earlier by Ludwig himself.3 Hoffa described irregular contractions induced by “faradization” (electrical stimulation), which persisted even after the termination of electrical stimulation and resulted in cardiac arrest that could not be checked by vagal stimulation. Intensive investigation of the newly described phenomenon led to the introduction of numerous terms, which aimed to capture the mechanistic and/or anatomic nature of the irregular contractions and resulting cardiac arrest.4 The main disagreement gravitated toward one question: is the phenomenon neurogenic or myogenic in nature? In other words, is irregular activity due to abnormal …

The vagus nerves of anesthetized dogs were stimulated once each cardiac cycle with a brief burst of pulses, and the timing of the stimulus bursts was changed by a fixed increment on successive cardiac cycles. The effects of such vagal stimulation on atrial contraction depended on the number of pulses per stimulus burst, on the interval between pulses within the burst, and on the timing of the stimulus bursts within the cardiac cycle. The vagal stimulus bursts had the least negative inotropic effect when they were given less than 100 msec before the next atrial depolarization, and they were most effective when given about 330 msec before the next atrial depolarization. This dependence of the inotropic response on the timing of the vagal activity within the cardiac cycle indicates that the following conditions must prevail with respect to the vagal innervation of the atrium: (1) the acetylcholine released from the vagal nerve endings is hydrolyzed at a critically rapid rate in the atrial tissues, (2) the neurally released acetylcholine must exert its major influence on atrial contraction during some preferential fraction of the cardiac cycle (presumably, during depolarization), and (3) after a vagal stimulus of a given strength, the concentration of acetylcholine in the region of the myocardial cells will attain its maximum value during this critical phase of the cardiac cycle when the vagal stimuli are given at the optimal time in the cardiac cycle.

We determined the effects of the timing of repetitive bursts of vagal stimulation on the positive chronotropic responses of the heart to trains of cardiac sympathetic nerve stimulation in open-chest anesthetized dogs. Trains of sympathetic stimulation alone, at frequencies of 2 and 4 Hz, decreased the cardiac cycle length by 176 +/- 19 msec (mean +/- SE) and 190 +/- 22 msec, respectively. When bursts of vagal stimuli were given once each cardiac cycle and they were placed at their least effective time in the cycle, sympathetic stimulation at frequencies of 2 and 4 Hz decreased cardiac cycle length by only 107 +/- 8 and 120 +/- 8 msec, respectively. However, when the bursts of vagal stimuli were delivered at their most effective time in each cycle, the same levels of sympathetic stimulation elicited much larger reductions in cardiac cycle length (285 +/- 32 and 330 +/- 32 msec, respectively). Therefore, the effects of sympathetic stimulation were significantly attenuated by the vagal stimuli when the vagal bursts were relatively ineffective. Conversely, the chronotropic effects of the sympathetic stimulation were exaggerated substantially when the vagal stimulus bursts were initially positioned at their most effective time in the cardiac cycle. This latter response is contrary to the characteristic "accentuated antagonism," wherein the effect of any given level of sympathetic stimulation is diminished as the level of vagal activity is increased. This vagally mediated enhancement of the positive chronotropic response to sympathetic stimulation occurs because the phase dependency of the response of the automatic cells to the bursts of vagal stimulation is altered by the increased sympathetic activity.

The role of vagal activity, potassium concentration, and temperature in overdrive suppression of ventricular pacemakers, was studied in anesthetized dogs with and without surgical atrioventricular block. The results were as follows. (1) Overdrive of an idioventricular pacemaker during vagal stimulation was followed by a second pause. (2) Hypothermia increased the duration of the vagal and overdrive pauses. (3) Coronary sinus plasma potassium concentration ([K] cs ) decreased with the onset of vagal stimulation and increased with overdrive during vagal stimulation. (4) The increase in [K] cs during ventricular overdrive was greater in hypothermia than it was in normothermia when the same absolute rate of overdrive was used and was unchanged when the same relative rate of overdrive was used. (5) Hypothermia increased overdrive suppression more than can be accounted for by the decrease in spontaneous rate, (6) By predriving the ventricles and then overdriving them, the increase in [K] cs was dissociated from the pause duration. It is concluded that during vagal stimulation ventricular arrest is due to the inhibition of idioventricular pacemakers by the sinus node by virtue of its faster rate and that changes in [K] cs are due to rate changes and not to acetylcholine release. Overdrive suppression can be dissociated from an increase in extracellular potassium concentration. Its prolongation in hypothermia provides evidence for a metabolism-dependent mechanism.

