Myocardial Infarction Period

Transcription

1 131 Ventricular Arrhythmias in the Subacute Myocardial Infarction Period High-Resolution Activation and Refractory Patterns of Reentrant Rhythms Mark Restivo, William B. Gough, and Nabil El-Sherif Patterns of activation, functional conduction block, and effective refractory periods during reentrant activation were investigated in a 4-day postinfarction canine model using a 64-channel high-resolution (1 mm) bipolar electrode array. Lower resolution (3-1 mm) isochronal activation maps of the entire epicardial surface were constructed from 126 sites during the initiation and sustenance of reentry and showed reentrant wave fronts that circulated around arcs of functional conduction block. During initiation of reentry by premature stimulation, high-density recordings from these same regions showed that conduction block occurred abruptly, within 1 mm, and without prior decrement of the impulse. Electrograms recorded in proximity to the arc of block were comprised of two deflections: a local activation potential and an electrotonic potential reflecting activation 1 mm away; the reverse order of activation and electrotonus was observed on the opposite side of the arc of block. The occurrence of functional conduction block during premature stimulation in this model was correlated with abrupt increases in effective refractory periods of 1-12 msec (27±24 msec; mean SD) within 1 mm or less. Neither the abrupt change of refractoriness nor functional conduction block appeared to depend on differences in excitability, the geometrical characteristics of the surviving epicardial layer, or the orientation of the myocardial fibers. During sustained reentrant activation, high-density recordings along the arcs of block showed split electrograms comprised of local activation and electrotonus, which were identical in morphology to those recorded during the initiation of reentry. The interval between the deflections was shorter at the ends of the arc and increased to a maximum value at the center of the arc. The activation potentials corresponded in time with activation of large isochronal regions on either side of the arc of block. There was evidence that at least part of the arc of block during sustained reentry represented thin discrete zones of constant block due to electrotonic influences of impulse penetration from both sides of the arc. Our results strongly suggest that continuous arcs of functional conduction block are a necessary prerequisite for both the initiation and the sustenance of reentrant activation in subacute canine myocardial infarction. Functional conduction block during the initiation of reentry was due to abrupt changes in refractoriness, within a distance of 1 mm or less. (Circulation Research 199;66: ) T he essential physiological requirements for reentry are unidirectional conduction block and slow conduction, which permit recovery of excitability proximal to block.1 A series of investigations, from this laboratory2-4 and others,5-9 have indicated that reentrant excitation occurs around From the Cardiology Divisions, Department of Medicine, State University of New York Health Science Center, and Veterans Administration Medical Center, Brooklyn, New York. Supported by National Institutes of Health grant HL-3668 and Veterans Administration Medical Research Funds. Address for correspondence: Mark Restivo, PhD, Brooklyn VA Medical Center, Cardiology Division (lla), 8 Poly Place, Brooklyn, NY Received March 3, 1989; accepted December 19, zones of dissociated conduction. We have defined these zones as arcs of functional conduction block.3.4 Functional conduction block and slow propagation of premature responses appear related to the spatial distribution of abnormal electrophysiological properties that have been identified in the ischemic heart. We have previously reported that the physiological substrate for functional conduction block in the 4-day postinfarction dog heart is a spatially nonuniform refractory distribution in which refractoriness increases monotonically from normal zone to ischemic zone.1 Based on measurements 5-1 mm apart, a refractory gradient of 2 msec/cm had been suggested as a threshold value for the formation of unidirectional conduction block.1

2 High-Density Electrode Recordings Figure 1 shows the arrangement of the highdensity bipolar electrode array (HDBEA). Briefly, the array was constructed of 64 bipolar electrodes, arranged in eight columns (1-8 in Figure 1) of eight bipolar electrode pairs (A-H in Figure 1). The interelectrode and interpolar spacing was 1 mm. The poles of the electrodes consisted of Teflon-coated silver wire (125,uM) embedded in a flexible acrylic material (Flexacryl, Kerr Industries, Romulus, Michigan). The material was polished flush, permitting only the exposed cross section of the wire to contact the tissue. All electrograms acquired and stored on the computer were amplified at a gain of 5 and were filtered at a bandpass of.5-5 Hz. High-resolution refractory and activation measurements were performed in those hearts in which functional conduction block, as defined by the above criteria, could be identified during whole-ventricle mapping. The HDBEA was then placed under the sock, over a region believed to exhibit functional conduction block in the whole ventricle maps. The HDBEA was oriented along or across the longitudi- The present study examines the relation of spatial variations of refractoriness with the occurrence of functional conduction block at a higher spatial resolution (1 mm). The criteria for accurate identification of conduction block, which will be outlined, are based on high-resolution refractory determinations and multiple, simultaneous, close bipolar electrogram recordings. We will show that a continuous arc of functional conduction block occurs during the initiation of reentrant excitation. Functional block of a premature impulse will be shown to occur without decremental conduction and to be unrelated to fiber orientation. Results obtained during premature stimulation will be used as a guide for the interpretation of electrical activity during sustained ventricular tachycardia. The morphological characteristics of activation and block will be compared with electrogram features during sustained ventricular tachycardia. Preliminary studies have been reported.11,12 Materials and Methods Surgical Preparation Details of the surgical preparation can be found elsewhere13 and will be described only briefly here. Experiments were performed on 27 heartwormnegative mongrel dogs, weighing 15-2 kg, in which the left anterior descending coronary artery was ligated distal to the anterior septal branch. Four days after ligation, the dog was reanesthetized with 3 mg/kg i.v. sodium pentobarbital and given supplemental doses as necessary. Supplemental anesthetic and saline were administered through a catheter placed in the cephalic vein. The dogs were ventilated with room air through a cuffed endotrachial tube using a positive pressure respirator (Harvard Apparatus, South Natick, Massachusetts). Electrocardiographic lead II and aortic blood pressure (Statham transducer, Gould, Cleveland, Ohio) were continuously monitored on a DR12 monitor (Electronics for Medicine, Pleasantville, New York). The heart was exposed through a left thoracotomy. Core temperature and intrathoracic temperature were monitored using two electronic thermometers (Yellow Springs Instrument, Yellow Springs, Ohio). To slow the sinus rate, a stimulator (model 588, Grass Instrument, Quincy, Massachusetts) was used to stimulate the right and left vagosympathetic trunks, through two pairs of Teflon-insulated silver wires (.1-in. diameter), with square wave pulses (.1-.5 msec) at a frequency of 2 Hz and at 1-1 V. Whole-Ventricle Isochronal Mapping A sock electrode array was placed on the ventricular surface for simultaneous recording at 126 epicardial sites. Each bipolar electrode consisted of a pair of silver wires (.5-in. diameter) sutured to the sock with an interpolar distance of 1-2 mm. The distance between electrodes was 4-1 mm, with a higher concentration of electrodes covering the zone overlying the infarction. Additional details of the Restivo et al High-Resolution Mapping 1311 electrode arrangement and recording technique are reported elsewhere.24 Programmed electrical stimulation was provided by a digital stimulator (model DTU-11, Bloom, Reading, Pennsylvania) through a bipolar plunge electrode consisting of two hooked stainless steel wires (enamel-coated,.5 -in. diameter) placed in a 23-gauge hypodermic needle. The control stimulation site was located in 1) the right ventricle, adjacent to the septal border of the infarct, or 2) the left ventricular base, near the basal lateral border of the infarct. After the surgical preparation and sock electrode placement, the ribs were approximated, and the chest cavity was closed. Once the core temperature had stabilized, programmed stimulation was applied to the control site to induce reentrant rhythms. The control stimulation sequence consisted of a train of eight basic driven beats (S,-S, of 36-4 msec), at twice diastolic threshold, followed by a premature stimulus, S2. The premature stimulus was introduced at decreasing coupling intervals, starting at 26 msec, until unstimulated ventricular responses were induced. Once a reproducible rhythm could be initiated, electrogram recordings were obtained from the data acquisition system and used to construct isochronal maps of epicardial activation. Isochrones were delineated by closed contours at 2-msec intervals beginning with the earliest detected time of activation. For the whole ventricle activation maps, functional unidirectional conduction block was defined, as in our previous studies,2'4 by an activation time difference between adjacent recording sites of more than 4 msec or by electrotonic deflections representing distant activation. A continuous line could be drawn through these regions and is defined as an arc of functional conduction block.

