tachyarrhythmias The effects of premature stimulation of the His bundle potentials in dogs susceptible to sustained ventricular

Transcription

1 LABORATORY INVESTIGATION ARRHYTHMI The effects of premature stimulation of the His bundle on epicardial activation and body surface late potentials in dogs susceptible to sustained ventricular tachyarrhythmias JOSEPH F. SPEAR, PH.D., DAVID A. RICHARDS, M.D., GEORGE J. BLAKE, PH.D., MICHAEL B. SIMSON, M. D., AND E. NEIL MOORE, D. V. M. ABSTRACT Experiments were performed on 20 anesthetized dogs to determine the effects of premature stimulation of the His bundle on epicardial conduction and late potentials recorded on the body surface. Fifteen dogs underwent occlusion of the left anterior descending coronary for 2 hr followed by reperfusion, and five that did not undergo operation served as controls. Animals were studied 2 to 52 weeks after the induction of infarction, and four animals with infarction and four control animals exhibited no sustained arrhythmias in response to programmed ventricular extrastimulation. Five dogs with infarction and one control dog had ventricular fibrillation while the six remaining dogs had inducible sustained ventricular tachycardia. All animals with ventricular tachycardia had late potentials in the terminal portion of the signal-averaged body surface QRS complexes during sinus rhythm and QRS durations in the animals were 64 msec or greater. The voltage in the last 20 msec of the QRS complex was 13.5,uV or less and the duration of late activity below 30,uV was 18.2 msec or more. These values did not overlap values in animals with no inducible arrhythmias. Ventricular fibrillation was a nonspecific end point in these experiments and values overlapped those in animals with no arrhythmias and those with ventricular tachycardia. All animals with infarction and late potentials associated with their QRS complexes also had delayed and prolonged epicardial electrograms that extended into the time of the late potentials recorded from 45 standard sites in the infarcted regions. A single premature beat evoked by His bundle pacing (coupling interval of 192 to 270 msec) had no significant effect on late potentials or their relationship to epicardial activation in the area of infarct. However, changes in the durations of electrograms in response to premature beats were different in animals with infarction and ventricular tachycardia than in those with ventricular fibrillation. In animals with ventricular tachycardia, electrograms at 15 of 210 sites increased in duration by more than 10 msec while those at 61 of 210 sites decreased in duration. In animals with ventricular fibrillation, electrograms at 40 of 207 sites were of increased duration while those at 26 of 207 decreased. The decreases in duration were usually due to components of fractionated electrograms "dropping out" and likely represent local conduction block near the recording electrode. Thus, the present studies demonstrate that delayed and prolonged activation in the infarcted epicardium correlates with late potentials on the body surface and that single premature His bundle beats have no significant effect on either late potentials or on their relationship with epicardial activation. Animals with ventricular tachycardia could be distinguished from those with no arrhythmias based on signal-averaged QRS criteria defining late potentials. Circulation 72, No. 1, , From the Department of Animal Biology. School of Veterinary Medi- STUDIES IN MAN in which high-resolution techcine, University of Pennsylvania, Philadelphia. niques were used to record body surface potentials Supported by grants HL-23071, HL-25213, HL-28393, HL n and HL from the National Heart, Lung, and Blood Institute and have shown that low-energy high-frequency late poby a grant from the W. W. Smith Charitable Trust. tentials occur during the terminal portion of the QRS Address for correspondence: Dr. Joseph F. Spear, Department of Animal Biology. School of Veterinary Medicine, University of Penn- complex in many postmyocardial infarction patients sylvania, 3800 Spruce St., Philadelphia, PA susceptible to sustained ventricular tachyarrhyth- Received Jan. 14, 1985; revision accepted April 18, mias. 6 Studies correlating ventricular activation with Dr. Simson is the Samuel Bellet Associate Professor of Medicine and Cardiology. body surface potentials suggest that late potentials re- 214 CIRCULATION