A supernormal phase of conduction between right and left atria was observed in the exposed hearts of anesthetized dogs. Interatrial conduction time of premature atrial responses was reduced by as much as 17% during the early phase of diastole, relative to the conduction interval of basic driven responses. The supernormal phase lasted from 60 to 140 msec and was greater at slower basic driving frequencies. A brief phase of supernormal excitability was also found in the specialized cells of Bachmann's bundle from puppy hearts in vitro. Supernormal conduction was not observed within the atrial appendages (i.e., in areas not supplied by specialized conducting bands); supernormal excitability was not demonstrable in cells of ordinary atrial myocardium. Although vagal stimulation abbreviated transatrial conduction time, atropine did not abolish the phase of supernormality.

The site of origin and the sequence of atrial activation were studied in 21 open-chested dogs anesthetized with sodium pentobarbital. Bipolar plunge electrodes were used to record atrial electrograms simultaneously with surface lead II and His bundle electrogram. In sinus rhythm the sinus node was activated first, followed sequentially by Bachmann's bundle, right atrial appendix, left atrial appendix, posterior left atrium, and the proximal portion of the coronary sinus. In 15 dogs left atrial beats and rhythms of various rate were recorded, including instances of left atrial exit block and of rapid left atrial rhythms. In such beats one of the left atrial electrodes was activated first and was followed by Bachmann's bundle, sinus node, proximal portion of the coronary sinus, and by right atrial appendix. Left atrial beats were observed occurring (1) spontaneously; (2) as a result of vagal stimulation; (3) after destruction of the sinus node; (4) during ventricular pacing. Left atrial pacing through the electrode recording initial activity reproduced the sequence and configuration of atrial electrogram similar to that seen in the ectopic left atrial beats. These results demonstrate that in dogs ectopic activity frequently originates in the left atrium.

Animals pretreated with appropriate doses of anticholinesterase agents develop auricular flutter and fibrillation following injection of acetylcholine or vagal stimulation. Depression of normal sinoauricular nodal function facilitates induction of fibrillation. During fibrillation, additional injections of acetylcholine, or vagal stimulation, increase the rate of fibrillation. Atropine in small doses reverts the fibrillation to normal sinus rhythm.

A set of equations is described which permit prediction of the time course of heart rate in a dog anesthetized with pentobarbital from the time course of frequency of stimuli applied to the distal cut ends of vagus and sympathetic efferent nerves to the heart. These equations, although based on assumptions as to the physiological mechanisms involved, were only tested for their ability to describe the kinetic relationship involved. The time course of heart rate resulting from combined vagus and sympathetic nerve stimulation (HR SV ) is given by the equation: HR SV = HR V + (HR S - HR 0 ) (HR V - HR min )/(HR 0 - HR min ), where the subscripted terms are heart rate values due to vagus stimulation (v), sympathetic stimulation (s), no stimulation (0) and the minimum (min) heart rate achievable by vagus stimulation alone. Clearly the vagus influence is capable of dominating the sympathetic in the control of heart rate.

The effects of right vagus nerve stimulation on heart rate were examined in 16 dogs anesthetized with chloralose (60 mg/kg, iv) and urethane (600 mg/kg, iv) after a bilateral cervical vagotomy. Two protocols were followed. In the first, trains of 1, 3, or 5 stimuli, 2 msec each in duration, 10 to 15 v in magnitude, and separated by 10 msec were delivered at predetermined times after "P" wave onset as a single stimulation. In the second, the stimulation was carried out repetitively for 2 minutes at a predetermined frequency. Typically, in the first, the later in the cycle the stimulation occurred, the greater was the prolongation of the included cycle. With one stimulus in the stimulation, a typical P-P interval was 574 msec with a "P" to stimulation delay of 4 msec, and 620 msec when the "P" to stimulation delay was 200 msec. With five stimuli moving the stimulation through a range of 200 msec, the change in P-P interval was 150 msec. There was a functional dependence of the P-P cycle length on the time of arrival of the inhibitory pulse. The slope of this relationship is a function of the number of stimuli in the stimulation. With repetitive stimulation in the second protocol, there were zones wherein the heart rate synchronized to the vagal stimulation rate not only at the fundamental frequency but also at harmonics and subharmonics. This range of entrainment was also dependent upon the number of stimuli in the stimulation. The data suggest the possibility of phase shift as a mechanism for vagal regulation of heart rate.