3 1312 Circulation Research Vol 66, No 5, May 199 o A * * * * * * * * * * * * * DO * * * * * * S O S c E * * * * * * * * * * * * * F OSS * * * * * * * * * * * * G H S 8 S S S * S S S S O S S O S O FIGURE 1. High-density bipolar electrode array consists of 64 electrodes arranged in eight columns (1-8) of eight bipolar pairs (A-H). Distance between all adjacent poles o o (filled circles) is 1 mm. Bipolar stimulation (open circles) can be applied from the midpoint of each side. The cathode of each stimulating pair is located closest to the recording electrodes. H- 1 mm nal fiber axis. After placement of the HDBEA, the chest cavity was again closed. Measurements were made once the intrathoracic temperature had returned to control level. In five hearts, the HDBEA was manually positioned at multiple locations in an open chest during sustained monomorphic ventricular tachycardia. Refractory periods were not determined during this procedure. Recordings were made during S1-S2 stimulation either from the control site or from the stimulating electrodes contained in the HDBEA. Bipolar stimulation could be applied along any of the four outer edges of the HDBEA to initiate propagation parallel or transverse to the fibers; conduction was in a normal to ischemic direction or in an ischemic to normal direction. To ensure uniform wave-front propagation, the stimulating electrode pair was oriented perpendicular to the array, with the cathode closest to the recording electrodes. Unipolar stimulation could be applied to any of 128 sites for refractory period determination (see below). At the end of the experiment, the HDBEA location was marked, and the heart was fixed in 1% methanol-free ultrapure formaldehyde (Polyscience, Niles, Illinois) for later histological study. The sections were stained with hematoxylin and eosin. The myocardial fiber axis and infarct in relation to the HDBEA recording locations were determined using light microscopy. Refractory Period Determination If the HDBEA electrograms revealed electrotonic deflections indicative of conduction block, effective refractory periods were determined. Refractoriness was determined at one or both poles of each electrode. By using a previously published method,1 effective refractory periods were measured at twice diastolic threshold after a train of eight basic driven beats. The measurements were determined at 5 -msec increments. The effective refractory period of a site was defined as the longest interstimulus interval, SI-S2, that failed to produce a propagated local response. A maximum diastolic threshold of 1.5 ma was used to ensure that the propagated response originated from within 1 mm of the test site and did not engage adjacent regions.14 Values above 1.5 ma were excluded from the study. Refractoriness was also measured by displaying the four sites (two parallel, two transverse) surrounding the test site. The left panel of Figure 2 shows the shortest S1-S2 interval (15 msec [effective refractory period plus 5 msec]) in which all four sites responded. At an S1-S2 interval of 145 msec (effective refractory period), all sites failed to respond. Conduction Block Determination by High-Resolution Mapping High-resolution mapping was performed to corroborate our interpretation of conduction block obtained during whole-ventricle mapping studies. To determine the site of conduction block, one must differentiate between activation potentials, representing a propagated impulse passing under the electrode poles, and electrotonic potentials, representing activity distant from the electrode. For an ideal bipolar recording, in which each pole is considered a point, we will consider it is possible to register one or two activation times (two activation times will be considered possible for asynchronous activation under each pole, as will be seen in "Results"). During premature stimulation, multiple high-resolution electrogram recordings and effective refractory period values were used for the decision. Any deflection that occurred before the expiration of refractoriness was considered electrotonic; the following deflection was considered an activation potential. Conduction block during sustained ventricular tachycardia was determined by electrogram analysis only. Effective refractory periods were not available during ventricular tachycardia; therefore, conduction block determination was not subject to the same criteria as during premature stimulation. Electrogram morphologies were compared with electrogram features of activation and block obtained during premature stimulation. Multiple electrogram recordings were used in the analysis of bidirectional block during

4 Restivo et al High-Resolution Mapping 1313 S 1 15 msec S 2 p1 2 P2 T2 p P1p P2 T2 TI * * T2 l ventricular tachycardia so that the direction of the impulse on opposing sides of the block could be ascertained and synchronous alignment of electrotonic deflections, distal to block, could be ascertained. Though alternate explanations for the electrogram recordings shown in this study will be postulated (see "Discussion"), for the sake of clarity, recordings that follow the above guidelines will be defined as conduction block in "Results." Statistical Analysis Effective refractory periods and diastolic thresholds between adjacent (1 mm) recording locations at sites of block and the difference in effective refractory period between recording locations at sites of block both parallel (septal infarct border) and transverse (basal lateral infarct border) to the fiber axis were tabulated. The data were compared using Student's unpaired t test. A confidence level of 95% was considered statistically significant. Results Twenty-seven dogs that survived the ligation and surgery were studied. Reentry could be induced in 19 dogs by a single extrastimulus at one or both of the control stimulation sites; multiple extrastimuli were necessary for inducing rhythms in five dogs; reentry could not be induced in three dogs. The activation sequence of the reentrant pathway could be traced entirely on the epicardial surface in 2 hearts. In 24 hearts, a continuous arc of functional conduction block was detected during control premature stimulation. Eight reproducible sustained monomorphic ventricular tachycardias were induced in six dogs. Five of these tachycardias resulted in satisfactory activation P2 S1 145 msec S2 y FIBER AXIS FIGURE 2. Recordings showing effective refractory period determination. Left panel: An S1-S2 interval of 15 msec (effective refractory period plus 5 msec) applied to the test site (open circle) resulted in propagation from the test site, which was recorded at four locations (closed circles) surrounding the site (two parallel to the fiber axis [P1,P2] and two transverse to the fiber axis [T1,J2]). Right panel: An S1-S2 interval of 145 msec (effective refractory period) applied to the test site did not propagate to any of the four recording sites. maps in which the total reentrant cycle could be traced on the epicardial surface. High-Resolution Identification of Functional Conduction Block During Premature Stimulation The high-density bipolar electrode array was used for accurate determination of the spatial refractory gradients and their relation to the occurrence of conduction block. Figure 3 shows activation patterns from a heart in which there was no indication of abnormal conduction or block during S,. A single extrastimulus, S1-S2 of 18 msec, produced an arc of functional conduction block (shown by the heavy line). Reactivation did not occur at this coupling interval. The right portion of the figure shows a portion of a whole-ventricle activation map in which the latest S2 activation distal to the arc occurred at 11 msec. The high-density bipolar electrode plaque was then placed between sock electrode recording sites that were believed to exhibit functional conduction block according to defined criteria (see "Materials and Methods"). In this heart, the HDBEA was oriented with the electrode columns along the longitudinal axis of the epicardial muscle fibers and approximately perpendicular to the arc of block. The arrows in each electrogram indicate the end of the effective refractory period relative to S, activation at each site. After an S2 delivered from the septal margin, conduction block occurred between sites c and d (in Figure 3), which were spaced only 1 mm apart. The difference in effective refractoriness between these closely spaced sites was 35 msec. Since site d was still refractory during activation at site c, the premature wave front traveled around the block and conducted retrograde on the opposite side of the