2 LABORATORY INVESTIGATION-ARRHYTHMIA sult from abnormal delayed conduction within the infarcted region.7 This abnormal conduction may also provide the substrate for the reentrant arrhythmias in these patients.8' 9 Thus, the presence of late potentials on the electrocardiogram can be used as a noninvasive marker for patients at risk of developing ventricular arrhythmias. Acute occlusion of the left anterior descending coronary artery in dogs causes delayed activity in the ischemic zone and produces late potentials on the body surface.'0 Changes in the degree of delay associated with the evolution of ischemia are accurately reflected by similar changes in the timing of the late potentials. '0 During the early stage of infarction (3 to 6 days) premature atrial beats slow conduction through the infarct and cause a corresponding delay in late potentials on the body surface." The effect of premature activation on late potentials in the setting of chronic, stable infarction at a time when inducible sustained ventricular arrhythmias occur has not been investigated in either animals or man. Since late potentials are not recorded in all patients with ventricular tachycardia,3 it was thought that premature activation might provide a method for exaggerating late potentials and therefore increase their manifestation on the body surface as well as provide additional insight into their mechanisms. In the present studies we used a canine preparation of infarction in which both delayed epicardial conduction in the infarcted region and pacing-induced ventricular tachyarrhythmias occur chronically after coronary artery occlusion and reperfusion. 12 Our objectives were first to establish that delayed conduction in the infarcted region produces late potentials on the body surface and second to determine if criteria similar to those described for late potentials in man can be used to identify animals prone to inducible ventricular tachycardia. In addition, we evaluated the effect of premature supraventricular beats on epicardial conduction and body surface late potentials to determine if premature beats could enhance the sensitivity and specificity of the technique. Methods Animal preparation. Experiments were performed on 20 mongrel dogs weighing between 8 and 16 kg. Fifteen animals were studied 2 to 56 weeks after experimental anterior myocardial infarction, and five animals that did not undergo operation served as controls. The technique for producing infarction has been described previously.'3 Briefly, animals were anesthetized with sodium pentobarbital (30 mg/kg) and through a limited left lateral thoracotomy, the anterior descending coronary artery was occluded in two stages for 2 hr followed by complete reperfusion. Body surface potentials. At the time of study, body surface electrocardiographic potentials were recorded in each animal with signal-averaging procedures and computer-based analyses described by Simson.3 The animals were anesthetized with 30 mg/kg sodium pentobarbital and before thoracotomy bipolar x, y, and z leads were recorded. The x lead was recorded between the right and left lateral chest wall at the fifth intercostal space, the y lead between the superior manubrium and the left leg, and the z lead between the anterior chest wall over the sternum and a corresponding posterior position over the vertebral column. The analog signals from each lead were sequentially analog-to-digital converted at 1000 Hz to 12-bit accuracy and stored on a Winchester hard disk (Hewlett-Packard 9834A) with a Hewlett- Packard 9836 microcomputer. Between 100 and 200 QRS complexes were stored for each lead. The digital signals for each lead were passed through a template recognition program to eliminate noisy or ectopic beats. The x, y, and z QRS complexes were averaged and then filtered with use of a bidirectional digital filter set at a high-pass of 50 Hz. This value was chosen to eliminate low-frequency signals that are present in the ST segment of the dog.'0 The filtered signals for the three leads were combined into a vector magnitude and the QRS onset and termination were determined by a computer algorithm. In the experiments involving premature His bundle stimulation, digital conversion of x, y, and z lead analog data was triggered to begin 20 msec before the premature His stimulus and was terminated 256 msec later. During the bidirectional digital filtering of the averaged QRS complexes, the forward digital filter was gated to begin 5 msec after the termination of the stimulus artifact to prevent filter ringing from obscuring the beginning of the QRS complex. Pacing and recording. His bundle stimulation was chosen as the mode for introducing premature beats since this permitted coupling intervals shorter than those attained by atrial stimulation to be induced and also prevented the obscuring of late potentials by ventricular pacing. To pace the bundle of His, a bipolar catheter was introduced into the root of the aorta by way of the left carotid artery. The catheter had an interelectrode distance of 10 mm and was positioned to obtain a stable His bundle electrogram. The catheter was then connected to a constant-current stimulus isolation unit with use of the distal electrode as the cathode. The current intensity was increased until effective His pacing occurred. His pacing was confirmed by noting that the stimulus artifact-to-q wave interval was comparable to the His bundle electrogramto-q wave interval recorded before stimulation and by the same QRS and T wave configurations as during sinus rhythm. After every 8 basic His pacing beats a premature His bundle stimulus was introduced as early as possible without causing aberration in the QRS complex. To prevent retrograde atrial activation or atrial echo beats from obscuring the terminal portion of the premature His QRS complex, the atrium was paced with a second catheter introduced into the right atrium by way of the left jugular