The main ventricular conducting fascicles were slightly injured in anesthetized dogs by gently scratching them with a blunt needle introduced through the ventricular wall. Initially the bundle branch block that resulted was rate independent (stage 1). When conduction returned to normal, both premature atrial and vagal stimulation reproduced the bundle branch block (stage 2). During stage 2 (5-15 minutes), three conduction ranges were documented: an early (phase 3) block range, a late (phase 4) block range, and an intermediate normal conduction range. The normal conduction range was narrow at the beginning but widened progressively, mostly at the expense of the phase 4 block range. Another period during which only phase 4 bundle branch block occurred (stage 3) preceded total normalization (stage 4). The escape beats that arose from the injured fascicle were most abundant and had the shortest coupling during stage 1, they were less common and their coupling became longer during stages 2 and 3, and they disappeared in stage 4. The tachycardia-dependent or phase 3 bundle branch block was related to a prolongation of refractoriness; the bradycardia-dependent or phase 4 bundle branch block was attributed to slight hypopolarization, enhanced spontaneous diastolic depolarization, and a shift in the threshold potential toward zero. These abnormalities were assumed to be secondary to the hypopolarization, which was probably the basic derangement. This hypothesis satisfactorily accounts for the observation that phase 3 and phase 4 bundle branch block commonly coexist in the same injured fascicle.

The dynamic interaction of simultaneously changing heart period and single vagal stimuli on atrioventricular conduction (AV interval) was quantified by subtracting the vagally induced responses of the paced heart preparation from those of the unpaced heart preparation. This difference was significantly greater (P less than 0.05) than the AV responses to changes in heart period (delta A) alone without vagal stimulation (using the identical delta AA recorded from the unpaced heart in the same preparation, but with a crushed sinoatrial node). That is, for a given increase in AA interval, the AV conduction time was considerably less when the change in AA was associated with increased vagal activity than in the absence of any vagal activity. Data from some dogs in which a complete AV block was produced for both paced and unpaced hearts suggested that one mechanism of the paradoxical response was located in AV nodal tissue. Data from other dogs, in which two surface atrial recording sites were used, indicated that shifts of pacemaker site and atrial activation patterns also are an important mechanism of the paradoxical response. The relative contribution of these two mechanisms is not fixed, but can vary considerably from animal to animal.

The chronotropic responses to repetitive bursts of vagal stimulation were determined in open-chest, anesthetized dogs. Either 5 or 10 electrical pulses were included in each stimulus burst, and the interpulse interval (deltat) was varied over the range of 5 to 150 msec. As the frequency of the stimulus bursts was progressively changed, the sinoatrial (SA) nodal pacemaker cells became synchronized with the repetitive bursts of stimuli over a certain range of burst frequencies. The magnitude of this frequency range varied with deltat. For 5 and 10 pulses/burst, the values of deltat that produced the greatest magnitude of this frequency range were 30.2 and 24.3 msec, respectively. Also, over the range of values of deltat from 5 to 50 msec, the magnitude of the negative chronotropic effect of the vagal stimulus burst varied directly with deltat. It is likely that, as the interpulse interval is increased within the range of values, either more acetylcholine is released from the vagal nerve endings per pulse or there is less saturation of the receptors on the pacemaker cell membranes during each burst.