5 1314 Circulation Research Vol 66, No 5, May 199 Si S 2 a., ~~ e A_ 23 1' After an S1-S2 coupling interval of 2 msec, the arc of block receded from the stimulating site (Figure 5). Conduction block occurred between sites e and f. An electrotonic deflection could be seen in electro- 3mV1 1 msec -/I'1 \ 1 FIGURE 3. High-density recordings across an arc offunctional conduction block. Right panel: Whole-ventricle activation maps ofbasic, S,, andpremature, S2, beats. Conduction block is not evident during S,, but after S2, a continuous arc of conduction block was formed. The high-density bipolar electrode array (HDBEA) was placed across the arc of block. The location of the HDBEA column, for the recordings shown in the left panel, is indicated. The star indicates the HDBEA stimulation location; the dots indicate sites a-e. Left panel: Electrogram recordings (1-mm interelectrode spacing) duping premature stimulation along one column of the HDBEA. The end of the effective refractory period for each site, relative to S, activation, is shown by the arrows. During S, stimulation, all deflections are smooth uniphasic signals. After a premature interval of 18 msec, S2 activation at site c preceded the expiration of refractoriness at site d and became blocked. There was a 35-msec difference in refractoriness, within 1 mm, at the site offunctional unidirectional conduction block. line of block. By the 7-msec isochrone, refractoriness had expired at sites d and e, and these sites were activated within 45 msec after activation at site c. This small difference in activation time prevented retrograde propagation across the arc from d to c. Figure 4 shows another experiment in which stimulation was applied near the septal border of the infarct. The left panel of Figure 4 shows a plot of refractoriness versus distance along one column of electrodes. The largest jump in refractoriness (25 msec) was between sites B and C. The right panel shows electrograms recorded along this row. During a basic cycle of 38 msec, the S, electrograms were smooth biphasic signals, devoid of notches or any evidence of discontinuous activity. Conduction following a premature interval of 17 msec resulted in conduction block between sites B and C. This is clearly illustrated by the electrotonic deflections in electrograms C and D (indicated by the stars), which occurred before the expiration of refractoriness at both sites. These simultaneous deflections corresponded in time with the latest activity before conduction block at site B and diminished in amplitude from the site of block. Since there was a diminished contribution to the signal from a receding impulse, the proximal electrograms, A and B,, became less biphasic and more uniphasic during S,. There was no indication of decremental conduction before block. Effect of Coupling Interval The effect of the premature coupling interval on the site of conduction block is illustrated in Figure 5. Electrode columns a-f at the top of the figure were oriented parallel to the activation wave-front direction. Expanded activation maps are shown in the lower portion of the figure; the HDBEA position, indicated by the shaded region, was fixed for a sequence Of S1-S2 intervals. After an S1-S2 coupling interval of 18 msec, conduction block occurred between sites c and d, with a refractory difference of 15 msec. The intrinsic deflection at site c occurred before the end of refractoriness at site d. Activation from the distal wave front arrived 6 msec later and conducted retrogradely to side d. Because the returning impulse arrived at site d 7 msec after the expiration of refractoriness at site c, retrograde conduction block occurred between sites d and c. Because of a diminished contribution to the signal from a receding activation wave front, the S2 deflection at sites c and d were largely monophasic. An electrotonic deflection simultaneous with the S2 activation potential at site d is seen in the electrogram from site c.

6 Restivo et al High-Resolution Mapping 1315 ERP versus 2 SITE OF BLOCK DISTANCE E A S, S2 lc 1 6 msec + 19 C 18 ICD B 165 msec f W 17 A B 15.,.. v A, DISTANCE (mm) FIGURE 4. Left panel: Effective refractory period (ERP) plotted as a function ofdistance from stimulation sites (A-E). The ERP ofthe stimulation site is shown at mm. After a premature interval of 17 msec, conduction block occurred between sites ofgreatest ERP difference (B and C). Right panel: Electrograms recorded at sites A-E. Activation at site B blocked to site C after S2. Electrotonic deflections (shown by stars) precede the expiration of refractoriness (shown by the vertical arrows) in electrograms C and D and are coincident in time with the activation potential ofsite B. Electrotonic potential in D is attenuated relative to C, which was closer to the site of block. gram f, which preceded the end of refractoriness for site f and reflected activation at site e. The arc was shorter than in the preceding example because of the longer premature interval. The activation wave front arrived distal to the block, at site f, 35 msec later. Electrotonus due to activation at site f is seen in electrogram e. At a longer premature interval of 25 msec, the activation map in Figure 5 shows that the arc of block had disappeared. Refractoriness at all sites, a-f, expired before the premature interval of 25 msec. The S2 electrograms were similar to S, activation. 16- Conduction Block Is Independent of Variations in Excitability and Direction of Impulse Propagation Effective refractory periods and diastolic thresholds were tabulated from 1-mm-spaced sites immediately proximal and distal to the conduction block during S2 stimulation. Figure 6 shows that the effective refractory period was significantly greater at sites distal to block ( msec [mean+sd]) than at proximal sites (197±32 msec [mean+sd]). On the other hand, mean thresholds distal to the block were generally lower compared with thresholds proximal to the block ( ma vs ma, 2 mv respectively); this difference, however, was not statistically significant. As further evidence that conduction block is due to conduction into a zone of prolonged refractoriness, stimulation was applied at opposite sides of the HDBEA. The left panel of Figure 7 shows electrograms during propagation in a normal to ischemic direction, from A to F. The electrograms show that after an S1-S2 interval of 25 msec, conduction block occurred between sites D and E, where the difference in refractoriness was 25 msec. The second S2 deflection of site E probably represents an activation potential, which accounts for the large electrotonic deflection of site F. The shortest premature interval that could be introduced for propagation in the opposite direction, from F to A, was 25 msec. The right panel of Figure 7 shows that when the impulse traveled down a refractory gradient, conduction was unimpaired. At this interval, the S2 electrograms were similar to S,. Conduction Block Occurs Parallel and Transverse to the Fiber Orientation Figure 8 shows HDBEA activation and refractory maps in which a wave front was propagated across

7 1316 Circulation Research Vol 66, No 5, May 199 Si 18 msec S2 S 2 msec S2 S1 25 msec S c;--- -c - 18 t i d t ~~~~25 1Oms Pc 15 F B E R A X S 1 F B A 2 FIGURE 5. Recordings and activation maps showing effect of coupling interval on the site of conduction block. Electrograms of sites a-f from one column of the high-density bipolar electrode array are shown for three premature intervals. Corresponding ventricular activation maps are shown in the lowerportion of each panel with the shaded areas indicating the location of the array column. Left panel: S1-S2 of 18 msec: Conduction block occurred between sites c and d, with an effective refractory period difference of 15 msec (end of effective refractory period indicated by the vertical arrows). Middle panel: S1-S2 of 2 msec: Conduction block occurred between sites e and f with an effective refractory period difference of 1 msec. Right panel: S1-S2 of 25 msec: Premature interval exceeds effective refractory period for all sites. Conduction block does not occur. the long axis of the fibers. The S, map (left panel) shows that conduction in this direction was slower than in the longitudinal direction. The isochrones were fairly uniform and equally spaced. After an S1-S2 interval of 18 msec, conduction was even slower and failed to conduct anterograde into the lower right region. The region distal to the block was activated 2-3 msec later by a circulating wave front. Had slow propagation occurred between the activation potentials occurring on either side of the block, a conduction velocity as low as.3 m/sec would have been needed. The middle and right panels show that the line of block (depicted by the heavy line) occurred longitudinal, transverse, and diagonal to the fiber axis, between sites of disparate refractoriness (35-6 msec). Conduction block, occurring in parallel and transverse fiber orientations, was examined at multiple locations along the arc of block. Figure 9 shows a heart in which reentry was initiated at an S1-S2 interval of 18 msec. Reactivation occurred within B E R 1 34 msec. The HDBEA was positioned parallel (location I) and transverse (location II) to the fibers at two locations that crossed the arc of block as determined from whole-ventricle mapping. Stimulation was applied to the HDBEA to produce activation wave fronts that propagated perpendicular to the control arc. At location I in Figure 9, conduction block probably occurred between sites d and e after an S1-S2 interval of 25 msec. The difference in refractoriness at the site of block was 7 msec. The S2 electrograms at sites a-c were biphasic and similar in morphology to the S, response. However, the S2 response at site d, immediately proximal to the site of block, was uniphasic due to the absence of a receding activation wave front. At location II, conduction block occurred between sites c and d after an S1-S2 interval of 185 msec. The difference in refractoriness between these sites was 3 msec. The peak of the electrotonic deflections at sites d, e, and f were not perfectly aligned, as in previous examples. This may have been due to the electrotonic f