vein. For the 8 basic drive beats, the atrial stimulation was coincident with His bundle pacing. For the premature His beat, the atrium was preexcited such that it would be refractory during the time of the premature QRS complex. The signal-averaging procedure was first carried out in the dogs during sinus rhythm with the pacing catheters in place. The mean cycle length during sinus rhythm was 396 msec and ranged between 307 and 507 msec in the experiments. After the sinus rhythm sequence, signal averaging was repeated for the premature His beat. The coupling interval for the premature His stimulus relative to the last basic His stimulus ranged from 192 to 270 msec, mean 224 msec. After the signal-averaging procedures each animal's chest was opened with a midstemal incision and the heart was exposed and supported in a pericardial sling. In animals that had undergone the occlusion-reperfusion procedure, the infarcted Vol. 72, No. 1, July

3 SPEAR et al. region was located visually on the anterior left ventricle and a 45-site plastic template was oriented in a standard way and sutured over the infarcted region (figure 1). In animals without infarcts the template was applied in a corresponding region with the same orientation. Electrograms were sequentially recorded at the 45 positions with a hand-held roving electrode array in dogs in sinus rhythm and during premature His bundle stimulation at the same coupling intervals as in the signal-averaging procedures. Bipolar plunge-wire electrodes were inserted subepicardially in normal areas of the right and left ventricles for recording reference electrograms and for stimulation during the arrhythmia induction protocol. The roving electrode array consisted of two orthogonally arranged bipolar leads (figure 1, insert). Each pole was 0.3 mm in diameter and the center-tocenter distance between pairs was 0.6 mm. The bipolar axes of the electrode pairs are indicated in figure 1, insert, by the lines and are designated as x and y. The positive poles of each bipole are also indicated. During all recording procedures, the orientation of the array was maintained with the axis of the x bipolar electrode parallel to the longitudinal axis of the 45-position template. An orthogonal array was used for recording electrograms to decrease the possibility of activity not being recorded at a site due to a wavefront simultaneously intercepting both poles of a single bipolar electrode. The electrograms were filtered with a band pass of 40 to 1000 Hz. The electrocardiogram, both orthogonal electrograms, and the right and left reference electrograms were displayed simultaneously on a memory oscilloscope (Tektronix D- 15) and recorded on a Gould electrostatic strip-chart recorder at a paper speed of 250 mm/sec.. 1 cm FIGURE 1. A schematic representation of the standard location of a 45- site template applied over the infarcted region (stippled area) of the anterior left ventricle and used to guide the mapping electrode. Each site was separated from its neighbors by 2.5 mm. The numbering system for the sites is indicated. Inset, The arrangement and dimensions of the poles of the orthogonal dual bipolar electrode used to record activity at each site. The poles were insulated except at the surface in contact with 216 heart. After the experimental procedures, all animals were evaluated for their susceptibility to ventricular tachyarrhythmias by a previously described pacing protocol.'2 After basic drive at cycle lengths of 350 and 300 msec, single, double, and triple ventricular extrastimuli were introduced. The stimulation protocol was first performed with right and then left ventricular stimulation at an intensity of twice diastolic threshold. If tachyarrhythmias were not induced, the entire procedure was repeated with a current of four times diastolic threshold. Once an arrhythmia was reproducibly induced the protocol was terminated. At the end of the experiments the animals were killed, their hearts were removed, and the area of infarction was verified by gross inspection. Data analysis. Dogs were separated into three groups. Group had no inducible ventricular tachyarrhythmias, group II had inducible ventricular fibrillation, and group III had inducible sustained (greater than 10 see) ventricular tachycardia during electrophysiologic testing. All 45 electrograms recorded during the mapping procedure were referenced to the beginning of the QRS complex and parameters were manually digitized with a Hewlett-Packard 9836 computer system. The system had a resolution of mv and 0.1 msec. The orthogonal bipolar electrode array was considered to be at a nonactivated epicardial site when the largest component in both electrograms did not exceed 0.1 mv. 12 The onset of activity at a site was taken as the point on the electrogram at which the signal deviated from the baseline by greater than 0.1 mv and the terminatlon of the signal was taken as the latest deflection exceeding 0. 