Episodes of tachycardia induced by strong vagal stimulation in spontaneously beating isolated atria of frog (Rana temporaria) were studied with multielectrode mapping technique. These episodes were inducible in 19 of 39 preparations. The arrhythmia started several seconds after cessation of vagal stimulation strong enough to cause sinus arrest, without electrical stimulation of the myocardium. The arrhythmia consisted of two to 20 beats (6 +/- 4, mean +/- SD, n = 42) with a cycle length of 100-500 msec. Recording from 32 sites with spatial resolution of 1-2 mm showed that the arrhythmia was due to intra-atrial circus movement. The estimated perimeter of the reentrant circuit ranged from 6 to 20 mm. In circuits of the minimal size, the average conduction velocity along the circuit was as low as 2-3 cm/sec. Paroxysms of the tachycardia were always preceded by vagally induced nonuniform depression of conduction, with some areas of atria being completely blocked. As the vagal influence decreased, the blocked areas recovered in an inhomogeneous manner, their unblocking being significantly (p less than 0.05) delayed after inhibition of tissue cholinesterase by proserine. The reentrant tachycardia was initiated when a sinus impulse arrived during certain phase of the unblocking. Unlike the well-known mechanism of reentrant excitation, which is based on inhomogeneous refractoriness and critically timed extrabeat(s), the circus movement in our model depended on vagally induced conduction block and could be launched by a single sinus impulse.

Efferent vagal stimulation in the anesthetized dog with constant ventricular rate caused either no change in coronary sinus outflow and cardiac work, or a decrease in coronary sinus outflow associated with a greater decrease in cardiac work. When the decreases were observed they could best be explained by a negative inotropic effect of vagal stimulation on ventricular muscle, rather than by a direct vasomotor effect of vagal stimulation on the coronary bed. Critical review of apparent divergent results previously reported regarding the effect of vagal stimulation on coronary flow reveals that they are not inconsistent with the above hypothesis.

In anesthetized dogs the vagus nerve was stimulated for 2 min; during the first minute the ventricles were driven at a rate higher than the control sinus rate; after discontinuation of the drive, the duration of asystole was prolonged. When the ventricles were driven at a rate lower than the sinus rate during vagal stimulation the subsequent asystole was shortened. Slowing sinus activity by graded vagal stimulation before maximal vagal stimulation led to a shorter asystole. Driving the ventricles at a rate higher than the sinus node rate before vagal stimulation resulted in longer asystole. In animals with chronic atrioventricular block, "overdriving" the ventricles resulted in subsequent temporary inhibition of ventricular pacemakers. In dogs with atrioventricular block, coronary sinus plasma potassium increased during the period of ventricular overdriving, and the magnitude of the rise was a function of the driving rate. These results support the concept that ventricular asystole results from the suppressive action of the fast rate imposed by the sinus node upon the slowly discharging ventricular pacemakers. Suppression of sinus node activity by the vagus reveals the rate-dependent inhibition of ventricular pacemakers. The mechanism of inhibition may be related to changes in ionic concentration gradients.

In chloroform-anesthetized dogs receiving epinephrine, slowing of the heart rate invariably accompanies conversion of arrhythmia to normal supraventricular rhythm by vagal stimulation, and restoring the faster heart rate reestablishes arrhythmia. Increasing the heart rate during epinephrine infusion can induce arrhythmia or increase the proportion of ventricular ectopic beats. These observations are taken to indicate that the vagal influence on ventricular arrhythmia in these conditions is indirect and mediated through the frequency of excitation entering the ventricle. It is suggested that during chloroform anesthesia an increased heart rate favors the emergence of ventricular arrhythmia when activity propagated after successively shorter intervals encounters delay or local conduction block in the Purkinje system, and re-entry excitation is made possible. It is proposed that in these conditions ventricular tachycardia is a self-sustaining arrhythmia that must maintain a critical frequency, either intrinsically or from external (supraventricular) input. The arrhythmia is susceptible to termination by vagal stimulation when atrial slowing decreases the probability of an atrial impulse entering the ventricle at a critical time.

The duration of the monophasic action potential (APD), the interval between atrial beats and the delay time between excitation of the right and left atria of dogs were measured, using coaxial suction electrodes, before and during vagal stimulation. The effect of vagal stimulation on the degree of shortening of APD varied at different recording sites in the atrium, the shortening being more pronounced in the right than in the left atrium. Threshold and rate of recovery after cessation of stimulation varied throughout the atrium. The results indicate that the vagal effect is nonuniformly distributed over the atrium. The time difference between excitation of the right and left atria increased with vagal stimulation under the present experimental condition, suggesting that a reduction of synchrony of excitation occurred in the atrial myocardial fibers as a whole. The same conclusion was derived from analysis of the pattern of the P wave before and during vagal stimulation.