8 Restivo et al High-Resolution Mapping 1317 E S cc 3 W 2 p < ± 32 T I 224 ± ±.37 NS C3 (.8 I- cc U- W2 -g.6 m m( Sr-.4 CY U. U. W-.2 Pf4RI DISTAL PROMUL DISTAL FIGURE 6. Bargraph showing effective refractoryperiods and diastolic thresholdfor 1-mm-spaced sites immediately proximal and distal to the site of conduction block. influence of the large propagation wave front proximal to the block and perpendicular to the electrode row. Had refractory values not been available, it may have been easily argued that slow conduction had occurred from sites c to d to e to f. Although the first deflections at sites d, e, and f might be interpreted as slowly conducted responses, they occurred before the expiration of refractoriness. However, at sites d, e, and f, a second, more rapid potential can be seen that follows the expiration of refractoriness. Therefore, these initial deflections were considered electrotonic. Had saltatory conduction occurred between sites c and d, an apparent conduction velocity of.15 m/sec would be indicated. The difference in refractoriness between sites of functional conduction block at the septal and basal lateral border of the infarct were determined at 132 sites of block in 22 dogs. Impulses were propagated. parallel to the fibers at the septal locations and transverse to the fibers at basal lateral locations; stimulation was applied from the HDBEA. The difference in refractoriness between 1-mm-spaced sites, immediately proximal and distal to the block, was significantly greater (p<.1) for impulses traveling along the fibers at the septal border of the infarct (34±29 msec) than at the basal lateral border during impulse propagation across the fibers (19±13 msec). High-Density Recording During Sustained Monomorphic Ventricular Tachycardia Figure 1 shows an experiment in which the HDBEA was placed at various locations along one of the arcs of block during a sustained ventricular tachycardia initiated by programmed premature stimulation. The activation pattern of the reentrant tachycardia was in the form of two wave fronts A B c D E F A B c D E F FIGURE 7. Recordings during S1-S2 stimulation at sites A-F 175 msec t Conduction block is independent of the tapering geometry of the surviving layer. Left panel: Prop- 185 msoc +agation in a normal to ischemic 195 misec (thicker to thinner) direction. Conduction block occurs between 4 sites D and E, with a refractory 25 msec difference of 25 msec (end of effective refractory period is indi- 23 msec cated by the vertical arrows). Right panel: Conduction block 25 msc does not occur forpropagation in * the opposite direction (thinner to 25 mcc T thicker). S, S2

9 1318 Circulation Research Vol 66, No 5, May 199 Si * * 25/3 FIBER S2 AXIS ERP x 22 x FIGURE 8. Activation and refractory maps of high-density bipolar electrode array. S]-S2 stimulation at 18 msec was applied across the longitudinal axis ofthe epicardialfibers. Sites of stimulation are shown by stars; dots indicate bipolar recording sites; arrow indicates wave-front direction. After S2, a line of conduction block (shown by the heavy line) occurred longitudinal, transverse, and diagonal to the epicardial fiber axis between sites of disparate refractoriness (ranging between 35 and 6 msec). ERP, effective refractory period. circulating around two arcs of functional conduction block. The two wave fronts coalesced once every cycle (shown in the map at the 12-msec isochrone) and conducted slowly through the ischemic zone. The isochrones within the common reentrant pathway were oriented perpendicular to both arcs of block. The HDBEA was manually positioned at multiple locations along the upper arc of block. Electrode site a was located proximal to the arc and common reentrant pathway. Site f (site e for location I) was located distal to the arc, within the common reentrant pathway. The lower arc of block was located near the apex of the heart and was inaccessible for adequate placement of the HDBEA. The upper portion of Figure 1 shows the electrograms along one column of bipolar electrodes for each HDBEA location. HDBEA locations II and III were situated near the center of the arc of block. At location II, conduction proximal to the arc was blocked between sites d and e, but site e failed to show an electrotonic potential. On the other hand, the distal activation impulse failed to conduct to site d, and an electrotonic deflection corresponding to activation at site e was seen in electrogram d. The isochronal activation difference of 61-8 msec corresponded to an activation time difference between proximal and distal electrogram deflections of 75 msec. At location III in Figure 1, the isochronal activation difference (81-1 msec) was larger compared with location IL and corresponded to an activation time difference between the proximal and distal electrogram deflection of 85 msec. In contrast to location II, the exact site of conduction block for both the proximal and distal wave fronts could not be precisely established because both deflections in electrogram d were equivocal. Conduction block of the proximal wave front could have occurred between sites c and d, in which case the initial deflections in sites d and e could have represented electrotonic potentials. However, it could be argued that conduc- tion block occurred between sites d and e, in which the initial deflection at side d would have represented an activation potential. Similar arguments apply for determining the exact site of block for the retrograde wave front, which could have occurred between sites e and d or sites d and c. Since the interval between the two deflections at site d was 85 msec, activation could not have occurred twice at that site. It is possible that both deflections represented electrotonic potentials, in which case the electrode at site d could have been overlying an inexcitable arc of block of finite dimension in which both deflections reflected bidirectional block from cells within 1 mm of the recording site. Location I in Figure 1 was situated at the septal end of the arc of functional conduction block. A difference between proximal and distal activation potentials of 3 msec corresponded to an isochronal difference of 21-4 msec. Conduction block of the proximal impulse probably occurred between sites d and e; an electrotonic deflection can be seen in electrogram e. Conduction block of the distal wave front appeared to have occurred between sites d and c. This interpretation suggests, however, that both deflections at site d represent an activation potential. This could only be possible if the arc of conduction block was situated between the two poles of electrode d. Such an observation will be illustrated in more detail in Figure 12. The recordings from site IV in Figure 1 suggest that the arc of functional conduction block may have been longer than predicted by our interpretation by whole-ventricle mapping. Since the interval between deflections is small, an isoelectric segment could not be discerned as in sites III and IV. Although slow conduction at this location cannot be discounted, there were two distinct deflections in each electrogram in the immediate vicinity of the arc of block. Electrograms recorded a few millimeters away from the arc (i.e., a, e, and f) showed predominantly single

10 I a S1 26 mc S S 2 Restivo et al High-Resolution Mapping d ~ ~ ~~~~~~~~~~~~~~~~8 lated to fiber orientation. The upper arc was approx- f 35 2 mv S1 186 "'eec S2 I~~~ dol 186me FIGURE 9. Recordings (left panel) and activation maps (right panel) showing conduction block occurring parallel and transverse to the fibers. The high-density bipolar electrode array was positioned at two locations along a continuous arc offunctional conduction block detected by whole-ventricle mapping. Location of stimulation site for each array is shown by the stars; dots indicate bipolar recording sites (a-f). Expiration ofrefractoriness, relative to S,, is indicated by the vertical arrows. At location I (septal), the electrode columns were oriented parallel to the longitudinal axis of the epicardial fibers and approximately perpendicular to the arc of block; propagation was initiated along the fibers. At location II (basal lateral), the electrode columns were oriented transverse to the longitudinal axis of the epicardial fibers and approximately perpendicular to the arc of block; propagation was initiated across the fibers. Expiration ofrefractoriness is indicated by arrows. An S1-S2 interval of25 msec, applied at location I, produced conduction block between sites d and e, with an effective refractoryperiod difference of 7 msec. An S-S2 interval of185 msec, applied at location II, produced conduction block between sites c and d, with a difference in effective refractory period of 3 msec. LAD, left anterior descending coronary artery. deflections that corresponded with the appropriate complicated arrangement of block from another isochronal intervals. heart during sustained monomorphic tachycardia. In the previous example, the arc of conduction During the tachycardia in this heart, there were three block was approximately parallel to the longitudinal isolated arcs of block, whose position appeared unreaxis of the epicardial fibers. Figure 11 shows a more