1 mv. Local activation time was measured to the peak or baseline crossing of the largest fast deflection in the electrograms. Electrogram duration was measured as the difference between the termination and onset of the electrogram. Electrogram amplitude was measured as the vector sum of the peak amplitude of the largest fastest deflections in the orthogonal electrograms x and y. Late potentials were characterized by parameters that are similar to those used in man3 4 and were calculated from the signal-averaged and filtered body surface QRS potentials. Parameters included the total QRS duration, the total root-meansquare (RMS) voltage in the QRS, the RMS voltage in the last 20 msec of the QRS complex, and the duration of activity in the terminal portion of the QRS complex below the 30 jtv level. Statistical methods. Electrographic data are presented for each experiment as the mean and SD for the 45 electrode sites. Mean and SD pooled data from all experiments within a group are also presented. Significant differences between groups for a given parameter were determined by analysis of variance. Differences in electrographic parameters recorded during sinus rhythm and His premature stimulation for the 45 sites were analyzed by Student's paired t test for each animal. Body surface QRS data were compared between groups by analysis of variance. A p value of less than or equal to.05 was considered to indicate a significant difference. QRS parameters during sinus rhythm and His bundle stimulation were compared by linear regression analysis. Results The characteristics of the arrhythmias for each animal are presented in table 1. Four of the control animals and four of the animals with infarction had no inducible ventricular arrhythmias. Six animals, one of which was a control dog, had inducible ventricular fibrillation. The six remaining dogs with infarction exhibited inducible sustained ventricular tachycardia CIRCULATION

4 LABORATORY INVESTIGATION-ARRHYTHMIA TABLE 1 Characteristics of arrhythmias Tachycardia Induction parameters Experi- Weeks Cycle Site of Basic cycle Number Current ment after length stimu- length of extra- (x dia- No. infarct Type (msec) lation (msec) stimuli stolic) Group I 1 LV LV LV LV LV LV LV LV Group II 1 - LV LV LV LV LV RV Group III 1 24 Poly 160 LV Mono 160 LV Mono 176 LV Mono 152 RV Mono 152 LV Mono 136 RV LV = left ventricle; RV = right ventricle; Poly = polymorphic sustained ventricular tachycardia; Mono sustained ventricular tachycardia. with cycle lengths ranging between 136 and 176 msec. With the exception of experiment 1, the type of ventricular tachycardia induced was monomorphic. The polymorphic ventricular tachycardia in experiment 1, although reproducibly induced and sustained for greater than 10 sec, eventually degenerated to ventricular fibrillation. Body surface QRS complex in sinus rhythm. Table 2 summarizes the characteristics of the filtered body surface QRS complexes recorded during sinus rhythm. It was noted when group I animals with no arrhythmias and group III dogs exhibiting ventricular tachycardia were compared that all animals with ventricular tachycardia had QRS durations of 64 msec or greater. The voltage in the last 20 msec of the QRS complex was 13.5 gv or less and the duration of activity below 30 gv was 18.2 msec or more in the group IlI animals. These values did not overlap with those in animals with no arrhythmias. Therefore, animals having inducible ventricular tachycardia had low-amplitude, highfrequency late potentials on the terminal portion of the filtered QRS complex that clearly distinguished them from animals having no arrhythmias. Vol. 72, No. 1, July 1985 monomorphic In the group II animals with ventricular fibrillation mean values for these parameters fell between those in groups I and III. However, values for individual animals with ventricular fibrillation ranged across those for animals with no arrhythmia and those with ventricular tachycardia. Therefore, the group II animals were a mixed population and could not be distinguished from those in groups I and III based on the filtered QRS complex. Epicardial activation during sinus rhythm. In table 3 the characteristics of the ventricular electrograms from dogs in sinus rhythm are presented for each parameter. Animals without arrhythmia (group I) differed from those with ventricular tachycardia (group III), with electrogram amplitudes being significantly depressed and the local activation time and duration of the electrograms being significantly prolonged in group III. These findings are similar to those previously reported for another series of experiments in which different recording techniques were used.10 The coefficients of variance (SD divided by the mean times 100) for the local activation times were significantly greater in the group III than in group I animals, indicating a greater 217