In the anesthetized, open-chest dog, electrical stimulation of the cardiac ends of the transected cervical vagus nerves produced effects on heart rate and A-V transmission that were dependent upon the P-St interval, i.e., the-time from the beginning of the P wave to the beginning of the stimulus. When one vagal stimulus was delivered per cardiac cycle, the pacemaker response curve (curve of the P-P interval as a function of the P-St interval) was sinusoidal in configuration, with a maximum at a P-St interval of 135 msec and a minimum at 349 msec. The mean P-P interval was 568 msec, and the mean amplitude of the pacemaker response curve was 58 msec. In any given experiment, the range of P-P intervals encompassed by the pacemaker response curve defined a range of frequencies over which S-A nodal rhythm became synchronized with the activity in the vagus nerves. Over such a range, increases in the frequency of vagal stimulation evoked paradoxical increases in heart rate. On either side of this range of synchronization, vagal stimulation elicited pronounced rhythmic oscillations of P-P and P-R intervals. The frequency of such oscillations was equal to the difference between the vagal stimulation frequency and the mean heart rate. The oscillations in P-R interval were approximately 180° out of phase with the oscillations in P-P interval.

The effects of reserpine and guanethidine on reflex venoconstriction were studied in a group of open-chest anesthetized dogs by the major vessel occlusion technique described by Bartelstone. Guanethidine (10 mg./Kg. intravenously) blocked reflex venous constriction following carotid occlusion and central vagal stimulation within, five minutes after injection in five dogs. Reserpine (0.5 mg./Kg. intravenously) required 50 to 110 minutes to abolish these venopressor reflexes in five animals. Pretreatment with either reserpine (0.5 mg./Kg. intraperitoneally on two consecutive days in four dogs) or guanethidine (10 mg./Kg. intravenously on two days in four dogs) also abolished reflex venoconstriction. These results indicate that reserpine and guanethidine are capable of blocking reflex venoconstriction, and it is suggested that this action may contribute to their effects in the treatment of hypertension.

The present experiments describe effects of transient increases in heart rate on ventricular pacemakers during digitalization. Fifteen anesthetized, vagotomized dogs were studied during ouabain infusion or after recovery from ouabain-induced ventricular tachycardia. Vagal stimulation was used to assess baseline ventricular automaticity and 30-second periods of atrial pacing were utilized to evaluate the response of ventricular pacemakers following different increments in rate. After an average dose of 45 ± 3 (SE) µg/kg of ouabain, vagal stimulation unmasked an automatic ventricular focus in 14 animals and pacing was followed by ventricular acceleration in all 15 animals. The Degrees of ventricular acceleration varied directly with the pacing rate. The increment in ventricular rate above the level unmasked by vagal stimulation averaged 59 ± 4% of the increment in atrial rate. At the higher driving rates, a transient postpacing depression was present before the emergence of a rapid ventricular focus. These findings were unaffected by beta-receptor blockade with 4-(2-hydroxy-3-isopropylamino propoxy) acetanilide (AY21,011). The results indicate that transient increases in heart rate may accelerate rather than depress ventricular pacemakers during digitalization.

Comparisons have been made in the open chest dog of the oxygen usage of the left ventricle whose external work has been reduced to zero by 4 different procedures. The average oxygen usage of the left myocardium/100 Gm./min. is about 2.0 cc. in complete arrest with vagal stimulation or with intracoronary potassium injection; 3.8 cc. in fibrillation; and 3.4 cc. in the empty but beating heart. The oxygen value in the arrested heart approximates 27 per cent of that in the control state and the uptake during vagal stoppage varies directly with the control metabolic level. In the fibrillating ventricle the higher values are related to the intensity of fibrillatory movements, and in the empty beating heart to the control metabolic rate and to the frequency of the heart beat. The experiments with vagal arrest also permit estimation of the oxygen debt of the myocardium.