11 132 Circulation Research Vol 66, No 5, May 199 l II a Ill a lv a a b b b b c c d d d d I-looms.c4 f f ECG >mv]mv FIGURE 1. Recordings of high-density bipolar electrode array at multiple locations (I-IV) along a continuous arc offunctional conduction block during sustained monomorphic ventricular tachycardia (top panel). A polarprojection ofisochronal activation map is shown in the lower left panel; the perimeter represents the atrioventricular anulus with the apex at the center of the map. Arrows indicate wave-front direction during sustained ventriculartachycardia. Both arcs are oriented approximatelyparallel to the longitudinal axis of the epicardial musclefibers. A portion ofthis activation map is shown in the lower right panel. The shaded rectangles represent the column for each array location. Electrograms recorded in proximity to the arc of block show split electrograms composed of two discrete potentials separated by a variable isoelectric interval: one deflection represents local activation; the other deflection is an electrotonic potential reflecting activation recorded 1 mm away. The interval between the two deflections is greatest at the center ofthe arc (locations II and III) where the difference in isochronal activation, by whole-ventricle mapping, is greatest. The interval between the two deflections is less as the reentrant impulse circulates around the end ofthe arc (location I). The electrographic characteristics offunctional conduction block are the same at location IV which indicates that the arc offunctional conduction block was longer than that predicted by the whole-ventricle mapping technique. imately parallel to the longitudinal axis of the fibers at both the septal and lateral ends; the midportion of the arc was approximately perpendicular to the longitudinal axis of the fibers. The activation wave front, which circulated around the septal end of the arc of block, was very slow, taking almost 1 msec to reach the opposite side of the arc. The electrode plaque was placed under the sock in this vicinity as shown in the right panel of Figure 11. Figure 12 shows HDBEA recordings from the septal end of the upper arc of conduction block shown in Figure 11. A high-density activation map, drawn at 1-msec isochronal intervals, is shown in the lower right panel. During the first isochrone of

12 the HDBEA map, which was on the distal side of the arc of block and corresponded to propagation within the common reentrant wave front, activation occurred within a relatively short period of time. By contrast, the activation wave front proximal to the arc of block was markedly slowed. Similar to previous examples, Figure 12 shows electrograms recorded on either side of the arc with two deflections, which generally corresponded to electrotonus and local activation. However, the location of the plaque in this experiment was such that the poles of some electrodes were bisected by an arc of block. Electrodes lb-lb show two distinct deflections corresponding to activation times ranging from 4 to 6 msec in a 4-mm2 region. In the left pole of electrode pairs from column 1, the impulse propagated from 1F to 1B. Electrode 1A was uniphasic and represented the turnaround point. The second deflections in electrograms 1B-lE reflect activity detected in the right-hand poles of electrodes from column 1 and showed a reversal in the direction of impulse propagation along the opposite side of the arc of block. The largest difference in activation time (56 msec) was between the poles of 1E. A small electrotonic deflection corresponding to the second activation could be seen in 1F. Electrograms recorded from columns 2 and 3 (upper right and lower left, respectively, of Figure 12) indicate that impulses from either side were propagating toward the arc of block. Electrograms 2D and 3D, which were recorded near the arc of block, were of lower amplitude (.24 mv and.21 mv, respectively). Similar to the example shown in Figure 11, the choice of an activation potential from either of the two deflections may be equivocal, and these sites may have represented a thin, inexcitable region of bidirectional block. Discussion Based on our analysis of electrograms and our determination of effective refractory periods, the Restivo et al High-Resolution Mapping 1321 FIGURE 1 1. Whole-ventricle activation map (left panel) and location block of high-density bipolar electrode array (right panel) during sustained monomorphic ventricular tachycardia. Tachycardia cycle length was 22 msec. Arrows indicate wavefront direction during sustained ventricular tachycardia. Three arcs (thick lines) detected during tachycardia occurred both parallel and transverse to the longitudinal axis ofthe fibers. A portion ofthe activation map is shown in the right panel. The array was placed at the septal end of the upper arc offunctional conduction block, as indicated by the shaded square. results presented here indicate that in the subacute phase of canine myocardial infarction, functional unidirectional conduction block occurs between sites of disparate refractoriness after premature stimulation. A difference of 1 msec in effective refractory periods between sites spaced 1 mm apart was sufficient for the occurrence of functional conduction block. In this model, conduction block during premature stimulation and sustained ventricular tachycardias appeared to occur suddenly, independent of the direction of impulse conduction and without a progressive decrement of impulse conduction before block. Methodological Considerations Effective refractory period detennination. Differentiation of conduction versus conduction block, during premature stimulation, depended on the validity of the effective refractory period determination. Though the technique of effective refractory period determination used here has some limitations, it remains the best possible method for determining the responsiveness of small regions of heterogeneous ischemic myocardium. Because of the heterogeneity of the preparations and the uncertainty of the conduction path to recording sites even 1 mm away, functional refractory periods would have been impossible to determine. Although strength-interval relations provide additional information and encompass changes in excitability,15 their determination is technically unfeasible, considering the large number of sites tested in each experiment, and they are not necessarily better indicators of the end of the refractory period. For reproducible measurement of up to 128 sites per location of the array, we chose to use a test current of two times diastolic threshold. Considering the magnitude of the refractory differences measured and the fact that threshold was reasonably uniform along the test columns, the possibility of small errors due to the method used here seemed negligible. To

13 1322 Circulation Research Vol 66, No 5, May 199 ECG 1A.62 1B.5 1C.53 1D.63 1E.64 1F.56 ECG 3A 85 3B.58 3C.57 3D.21 3E.42 3F.41 A,/'- //\ A--A-- A-- J An-Av=> 1 k,. be sure that a small amount of myocardium was being measured, a small-diameter (125 -gm) electrode material was selected in which only the uninsulated tips of each pole (which were polished flush before each experiment) were in contact with the tissue. Since the diastolic threshold was low (.687±.32 ma) and fairly uniform along the test rows, it was unlikely that the test current engaged cells distant from the test site14 or was shunted through the extracellular fluid. In a recent review, it was suggested that there are inherent inaccuracies in refractory period determination by external stimulation.16 The basis of the argument is whether punctate stimulation, at two times diastolic threshold, realistically simulates the same current source of a broad-depolarization wave front. This has posed, for many years, a theoretically difficult problem because it requires characterization of all intracellular potentials and the electrical pathways between the cells. To determine the safety factor for propagation at the test site, the theoretical A B C D E F ECG 2A.67 2B 52 2C.56 2D.24 2E.5 2F.67 F B ER * 75* S 6 55 * * * 1 *. 4 5 * * 1 S * 1 * * 8 * * X S 2 2 msec -1 _. P--. FIGURE 12. High-density mapping of conduction around the end of an arc of block during sustained ventricular tachycardia. The high-density recordings consisted of 18 electrodes arranged in three columns (1-3) of six bipolar pairs (A-F) and was positioned at the location indicated in Figure 11. Lower right panel: High-density activation map. Conduction delay of 56 msec between poles of electrode ie is due to impulse delay around the end ofan arc offunctional conduction 3 block. Pathway for the delay was contained within 4 mm2. Upper left panel: 85 Split electrograms comprised of two activation potentials recorded from each pole of the bipolar electrode. Variable interval between deflections results from impulse *6. delay during conduction around the end of the arc of block. Upper right panel and 8 * lower left panel: Split electrograms comprised ofactivation and electrotonicpoten- 8 * tials, similar to previous examples threshold stimulus strength of the depolarizing wave front must be known. At the present time there are no data available relating the current source of a broad wave front to an externally applied current. Ursell et al17 have shown that cells from the surviving layer of infarcted canine hearts can become widely separated, with varying degrees of interstitial edema and fibrotic infiltration, which may affect the flow of current between cells in these regions. Spear et al'8 have measured decreased space constants in infarcted epicardium. A decreased space constant would indirectly indicate an increase in the effective input resistance and a reduction in the current necessary to bring the cell to threshold. High-resolution mapping: Importance of recording technique. Bioelectric signals are influenced by the recording method (i.e., electrode type, filter settings, and interelectrode spacing).19,2 Depending on the recording technique employed, fragmented multicomponent and long-duration electrograms can be recorded from diseased hearts and have been used as