5 SPEAR et al. TABLE 2 Characteristics of the signal-averaged QRS complex during sinus rhythm QRS Experiment duration Vt V20 D30 No. (msec) (,V) (iv) (msec) Group I Mean ± SD 55.8 ± ± ± ± 2.4 Group II Mean ± SD 65.7 ± ± ± ± 14.2 p value NS NS Group III Mean+SD ± ± p value <.001 NS <.002 <.001 p value by analysis of variance versus group I. Vt = total RMS voltage in the QRS; V20 = RMS voltage in the last 20 msec of the QRS; D30 = Duration of late activity below 30,uV. degree of disparity in the times of activation among the 45 sites for animals with ventricular tachycardia. As was the case for the filtered QRS parameters, individual animals with ventricular fibrillation (group II) had electrographic characteristics that ranged between values for animals with ventricular tachycardia and those for dogs with no arrhythmias. The relationship between epicardial activation and the filtered QRS for each experiment is graphically displayed in figures 2, 3, and 4. Sinus rhythm results are shown at the left in each case. The timing of onset and termination of the electrograms relative to the beginning of the QRS complex are displayed as horizontal lines for each of the 45 sites, with the time of local activation of each site indicated by an open circle. Sites are displayed in sequence from 1 through 45 from bottom to top with each nonactivated site being indicated by an x at the beginning of the QRS complex. The three vertical lines on the abscissa, from left to right, indicate the time of onset of the QRS complex, the time at which the voltage dropped below the 30 pv level, and the end of the QRS complex. The area of the QRS complex between the time at which the voltage dropped below 30,V to the end of the complex is emphasized by the dark shading. In group I animals epicardial local activation occurred within the high-voltage region of the QRS complex (figure 2; table 4). With the exception of in experiment 6, electrograms tended to be of short duration and few extended into the terminal portion of the QRS. In contrast, group III animals (figure 4) had prolonged or delayed electrograms that extended into the terminal portion of the QRS complex (table 4). Group II animals (figure 3) displayed patterns that ranged between those observed in group I and group III dogs. Figures 2 to 4 and table 4 demonstrate that animals with delayed or prolonged activation also had late potentials and prolonged QRS durations. However, the correlation between epicardial activation and late potentials was not perfect. There were epicardial sites the activity of which extended beyond the end of the QRS and did not appear to contribute potentials on the body surface. It is probable that there is some vector cancellation of epicardial activity involved and also that there is an energy "threshold" that must be exceeded to allow activity to be detected on the body surface. The latter point is supported by experiment 2 in figure 3 and experiment 1 in figure 4. In these cases, there were a considerable number of sites on the epicardial surface activated beyond the end of the QRS complexes and this late activity was not detected on the body surface. In these two experiments the lowest mean electrogram amplitudes were noted (table 3), suggesting that the total energy was insufficient for detection. The effect of premature His bundle pacing. Technically acceptable His bundle stimulation was not accomplished in seven of the 20 animals (figures 2 to 4, table 3). Inspection of the records after the experiments revealed the following problems. In experiment 1 in the group I animals His bundle pacing was confounded by simultaneous ventricular septal activation during the epicardial mapping procedure. This occurred also during the signal-averaging procedure in experiments 4, 5, and 8. In experiment 7, a retrograde p wave obscured the terminal portion of the QRS complex during signal averaging of the premature His beat. In the group II animals (figure 3, table 3), ventricular septal pacing occurred in experiment 6. For the group III animals (figure 4, table 3) ventricular septal stimulation occurred in experiment 4 and a retrograde p wave 218 CIRCULATION

6 TABLE 3 Characteristics of the electrograms during sinus rhythm and premature stimulation of the His LABORATORY INVESTIGATION-ARRHYTHMIA Activation coefficient of Experiment Amplitude (mv) Local activation time (msec) Duration (msec) variance (%) No. Sinus His Sinus His Sinus His Sinus His Group I ±4.5A ±3.OA ±2.OA ± 1.7A ±3.7A ±7.1A Mean+SD Group II I A A ±0.6A ±36.7A ±45.3A A OA ± 33.8A A Mean+lSD 4.6± ± p value NS.003 <.001 <.001 <.001 <.001 NS NS Group III ±13.8A A ± ± Mean+SD 2.6_ _ _ p value < <.001 <.001 <.001 < p value by analysis of variance. ANot significant by paired t test. obscured the terminal QRS in experiment 5. In some experiments the end of the p wave from the atrial preexcitation obscured the beginning of the premature His complex (for example, experiments 1 and 2, figure 4). In these cases the beginning of the QRS was estimated by overlaying the sinus complex. In animals in which premature His bundle stimulation was successful there were no significant differences in the QRS duration (figure 5, A), the voltage in the last 20 msec of the QRS (panel B), or in the duration of activity below 30,uV (panel C) during the premature His beats compared with sinus rhythm. Premature stimulation of the His bundle had a variable effect on the characteristics of electrograms recorded from the infarcted epicardial surface (table 3 and figures 2 to 4). In individual experiments, premature stimulation of the His bundle increased, decreased, or caused no significant change in the parameters measured. The effect of prematurity on the electrogram duration was different in