14 indirect evidence suggestive of reentry. Fragmented multicomponent and long-duration electrograms have been recorded from composite electrodes placed over large areas of infarcted ventricle,13 ventricular endocardial surface in humans using wide bipolar (1-mm interpole) catheters,21 the epicardial border zone of infarcts using 1-mm disk electrodes (2-mm interpole),5 epicardial tissue from healed infarcts studied in vitro,22 epicardial tissue from occlusion-reperfusion infarcts during sustained rhythms studied in vitro,23 and human atrial bundles infiltrated with collagenous septa.24 The interpretation of fragmented long-duration electrograms has been criticized largely because of an inability to sequentially trace the impulse.2,7 Spach et a124 cautioned that multiple deflections can arise from bundles of varying diameter and have suggested that a resolution of 2,um is necessary to uncover the fine details of the activation sequence causing multicomponent electrograms. We chose a resolution of 1 mm because of the combined constraints of the number of available multiplexer channels and total area of myocardium to be simultaneously mapped. Since extracellular electrograms are complex, we chose to have the minimal amount of signal processing caused by the recording methodology. We used wideband filtering (.5-25 Hz) and small electrode size (125 gm). Even with these precautions, there were cases, such as seen in Figure 12, in which the poles of a bipolar electrode actually spanned the line of block. In the electrograms from column 1 of Figure 12, the duration of the impulse that could be sequentially traced was 56 msec and was within an area of only 4 mm2. If the poles of the electrode were of a larger dimension, the recording from this area would have resulted in a long-duration multicomponent fragmented electrogram, and the details of the activation sequence could not have been discerned. The highresolution mapping technique employed here may explain the discrepancies, in the same animal preparation, between our results and others.5,25 We find that to attribute multiple deflections to saltatory conduction or conduction block, each deflection must be correlated with nearby activation by multiple recordings in proximity. Attributing individual deflections to local activation or conduction block can be further enhanced if the refractory period of the site is known. Spatial Gradation of Effective Refractory Pertiods In a previous study,1 the spatial patterning of refractoriness in the 4-day postinfarction canine heart was determined. Refractory gradients were interpolated from contour maps of refractoriness from both ventricles. Based on measurements spaced 5-1 mm apart, a linear refractory gradient of 2 msec/cm was suggested as a threshold for the occurrence of functional conduction block.1 The highresolution techniques employed in the present study showed the gradation of refractoriness to be much steeper, occurring over distances of 1 mm or less. Restivo et al High-Resolution Mapping 1323 At the present time we cannot ascertain the specific reasons for the apparent jump in refractoriness. We were unable to find any relation between gross infarct geometry with both the apparent jump in refractoriness or the occurrence of functional conduction block. Myocardial infarction in this canine model is characterized by a surviving epicardial layer of electrophysiologically abnormal cells. This surviving layer is wedge shaped and tapers over the core of the infarct.4 Prolonged refractory periods were measured within this region. Along the septal border of the infarct, there was usually a rapid transition in the thickness of the surviving epicardial layer.4 This transition region was encountered by impulses traveling along the fiber axis; this region was also where the greatest refractory differences were measured. Though specific anatomic abnormalities that account for the local disparities in refractoriness have not been found, it is known that the area overlying the infarction receives a reduced coronary blood flow.26 There is evidence that the boundary between normal and abnormally perfused cells is not gradual but discrete It may be possible that these discrete boundaries may account for the sharp discontinuities (refractory period differences up to 12 msec within 1 mm at the septal infarct border) measured between the sites that were blocked. Careful histological and biochemical studies that take into account differences in coronary circulation or morphometric analysis may be necessary to provide the answer to these questions. Characteristics ofarcs of Functional Conduction Block During premature stimulation, multiple highdensity bipolar recordings along arcs (as defined by whole ventricular mapping) and refractory period measurements indicated that functional conduction block occurred along a continuous line. The characteristics of electrograms recorded during ventricular tachycardia also suggested to us that conduction block also occurs along a continuous line. Though technical limitations prevented simultaneous highresolution mapping at all sites along the arc of block, in the many random placements of the electrode array, we found no indication of an interruption of what we considered an arc of functional conduction block. Recordings from arcs of conduction block invariably showed paired split electrograms and not multicomponent long-duration electrograms. These split electrograms were comprised of a local activation potential and an electrotonic deflection representing activation 1 mm away and were separated by an isoelectric interval. Conversely, electrograms recorded on the opposite side of the arc showed the reverse order of activation and electrotonus. It is conceivable that these paired deflections do not reflect activation and conduction block potentials. It is possible that the sites were activated sequentially, with either extremely prolonged or saltatory conduction delay. This would have required conduction

15 1324 Circulation Research Vol 66, No 5, May 199 velocities as low as.1 m/sec localized to a 1- mm-wide region. It is our interpretation that the increasing interval between the two deflections from the ends to the center of the arc, the opposing direction of the impulse on either side of the arc, and the correlation of the activation potentials with large isochronal zones on either side of the arc of block provide strong evidence for a continuous arc of functional conduction block. It has been suggested that large differences in activation time between the lower density recording sites used in many previous mapping studies may not necessarily indicate conduction block.5 Dillon and coworkers5,25 have proposed the concept of pseudo-conduction block due to structural abnormalities of the infarct border zone as a mechanism of sustained ventricular tachycardia. The authors proposed that during reentrant activation, centripetal impulses were conducted very slowly, within long, narrow zones, across the longitudinal fiber axis. We failed to record multicomponent electrograms during both premature stimulation and during sustained ventricular tachycardia. However, it should be noted that the primary locations of our recordings were along arcs of block at the septal and basal lateral borders of the infarct during premature stimulation and that technical limitations prevented us from recording from the apical arc during sustained "figure of 8" reentry. Thus, we cannot exclude the possibility that regions of discontinuously activated myocardium occur in other regions, where the disparity in activation time across the arc is less. There is further evidence for functional conduction block in this model: 1) The site of block could be shifted by varying the premature interval, as shown in Figure 5; similarly, the location and extent of the arcs of block, as analyzed by whole-ventricle mapping, have been shown to be cycle-length dependent.1 2) Shifting of arcs of block from one stable location to another has been demonstrated by resetting or entrainment of sustained reentrant ventricular tachycardias.3 3) The dependence of the location and extent of arcs of conduction block on the spatial distribution of recovery of excitability has been further confirmed by modifying recovery time distributions by dual S, stimulation.31 These observations would be highly unlikely if the arcs of functional conduction block represented zones of extremely slow conduction and were related only to structural abnormalities of an infarcted myocardial matrix. Mechanisms of Functional Conduction Block Functional conduction block can occur in a variety of circumstances. Block may be due to 1) abrupt changes in cardiac geometry,32,33 2) decremental conduction leading to propagation failure,34,35 3) differences in conduction properties relative to fiber orientation,35 or 4) regional differences in refractory period.1,36-38 The aim of this study was to investigate a functional obstacle (a continuous arc of block) that is necessary to support reentrant activation. In our experiments, we found that regional differences in refractory periods were the overriding cause of functional conduction block during the initiation of reentrant activation. Although the remaining mechanisms were not observed, they may be operative in other regions unrelated to the functional obstacle supporting reentrant activation. As mentioned earlier, the site of functional conduction block in our experiments was not dependent on propagation into an epicardial layer of tapering geometry. Further evidence against a change in geometry as the underlying mechanism was the fact that the site of block could be shifted by varying the coupling interval of the premature beat (see Figure 5). A sudden change in geometry would, in fact, favor block during propagation from the thin surviving layer to the normal myocardium.3233 Yet, conduction block always occurred during propagation in a normal to ischemic direction. As shown in Figure 7, conduction block did not occur even when the shortest coupled impulse was propagated from the ischemic layer. Since the magnitude of the depolarizing current source from active cells becomes progressively attenuated at shorter intervals during the relative refractory period, one may argue that conduction block is simply due to the reduced current source of active cells encountering cells with more depressed membrane properties. This is unlikely for two reasons. First, a gradual slowing of the impulse leading to conduction failure (i.e., decremental conduction) was not observed. Second, the proximal impulse did not encounter zones of increasingly depressed excitability because there was no difference in diastolic threshold between sites that were blocked. Functional conduction block occurred only when the impulse arrived at a region that was still refractory. The magnitude of the refractory barrier appeared to be responsible for the abrupt conduction block and prevented increasingly delayed slow responses to propagate along the gradient of refractoriness. A difference in safety factor relative to fiber orientation and propagation direction has been suggested by Spach et a135 as a mechanism of functional conduction block. After infarction, there may be disturbances in electrical connections between cells, which accentuate anisotropic electrophysiological properties. Though anisotropic conduction properties might be considered important in the surviving epicardial layer, in which one orientation of myocardial fibers may remain, we have shown that conduction block in this layer occurred between sites of disparate refractoriness, both across and parallel to the fiber axis. As shown in Figure 9, even when activation was initiated only a few millimeters away, conduction block also occurred in the same location as control regardless of the direction of the wave front. Thus, regardless of the orientation of the fibers or the propagation direction, we found no evidence that conduction block along the arc occurred without a corresponding abrupt change in refractoriness. It is interesting that Spach et a139 have recently reported that spatial inhomogeneities in

16 repolarization relative to fiber orientation may influence conduction block formation. Though the results presented here indicate that functional conduction block is due to disparate refractoriness, prolongation of refractoriness in the ischemic zone is not accompanied by a concomitant increase in action potential duration. Postrepolarization refractoriness has been observed in ischemic Purkinje and myocardial fibers by which full recovery of responsiveness exceeds the action potential duration.4-43 Postrepolarization refractoriness can be attributed to abnormal membrane properties, resulting in prolonged repriming of ionic channels, and/or abnormalities that affect the intercellular connections.41,42 Intracellular recordings from the surviving ischemic epicardial layer overlying 3-5-day-old canine infarcts show that the cells have variable degrees of partial depolarization, reduced action potential amplitude, and decreased upstroke velocity.17,43,44 Though an abnormal sodium channel has been implicated,434" the altered membrane properties of ischemic myocardium that may result in postrepolarization refractoriness have not been fully explored. The consequences of infarction are likely to affect the passive electrical properties of ischemic myocardium. Although cells from the surviving layer appear normal upon microscopic examination, there is considerable infiltration and separation of individual muscle fibers.4,5"17,25 Spear et al18 have reported that there is a reduction in the space constant of ischemic epicardium. It may be tempting to associate conduction abnormalities with disruptions in intercellular connections after infarction. However, in a preliminary report45 we found that refractoriness in the surviving epicardial layer decreased significantly with time (4-8 weeks) after infarction, a time when the derangements of the interconnective cellular matrix become increasingly prominent.17'22 Normalization of refractoriness in this period corresponded with a normalization of action potential characteristics and an inability to induce reentry in the epicardial layer. The relative contributions of abnormal membrane characteristics and disturbances in the intracellular connections to the prolongation of refractoriness are complex and must await further investigation. Restivo et al High-Resolution Mapping 1325 Patterns ofactivation and Functional Conduction Block During Ventricular Tachycardia Refractoriness could not be measured during reentrant ventricular tachycardia. As in the initiation of reentry, functional conduction block during sustained reentrant activation occurred along continuous lines, as indicated by discrete split electrograms of variable intervals. The morphological characteristics of electrograms recorded during ventricular tachycardia were strikingly similar to those recorded during the initiation of reentry. In studies from Cardinal et a18 and Dillon et al,5 activation impulses during sustained reentry were commonly observed to circulate around narrow zones oriented parallel to the long fiber axis. During sustained reentrant activation, arcs of block were usually repositioned from their location during premature stimulation and the first few unstimulated beats.3'4 The arcs usually shifted toward the center of the surviving epicardial layer during the sustained rhythm. Refractory gradients for S, were significantly greater along the septal border of the infarct (34±29 msec/mm) compared with the basal lateral margin (19±13 msec/mm). During S2, the arc of block occurred along both borders. However, during sustained reentrant activation, arcs of conduction block rarely occurred along the septal border and were usually oriented parallel to the fiber axis, where smaller refractory differences were measured. This shift has been attributed to the differential shortening of refractoriness during successive short cycles of the sustained rhythm.46 The orientation of the arcs during ventricular tachycardia and the low incidence of arcs along the septal border may also be attributed to the distribution of refractoriness in the heart preceding the tachycardia. During initiation, we found that the arc of functional conduction block usually occurred in a horseshoe pattern along both borders and that the site of breakthrough usually occurred at the septal border.3'4'8 Since the longest refractory periods were recorded in these regions, they were activated late during the initiating cycle. During the first reentrant cycle, activation traveled down the gradient of refractoriness. The preponderance of arcs along the fibers can then be explained by the bidirectional invasion of these zones by impulses from the common reentrant pathway and the proximal zones on either side. In other words, when the arc had stabilized, activation of zones along one side of the arc prevented penetration by impulses on the opposite side of the arc. Electrotonic influences, by constant bidirectional invasion of these impulses, may create finite zones of constant block, as suggested by Figures 1 and 12, similar to those reported by others.44247,48 It is possible that the location of arcs of block during sustained tachycardia is a result of the combined influences of differential refractoriness and the conduction pattern of the circuit that maintains them. If the location of arcs of block during the reentrant tachycardia is determined by the refractory pattern, which, in turn, is determined by the particular model of infarction, then the preponderance of arcs parallel to the longitudinal fiber axis in this model may be coincidental. Although we found no evidence for anisotropy as a mechanism of conduction block during the initiation of reentry, we cannot exclude the possible influence of directional differences in conduction velocity to formation of arcs of functional conduction block during sustained reentrant activation. In summary, continuous arcs of functional conduction block were found to be a necessary prerequisite for both the initiation and sustenance of reentrant excitation in the subacute phase of canine myocardial infarction. Functional conduction block, necessary for the initiation of reentry, was due to abrupt