the group II and group III animals. Table 4 lists electrographic data that indicate changes in duration greater than 10 msec for each experiment. Group II animals had a larger percentage of electrograms showing an increase in duration than group III dogs. The opposite was true for the group III animals, in which duration was more often decreased in response to a premature beat. This difference was significant by chi-square analysis. In these cases a decrease in the duration of a fractionated electrogram was usually due to components of the electrogram dropping out in response to the premature beat (figure 6). The likely mechanism of this effect involves local block of conduction near the recording electrode. Discussion Arrhythmias and late potentials. All animals with inducible sustained ventricular tachycardia (group III) had well-defined infarcts, prominent delays in epicar- Vol. 72, No. 1, July

7 SPEAR et al. 1 HIS PREMATURE HIS PREMATURE 5 2nL t 6 3 A-1-~~ 7 ip wave uvl 50 msec FIGURE 2. Signal-averaged, filtered QRS complexes and epicardial electrographic parameters for eight animals without inducible arrhythmias. Experiments 1 through 4 were controls, and dogs in experiments 5 through 8 underwent infarction. Data obtained during sinus rhythm are shown at the left and those obtained during premature His bundle stimulation at the right. See text for discussion. dial activation, prolonged and fractionated electrograms, and late potentials in the filtered body surface QRS. These findings are in marked contrast to those in group I animals in which no arrhythmias could be induced. Group II animals, in which ventricular fibrillation was induced, included those animals with infarcts and delayed and prolonged epicardial activation and late potentials (experiments 2 and 3) as well as those without late potentials and minor or moderate abnormalities in epicardial activation (experiments 4, 5, and 6). Included in this group was a control dog (experiment 1). In this animal, induction of ventricular fibrillation required 3 premature beats delivered to the left ventricle at four times diastolic threshold after pacing at a basic cycle length of 350 msec. Before ventricular fibrillation ensued, there were no extra beats induced by the premature stimuli. Induction of ventricular fibrillation in normal dogs by aggressive stimulation protocols have been reported by others.'14 15 In 220 experiments 5 and 6 in group II fibrillation also was induced by similar aggressive pacing and was not preceded by nonsustained runs of ventricular beats (table 1). Thus, because ventricular fibrillation was a nonspecific end point in these experiments, group II must be considered a mixed population of animals. The range of values for the filtered QRS and electrographic parameters in group II extended across values obtained in group I and in group III. In contrast, all animals with inducible ventricular tachycardia could be distinguished from those with no arrhythmia based on abnormal activation in the infarct and the following body surface QRS criteria: a QRS duration of 64 msec or greater, a RMS voltage in the last 20 msec of the QRS complex of 13.5,V or less, and a duration of terminal activity below 30,uV of 18.2 msec or greater. The similarity between these data and the findings in humans' further supports this animal preparation as a useful analog for studying CIRCULATION

8 LABORATORY INVESTIGATION-ARRHYTHMIA 1 HIS PREMATURE 4 HIS PREMATURE 2 o i = 5 5 fl p,, ===9-0 f- i 0to- Mh F * 0 k 0*,,.., CL _ 30 uv 50 msec FIGURE 3. Signal-averaged, filtered QRS complexes and epicardial electrographic parameters for six animals with inducible ventricular fibrillation. Experiment 1 was from a control animal. The data are arranged as in figure 2. 1 HIS PREMATURE ' 1'''b ' ' #ipl 2 5 the mechanisms of sustained ventricular tachycardia. Epicardial activation and late potentials. A consistent finding in all experiments was that animals with late potentials had delayed and fractionated epicardial electrograms that persisted into the time of the late potentials (table 4). Although all regions of the infarct were not sampled and some epicardial activity extended beyond the end of the QRS, these correlations suggest HIS PREMATURE 30 uv 50 msec *Pwave 3 91 LT -.1!1 0 :%- l1 -W - ---r -~~~~~~~~~~~~~~I - ; ~~~~~~~~~~~~~~~ J L I -Z- -d.j FIGURE 4. Signal-averaged, filtered QRS complexes and epicardial electrographic parameters which ventricular tachycardia was inducible. The data are arranged as in figure 2. for six infarcted animals in Vol. 72, No. 1, July

9 SPEAR et al. TABLE 4 Effect of premature stimulation of the His on electrogram local activation and duration Proportion of local activations Proportion of electrograms Changes in electrogram duration Experiment occurring after the onset of D30 extending into D30 >10 msec No. Sinus His Sinus His Increasing Decreasing Group I 1 0/45 5/45 2 0/45 0/45 6/45 15/45 1/45 1/45 3 0/45 0/45 1/45 1/45 0/45 0/45 4 0/45 6/45 5 0/45 0/45 6 0/45 0/45 21/45 26/45 0/45 0/45 7 0/45 11/45 0/45 0/45 8 0/45 8/45 Total 0/360 0/135 58/360 42/135 1/180 11/180 (0.0%) (0. 0%) (16.1%) (31.1%) (0.6%) (6.1%) Group I1 1 0/45 0/45 5/45 11/45 2/45 2/ /34 13/34 31/34 26/31 15/31 5/ /43 22/41 34/43 38/41 17/41 8/ /45 20/45 45/45 44/45 6/45 11/45 5 5/45 0/45 41/45 39/45 0/45 0/45 6 0/45 3/45 Total 60/257 55/ / /207 40/207 26/207 (23.3%) (26.6%) (61.9%) (76.3%) (19.3%) (12.6%) Group III 1 32/45 39/43 44/45 43/43 12/43 11/ /42 11/41 39/42 25/41 0/41 24/ /38 24/38 36/38 33/36 0/36 8/ /45 44/45 5 0/45 23/45 1/45 7/ /45 13/45 29/45 30/45 2/45 11/45 Total 106/305 87/ / /165 15/210 61/210 (34.8%) (52.7%) (70.5%) (79.4%) (7.1%) (29.1%) Abbreviations are as in table 2. that abnormal conduction within the infarcted epicardium contributes to the late potentials on the body surface. The low mean electrogram amplitude in experiment 2 in group II dogs and experiment 1 in group III animals and the poorer correlation between delayed epicardial activation and late potentials in these cases emphasizes that there is an energy threshold below which delayed activity is not detected on the body surface. The findings of the present studies contrast in one respect with data derived from similar studies in humans. In patients with ventricular tachycardia and late potentials, fragmented electrograms are seen primarily at endocardial sites.' The focus of the arrhythmia is also usually localized to the endocardium in these patients. 16 In the animal preparation the epicardium appears to be the preferential region for these phenomena. 12 Effects of premature beats. The effect of premature 222 stimulation of the His bundle on late potentials was minimal. While the configuration of the late potentials in the filtered signal was visibly changed in some cases (for example, experiment 3, figure 3; experiment 1, figure 4), the criteria used to define their presence were not significantly altered (figure 5). In the present experiments a single premature beat did not facilitate identification of late potentials. The degree of effect of a premature beat on epicardial activation was also relatively minor. In the group II and group III animals, some sites showed prolongation of activation time and/or duration while in others these values shortened or showed no change. However, in individual experiments, the mean values during the His premature beat were comparable to those during sinus rhythm (table 3). This correlates with the lack of effect of a premature beat on late potentials. The present study does not address the question of the effects of multiple His bundle or ventricular extra- CIRCULATION

10 A v -- a o NO ARRHY. a V. FIB. a V. TACH. I 0 y= x r= U4ATION, SINUIS AIYTW l0sec/ LABORATORY INVESTIGATION-ARRHYTHMIA stimuli on the above parameters. At least two ventricular extrastimuli were required to induce ventricular tachycardia in the group III animals (table 1). Since these were delivered at shorter coupling intervals than the His bundle extrastimuli their effects on conduction would be expected to be more severe. However, the effects of the single premature His beat on individual electrograms does provide some insight into the mechanisms underlying the induction of tachycardia by premature beats. B 60 ECG so o NO ARRHY. e V. FIB. e V. TACH. 0 eo 40 1 C O V20 SINIUS RHfTHM (iicrovoltsi o NO ARRHY. e V. FIB. a V. TACH. ECG 0 a y= x r= Y= x r= l I1 x _ - T l1 X ~-1 ia_ Y P wave 50 T HIS PREMATURE A 1 mv 100 msec sims PYrTMI (esec] FIGURE 5. A comparison of the signal-averaged QRS complexes during sinus rhythm and during premature stimulation of the His bundle. A, The duration of the QRS complex, B, The amplitude in the last 20 msec of the QRS (V20). C, The duration of late activity below 30,uV (D30). Also shown are the equations for the regression lines and their r values. Vol. 72, No. 1, July 1985 FIGURE 6. Analog records demonstrating the shortening of the duration of electrograms at a site within the infarcted region during premature stimulation of the premature His. The cycle length during sinus rhythm was 467 msec and the coupling interval of the premature His beat was 220 msec. 223

11 SPEAR et al. The shortening of some electrogram durations in response to a premature Hisbeat (figure 6) canbe explained by local block in part of the conducting path responsible for the fractionated electrogram in the region of the recording electrode. In this regard, block of conduction in the infarcted region seems analogous to what occurs in normal fast-conducting tissue such as muscle or the His-Purkinje system, where relatively little prolongation in conduction time occurs in response to the premature beat before the coupling interval that induces the block. This is in contrast to conduction in earlierinfarcts,' where a premature beat delays both activation and late potentials. These tissues behave like the atrioventricular node, where beats of increasing prematurity cause progressive delays in conduction before the point of block is reached. The sudden local block in the present experiments was not unanticipated since previous studies have shown that action potential depolarizations are relatively normal in the infarcted tissues beyond 2 weeks.'7 18 At that time the major mechanism of the abnormal discontinuous conduction in the infarct is probably disruption in cell-to-cell electrical continuity'9 and not depression in membrane depolarization. More epicardial sites of groupiii dogs than of group IL dogs exhibited electrogram shortening in response to a premature beat (table 4). In the infarcts of animals capable of sustaining ventricular tachycardia, conduction seems more prone to undergo local block in response to a premature beat. One implication of this may be that it is not the overall degree of delay in conduction in the infarcted region that determines the inducibility and character of the arrhythmia but rather these are determined by the way local areas interact. The most prominent late potentials in this series were observed in experiment 3 of figure 3. However, only eight of the 41 viable epicardial sites showed decreased duration in response to a premature beat (table 4) and the arrhythmia in this dog was ventricular fibrillation. Experiment 2 of figure 4 illustrates results in a dog with less prominent late potentials. In this case, however, 24 of 41 sites showed decreased duration and the arrhythmia was ventricular tachycardia. Thus, delayed and dissociated conduction sufficient to produce prominent late potentials on the body surface in dogs in sinus rhythm may not be sufficient to sustain a reentrant circuit producing ventricular tachycardia. This is consistent with clinical findings that some patients with late potentials do not have ventricular tachycardia.3 We ackowledge the expert technical assistance of RalphIannuzzi, William Moore, Bejay Moore, and Charles Prood. References 1. Uther JB, Dennett CJ. Tan A: The detection of delayed activation signals of low amplitude in the vectorcardiogram of patients with recurrent ventricular tachycardia by signal averaging. In Sandoe E, Julian DG, BellJW, editors: Management of ventricular tachycardia role of mexiletine. Amsterdam, 1978, Excerpta Medica, p Rozanski JJ, Mortara D, Myerburg RJ, Castellanos A: Body surface detection of delayed depolarizations in patients with recurrent ventricular tachycardia and left ventricular aneurysm. Circulation 63: 1172, Simson MB: Use of signals in the terminal QRS complex to identify patients with ventricular tachycardia after myocardial infarction. Circulation 64: 235, Breithardt G, Becker R, Seipel L, Abendroth R, OstermeyerJ: Non-invasive detection of late potentials in man a new marker for ventricular tachycardia. Eur Heart J 2: 1, Denes P, Santarelli P, Hauser RG. Uretz EF: Quantitative analysis of the high-frequency components of the terminal portion of the body surface QRS in normal subjects and in patients with ventricular tachycardia. Circulation 67: 1129, Cain ME, Ambos HD, Witkowski FX, Sobel BE: Fast-fourier transform analysis of signal-averaged electrocardiograms for identification of patients prone to sustained ventricular tachycardia. Circulation 69: 711, Simson MB, Untereker WJ, Spielman SR, Horowitz LN, Marcus NH, Falcone RA, Harken AH, Josephson ME: Relation between late potentials on the body surface and directly recorded fragmented electrograms in patients with ventricular tachycardia. Am J Cardiol 51: 105, Wellens HJJ, Durrer DR, Lie KI: Observations on mechanisms of ventricular tachycardia in man. Circulation 54: 237, Josephson ME, Horowitz LN, Farshidi A. Spielman SR, Michelson EL, Greenspan AM: Sustained ventricular tachycardia. Evidence for protected localized reentry. Am J Cardiol 42: 416, Simson MB, Euler D, Michelson EL, Falcone RA, Spear JF, Moore EN: Detection of delayed ventricular activation on the body surface in dogs. Am J Physiol 241: H Berbari EJ, Scherlag BJ, Hope RR, Lazzara R: Recording from the body surface of arrhythmogenic ventricular activity during the S-T segment. Am J Cardiol 41: 697, Richards DA, Blake GJ, Spear JF, Moore EN: Electrophysiologic substrate for ventricular tachycardia: correlation of properties in vivo and in vitro. Circulation 69: 369, Michelson EL, Spear JF, Moore EN: Electrophysiologic and anatomic correlates of sustained ventricular tachyarrhythmias in a model of chronic myocardial infarction. Am J Cardiol 45: Wetstein L, Michelson EL, Simson MB, Moore EN, Harken AH: Initiation of ventricular tachyarrhythmia with programmed stimulation: sensitivity and specificity in an experimental canine model. Surgery 92: 206, Hamer AW, Karagueuzian HS, Sugi K, Zaher C, Mandel WJ, Peter T: Factors related to the induction of ventricular fibrillation in the normal canine heart by programmed electrical stimulation. J Am Coll Cardiol 3: 751, Marcus NH, Falcone RA, Harken AH, Josephson ME, Simson MB: Body surface late potentials: effects of endocardial resection in patients with ventricular tachycardia. Circulation 70: 632, Spear JF, Michelson EL, Moore EN: Cellular electrophysiologic characteristics of chronically infarcted myocardium in dogs susceptible to sustained ventricular tachyarrhythmias. J Am Coll Cardiol 1: 1099, Gardner P, Ursell PC, Fenoglio JJ, Allessie MA, Bonke FIM, Wit AL: Structure of the epicardial border zone in canine infarcts is cause of reentrant excitation. Circulation 64: 320, 1981 (abst) 19. Spear JF, Michelson EL, Moore EN: Reduced space constant in slowly conducting regions of chronically infarcted canine myocardium. Circ Res 53: 176, 1983 a 224 CIRCULATION