17 1326 Circulation Research Vol 66, No 5, May 199 changes in refractoriness, occurring over distances equal to or less than 1 mm. Acknowledgments We thank Gerald Cohen and his staff in the animal laboratory for their assistance and Connie Mead of Medical Media for the photography. We thank Dr. Dennis White for his expert advice with regard to the design of the electrode array and Kaveh Gooyandeh for constructing the electrodes. We thank Dr. Kazem Fani for the histology. We also wish to thank Drs. Celvin Williams, Keui-Meng Wu, Wolfgang Schoels, Mohamed Boutjdir, and Raphael Henkin for their invaluable assistance and advice. References 1. Mines GR: On dynamic equilibrium in the heart. J Physiol (Lond) 1913;46: El-Sherif N, Smith RA, Evans K: Canine ventricular arrhythmias in the late myocardial infarction period: 8. Epicardial mapping of reentrant circuits. Circ Res 1981;49: El-Sherif N, Mehra R, Gough WB, Zeiler RH: Ventricular activation patterns of spontaneous and induced ventricular rhythms in canine one day old myocardial infarction: Evidence for focal and reentrant mechanisms. Circ Res 1982;51: Mehra R, Zeiler RH, Gough WB, El-Sherif N: Reentrant ventricular arrhythmias in the late myocardial infarction period: 9. Electrophysiologic anatomic correlation of reentrant circuits. Circulation 1983;67: Dillon SM, Allessie MA, Ursell PC, Wit AL: Influences of anisotropic tissue structure in reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res 1988; 63: Pogwizd SM, Corr BP: Reentrant and nonreentrant mechanisms contribute to arrhythmogenesis during early myocardial ischemia: Results using three-dimensional mapping. Circ Res 1987;61: Wit AL, Allessie MA, Bonke FIM, Lammers W, Smeets J, Fenoglio JJ: Electrophysiologic mapping to determine the mechanism of experimental ventricular tachycardia initiated by premature impulses: Experimental approach and initial results demonstrating reentrant excitation. Am J Cardiol 1982; 49: Cardinal R, Vermeulen M, Shenasa M, Roberge F, Page P, Helie F, Savard P: Anisotropic conduction and functional dissociation of ischemic tissue during reentrant ventricular tachycardia on canine myocardial infarction. Circulation 1988; 77: Chen PS, Wolf PD, Dixon EG, Daniely ND, Frazier DW, Smith WM, Ideker RE: Mechanism of ventricular vulnerability to single premature stimuli in open-chest dogs. Circ Res 1988;62: Gough WB, Mehra R, Restivo M, Zeiler RH, El-Sherif N: Reentrant ventricular arrhythmias in the late myocardial infarction period in the dog: 13. Correlation of activation and refractory maps. Circ Res 1985;57: Restivo M, Gough WB, Wu KM, Williams C, El-Sherif N: Role of abrupt changes in refractoriness and fiber orientation in the formation of functional conduction block (abstract). Circulation 1987;76(suppl IV):IV Restivo M, Gough WB, El-Sherif N: Correlation of abrupt changes in refractoriness and functional conduction block by high resolution electrode recordings in the post infarcted canine heart (abstract). JAm Coll Cardiol 1987;9: El-Sherif N, Sherlag BJ, Lazzara R, Hope RR: Reentrant ventricular arrhythmias in the late myocardial infarction period: 1. Conduction characteristics in the infarction zone. Circulation 1977;55: Sepulveda NG, Roth BJ, Wikswo JP Jr: Current injection into a two-dimensional anisotropic bidomain. Biophys J 1989; 55: Michelson EL, Spear JF, Moore EN: Strength-interval relations in a chronic canine model of myocardial infarction: Implications for the interpretation of electrophysiologic studies. Circulation 1981;63: Janse MJ, Wit AL: Electrophysiological mechanisms of ventricular arrhythmias resulting from myocardial ischemia and infarction. Physiol Rev 1989;69: Ursell PC, Gardner PI, Albala A, Fenoglio JJ, Wit AL: Structural and electrophysiological changes in the epicardial border zone of canine myocardial infarcts during infarct healing. Circ Res 1985;56: Spear JF, Michelson EL, Moore EN: Reduced space constant in slowly conducting regions of chronically infarcted canine myocardium. Circ Res 1983;53: Ideker RE, Mirvis DM, Smith WM: Late fractionated potentials. Am J Cardiol 1985;55: Waxman HL, Sung RJ: Significance of fragmented ventricular electrograms observed using intracardiac recording technique in man. Circulation 198;62: Josephson ME, Horowitz LN, Farshidi A: Continuous local electrical activity: A mechanism of recurrent ventricular tachycardia. Circulation 1978;57: Gardner PI, Ursell PC, Fenoglio JJ Jr, Wit AL: Electrophysiologic and anatomic basis for fractionated electrograms recorded from healed myocardial infarcts. Circulation 1985; 72: Richards DA, Blake GJ, Spear JF, Moore EN: Electrophysiologic substrate for ventricular tachycardia: Correlation of properties in vivo and in vitro. Circulation 1984;69: Spach MS, Dolber PC: Relating extracellular potentials and their derivatives to anisotropic propagation at a microscopic level in human cardiac muscle: Evidence for electrical uncoupling of side to side fiber connections with increasing age. Circ Res 1986;58: Dillon SM, Ursell PC, Wit AL: Pseudo-block caused by anisotropic conduction: A new mechanism for sustained reentry (abstract). Circulation 1985;72(suppl III):III Hirzel HO, Nelson GR, Sonnenblick EH, Kirk ES: Redistribution of collateral blood flow from necrotic tp surviving myocardium following coronary occlusion in the dog. Circ Res 1976;39: Axford-Gately RA, Wilson GJ: The 'border zone' in myocardial infarction: An ultrastructural study in the dog using an electrondense blood flow marker. Am J Pathol 1988;131: Factor SM, Okun EM, Kirk ES: The histological lateral border of acute canine myocardial infarction: A function of microcirculation. Circ Res 1981;48: Okun EM, Factor SM, Kirk ES: End capillary loops in the heart: An explanation for discrete myocardial infarctions without border zones. Science 1979;26: El-Sherif N, Gough WB, Restivo M: Reentrant ventricular arrhythmias in the late myocardial infarction period: 14. Mechanisms of resetting, acceleration, or termination of reentrant tachycardia by programmed electrical stimulation. PACE 1987;1: Restivo M, Gough WB, El-Sherif N: Reentrant ventricular arrhythmias in the late myocardial infarction period: prevention of reentry by dual stimulation. Circulation 1988;77: Sasyniuk BI, Mendez C: A mechanism for reentry in canine ventricular tissue. Circ Res 1971;28: Mendez C, Mueller WJ, Meridith J, Moe GK: Interaction of transmembrane potentials in canine Purkinje fibers and at Purkinje fiber-muscle junctions. Circ Res 1969;24: Cranefield PF, Wit AL, Hoffman BF: Conduction of the cardiac impulse: III. Characteristics of very slow conduction. J Gen Physiol 1972;59: Spach MS, Miller WT III, Dolber PC, Kootsey JM, Sommer JR, Mosher CE Jr: The functional role of structural complexities in the propagation of depolarization in the atrium of the dig: Cardiac conduction disturbances due to discontinuities of effective axial resistivity. Circ Res 1982;21:

18 36. Allessie MA, Bonke FIM, Schopman FJG: Circus movement in rabbit atrial muscle as a mechanism of tachycardia: II. The role of nonuniform recovery of excitability in the occurrence of unidirectional block as studies with multiple microelectrodes. Circ Res 1976;39: Boineau JP, Schusser RB, Mooney CR, Miller CB, Wylds AC, Hudson RD, Borreman MJ, Brockus CW: Natural and evoked atrial flutter due to circus movement in dogs: Role of abnormal atrial pathways, slow conduction, non-uniform refractory period distribution and premature beats. Am J Cardiol 198; 45: Levites R, Banka VS, Helfant RH: Electrophysiologic effects of coronary occlusion and reperfusion: Observations of dispersion of refractoriness and ventricular automaticity. Circulation 1975;53: Spach MS, Dolber PC, Heidlage JF: Interactive roles of inhomogeneities of repolarization and anisotropic propagation in atrial reentry (abstract). Circulation 1988;78(suppl II):II Lazzara R, El-Sherif N, Scherlag BJ: Disorders of cellular electrophysiology produced by ischemia of the canine His bundle. Circ Res 1975;36: Davidenko JM, Antzelevitch C: Electrophysiological mechanisms underlying rate-dependent changes of refractoriness in normal and segmentally depressed canine Purkinje fibers: The characteristics of post-repolarization refractoriness. Circ Res 1986;58: Rozanski GJ, Jalife J, Moe GK: Determinants of postrepolarization refractoriness in depressed mammalian ventricular muscle. Circ Res 1984;55: Restivo et al High-Resolution Mapping El-Sherif N, Lazzara R: Reentrant ventricular arrhythmias in the late myocardial infarction period: Effect of verapamil and D6 and the role of the 'slow channel.' Circulation 1979; 6: Lazzara R, Scherlag BJ: Role of the slow current in the generation of arrhythmias in ischemic myocardium, in Zipes DP, Bailey JC, Elharrar V (eds): The Slow Inward Current and Cardiac Arrhythmias. Boston, Martinus Nijhoff Publishing, 198, pp El-Sherif N, Gough WB, Restivo M: Recovery of ischemic epicardium correlates with normalization of refractoriness and failure to induce reentry in the canine post-infarction heart (abstract). JAm Coll Cardiol 1987;9: El-Sherif N, Gough WB, Zeiler RH: The effect of differential shortening of refractoriness in successive short cardiac cycles on the initiation and termination of reentry in the ischemic canine heart (abstract). JAm Coll Cardiol 1984;3: El-Sherif N, Sherlag BJ, Lazzara R, Samet P: Pathophysiology of tachycardia and bradycardia dependent block in the canine proximal His-Purkinje system after acute myocardial ischemia. Am J Cardiol 1974;33: Allessie MA, Bonke FIM, Schopman FJG: Circus movement in rabbit atrial muscle as a mechanism of tachycardia: III. The 'leading circle' concept: A new model of circus movement in cardiac tissue without the involvement of an anatomical obstacle. Circ Res 1977;41:9-18 KEY WORDS * high-density mapping * ventricular tachycardia gradient of refractoriness ^ anisotropy * functional conduction block

19 Ventricular arrhythmias in the subacute myocardial infarction period. High-resolution activation and refractory patterns of reentrant rhythms. M Restivo, W B Gough and N el-sherif Circ Res. 199;66: doi: /1.RES Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX Copyright 199 American Heart Association, Inc. All rights reserved. Print ISSN: Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: Subscriptions: Information about subscribing to Circulation Research is online at: