Characteristics of the Temporal and Spatial Excitable Gap in Anisotropic Reentrant Circuits Causing Sustained Ventricular Tachycardia
Abstract—The excitable gap of a reentrant circuit has both temporal (time during the cycle length that the circuit is excitable) and spatial (length of the circuit that is excitable at a given time) properties. We determined the temporal and spatial properties of the excitable gap in reentrant circuits caused by nonuniform anisotropy. Myocardial infarction was produced in canine hearts by ligation of the left anterior descending coronary artery. Four days later, reentrant circuits were mapped in the epicardial border zone of the infarcts with a multielectrode array during sustained ventricular tachycardia induced by programmed stimulation. During tachycardia, premature impulses were initiated by stimulation at sites around and in the reentrant circuits, and their conduction characteristics in the circuit were mapped. All circuits had a temporal excitable gap in at least part of the circuit, which allowed premature impulses to enter the circuit. Completely and partially excitable segments of the temporal gap were identified by measuring conduction velocity of the premature impulses; conduction was equal to the native reentrant wave front in completely excitable regions and slower than the reentrant wave front in partially excitable regions. In some circuits, a temporal gap existed throughout the circuit, permitting the entire circuit to be reset over a range of premature coupling intervals, although the size of the gap varied at different sites. In other circuits, the gap became so small at local sites that even though premature impulses could enter the circuit, the circuit could not be reset. Premature impulses could terminate reentry in circuits that could be reset or not. We also found a significant spatial gap, which was identified by determining the distance between the head of the circulating wave front, which could be located on the activation map, and its tail, which was the site most distal from the head as located by the site of entry of the premature wave front into the circuit. The spatial gap could also vary in different parts of the circuit. Therefore, nonuniform anisotropic reentrant circuits have both a temporal and spatial excitable gap with fully and partially excitable components that change in different parts of the circuit.
The excitable gap in a reentrant circuit, ie, the excitable region that precedes the propagating reentrant wave front,1 has both temporal and spatial properties. It has a time duration at any point in the circuit, the temporal gap, and it has length at any moment in time, the spatial gap. One method that has been used to determine the temporal gap in the in situ heart has been to determine the range of coupling intervals of stimulated premature impulses that can reset the circuit, ie, advance activation at all sites in the circuit (resetting curves).2 3 4 Another method has been to compare the cycle length of reentry with the refractory period in a region of the reentrant circuit determined by the maximum rate at which this site will respond to pacing.5 6 When the cycle length of reentry is longer than the local refractory period, the difference between the two has been designated the (temporal) excitable gap.5 6 7 8 Although these methods have been useful for understanding properties of reentrant circuits, they do not take into consideration the potential heterogeneity of the temporal excitable gap in different regions of the reentrant circuit. For example, the temporal gap would be expected to have different values in different parts of the circuit if there were different refractory periods at different sites in the circuit. Yet the first method described above (resetting curves) shows the minimal excitable gap only at some undefined site in the circuit, whereas the second method gives a measure of the excitable gap only at the site where the refractory period is determined.
In contrast to the temporal gap, the spatial gap of a reentrant circuit is that part of the circuit path length that is excitable at any instant in time. It is the distance between the reentrant wave front and its refractory tail from the previous excitation of the circuit. The spatial gap has not been studied in the in situ heart, although it has been investigated in in vitro models of reentry and isolated perfused hearts.9 10 In regions of the circuit where conduction slows or the refractory period is short, the spatial excitable gap should expand, whereas in regions where there is speeding of conduction or lengthening of the refractory period, the spatial excitable gap should decrease.
In a previous study, we found that anisotropic reentrant circuits in in situ infarcted canine hearts11 12 could be entrained, indicating the presence of an excitable gap.13 However, modification of the reentrant circuit by the repetitive overdrive stimuli used during entrainment limits the usefulness of this stimulation technique for determining the properties of the excitable gap.14 In the present study, we have used single stimulated premature impulses delivered from multiple sites around and within anisotropic reentrant circuits in combination with activation mapping to determine the temporal and spatial properties of the excitable gap. A preliminary report of our results has previously been presented.15
Materials and Methods
Canine Model of Myocardial Infarction
Myocardial infarcts were produced in adult mongrel dogs (weight, 25 to 35 kg) by ligation of the left anterior descending coronary artery (LAD) near its origin.11 13 16 Four days after surgery, the dogs were reanesthetized with sodium pentobarbital (25 to 35 mg/kg) for the electrophysiological study. The left femoral artery was cannulated for constant monitoring of blood pressure, which was displayed along with leads II and III of the surface ECG on an oscillographic recorder (model DR 12, Electronics for Medicine). The left femoral vein was cannulated for administration of fluids. The chest was opened through a median sternotomy. A 9×13-cm flexible polymer sheet, 0.3 mm thick, containing 292 bipolar electrodes was fixed to the left ventricular epicardium over the anterior and lateral aspects of the infarct with 2–0 silk sutures to map impulse propagation in the epicardial border zone of the infarct.13 16 The chest was covered with a clear plastic sheet to maintain body temperature. The epicardial temperature was monitored with a thermistor probe (model 409 A, Yellow Springs Instruments) and maintained at 37°C to 38°C with a heating lamp.
Recording and Stimulating Electrodes and Instrumentation
The electrodes in the 292 bipolar array were made from silver disks with a diameter of 1.0 mm. The distance between the centers of two disks forming a bipolar pair was 2.0 mm. We could record simultaneously from 196 of the 292 electrodes at any one time with a computerized mapping system.13 16 Bipolar stimulating electrodes were positioned around the centermost electrodes, at the basal and lateral margins of the recording array, and on the right ventricular epimyocardium adjacent to the LAD margin of the electrode array.13 16
Ventricular tachycardia was induced by either single or double premature stimuli (constant current, 2 ms in duration and 2× diastolic threshold) delivered during basic drive from one of the stimulating electrodes (central, basal, LAD, and lateral). Sustained ventricular tachycardia lasting for at least several minutes (all tachycardias were monomorphic) was required in order to study the properties of the excitable gap of the reentrant circuit. In addition, we only included experiments in which the entire reentrant circuit could be mapped in the epicardial border zone underlying the electrode array. Eight complete circuits11 13 16 were recorded during sustained tachycardia in the seven dogs that formed the experimental group. In one of these experiments, two different reentrant circuits occurred.
When a stable reentrant tachycardia was established, single stimuli (2-ms duration and 2× diastolic threshold) were introduced from one of the stimulating electrodes at intervals spanning the full cycle length, approximately every 10 tachycardia cycles (Fig 1A⇓). If tachycardia terminated spontaneously or was terminated by one of the stimuli (bottom ECG in Fig 1A⇓) before completion of stimulation throughout the entire cycle length, tachycardia was reinitiated after several minutes, and the protocol was completed. We tried to repeat this procedure at each of the four different stimulation sites, ensuring that the tachycardia had the same QRS morphology and reentrant circuit, but did not succeed in completing the entire protocol in all experiments.
The magnetic tapes from each experiment were replayed for off-line data analysis. Our methods for determining local activation times, drawing isochrones, designating regions of apparent conduction block, and locating reentrant circuits have been described in detail.11 13 16
We determined the temporal and spatial excitable gap by mapping the propagation of stimulated premature impulses into the circuit during ventricular tachycardia. We describe these methods in more detail along with the results for greater clarity.
The Temporal Excitable Gap in Anisotropic Reentrant Circuits
Characteristics of the Temporal Excitable Gap
We first describe the results of a representative experiment showing that anisotropic circuits have a temporal excitable gap and that the temporal gap at different sites in the reentrant circuit (local temporal excitable gap) is different, one of the major findings of the present study (Table 1⇓, experiment 1A). The activation map in Fig 1B⇑ shows the stable pattern of reentry causing the sustained tachycardia in Fig 1A⇑. In the time window shown, the sequence of isochrones (designated by the black arrows) shows a wave front (the “native” reentrant impulse) initially moving toward the LAD margin, then dividing in two, and continuing around two lines of functional block (thick black lines) parallel to the long axis of the myocardial fiber bundles, which were not present in the absence of tachycardia, to form a typical figure-of-eight anisotropic reentrant circuit.11 12 17 The activation pattern around the left line of block is also shown by the sequence of electrograms in Fig 1C⇑.
Fig 2⇓ illustrates the propagation of a long-coupled premature impulse initiated from the stimulating electrodes at the LAD site (pulse symbol in Fig 2A⇓) during a complete scan of the cardiac cycle that was used to determine the properties of the temporal excitable gap of the reentrant circuit described in Fig 1B⇑. In this and subsequent figures, we focus only on one of the reentrant loops of the figure-of-eight circuits to keep our description as brief as possible, but our analysis has shown similar phenomena occurring in both loops. The time window illustrated by the activation map in Fig 2A⇓ begins at the vertical line in Fig 2B⇓, which is the time that the stimulus was applied. The long arrows in Fig 2A⇓ represent the stimulated propagating wave front impinging on and entering the circuit in a region that we call “the zone of preexcitation.” We define the zone of preexcitation as the region where the stimulated propagating premature impulse enters the circuit before it becomes the reentrant wave front. The greatest degree of preexcitation in the circuit, ie, the most orthodromic electrogram showing the shortest premature coupling interval, defines the limit of the zone of preexcitation in the orthodromic direction, since it is the most orthodromic site to be excited by the premature wave front entering the circuit before it becomes the new reentrant wave front. In Fig 2⇓, it is electrogram 54 (circled in Fig 2B⇓ and shown with a darkened electrogram trace) with a coupling interval of 250 ms that is the most premature electrogram, 28 ms earlier than activation of that site by the native impulse. Electrogram 54 is activated simultaneously or before electrograms recorded from sites that precede it in the circuit (electrograms 50 and 52), showing that it is not activated by an orthodromically propagating reentrant wave front. The electrogram with the shortest coupling interval (electrogram 54) is, therefore, the entrance into the circuit from which the stimulated wave front propagates in the orthodromic direction through the rest of the circuit, as indicated by the expected sequence of activation and electrogram morphology of sites in advance of this site in the circuit (electrograms 82, 96, etc). Electrogram 54 is called the circuit entry electrogram.
Premature activation at electrode 54 results in an electrogram morphology similar to that produced by the native reentrant wave front (Fig 2B⇑), because the directions of propagation of the native and stimulated wave fronts at that site are similar. However, unlike the native cycle, the slight alteration in pattern of activation of this region by the stimulated wave front results in electrogram 54 preceding electrogram 52. At this stimulated coupling interval, the stimulated premature wave front enters the circuit when the head of the native reentrant wave front (represented by the stippled region in Fig 2A⇑) is somewhere near sites 37 and 50. This conclusion is based on the observation that electrogram 37 (Fig 2B⇑) is activated at the native cycle length with a nearly normal morphology; it is not perturbed by the premature wave front. Electrogram 50 is activated at a cycle length that is 2 to 3 ms less than the tachycardia, with a slightly altered morphology, but is coincident with activation of the preexcited zone just ahead of it. Therefore, there may be collision of the propagating stimulated wave front and the native reentrant wave front near electrograms 37 and 50 (diagrammed in Fig 2A⇑). The zone of preexcitation extends from electrograms 50 to 54.
The premature impulse propagates through the circuit with a pattern similar to the native reentrant impulse (Fig 2A⇑, curved arrows) but with slower activation, as indicated by the coupling intervals of the premature activation increasing with propagation from the circuit entry electrogram, electrode 54 to the subsequent representative electrograms around the circuit (Fig 2B⇑). Although much of the circuit is activated prematurely, as a result of this slowing of activation in the circuit, the premature impulse falls back on time with respect to the native intervals by the time it completes a full reentrant cycle. The next, or return, coupling interval at circuit entry electrogram 54 is 303 ms (Fig 2B⇑), ie, 25 to 26 ms longer than the native cycle length. At this recording site, the sum of the premature and return cycle lengths is nearly twice the basic tachycardia cycle length; therefore, the amount of slowing of activation has balanced the degree of prematurity. Resetting has not occurred. Similar patterns of activation occurred for coupling intervals between 250 ms (Fig 2B⇑) and 270 ms; the premature impulse entered the circuit and preexcited part of it but not all of it.
Shorter coupled premature impulses stimulated at the LAD margin (coupling intervals between 250 and 209 ms) during this tachycardia also entered the circuit at electrode 54 (circuit entry electrogram) but preexcited the entire reentrant circuit. Fig 3⇓ illustrates the pattern of propagation of these shorter coupled premature impulses. In Fig 3A⇓, propagation of the stimulated impulse occurred in a broad wave front (long arrows) that resulted in early and nearly simultaneous activation of electrograms 37, 50, 52, and 54 (darkened electrogram traces in Fig 3B⇓). Collision of the stimulated wave front with the native reentrant wave front occurred between electrodes 37 and 39; electrogram 39 has the cycle length of the tachycardia and a slightly altered morphology, whereas the cycle length at electrode 37 is decreased, and the electrogram morphology is significantly changed (Fig 3B⇓). The greatest degree of preexcitation in the circuit (ie, the electrogram showing the shortest premature coupling interval) was at the distal (orthodromic) end of this zone of preexcitation, still electrogram 54 (coupling interval 209 msec in Fig 3B⇓), which was responsible for propagation of this premature wave front around the remainder of the circuit. Electrogram 54 remained the circuit entry electrogram. The pattern of activation remained nearly identical to the native reentrant circuit, although activation of the circuit by the premature impulse was slower than by the native reentrant wave front, as shown by coupling intervals that are decreasingly premature (the coupling intervals lengthen) with propagation from the entrance point (Fig 3B⇓). For coupling intervals between 250 and 209 ms (not shown), the premature wave front finished a complete excursion around the circuit, with the next, or return, coupling intervals at circuit entry electrogram 54 in advance of where the native wave front would have been if the circuit had been left undisturbed, despite the conduction delays. That is, the sum of the preexcited intervals at electrogram 54 and the return intervals were less than twice the ventricular tachycardia cycle length, and the circuit was therefore reset. At the shorter premature coupling interval of 209 ms shown in Fig 3⇓, the stimulated premature impulse blocked in the circuit, causing termination of tachycardia (see Fig 1A⇑, bottom ECG trace). Recording site 41 was preexcited at a coupling interval of 238 ms (Fig 3B⇓), earlier than it was activated by any other premature wave front. This resulted in conduction block of the premature wave front in the region just distal to electrogram 41, indicated by the horizontal dashed line between sites 41 and 39 in Fig 3A⇓ and the solid horizontal line between these electrograms in Fig 3B⇓. Therefore, the coupling interval of 238 ms at electrogram 41 approximates the effective refractory period of the tissue at the site where block occurred, just distal to 41, and is the longest local effective refractory period in the circuit. It represents the inner boundary of the local temporal excitable gap at that site. The overall duration of the excitable gap just distal to electrode 41 is the difference between the native cycle length (278 ms) and the interval that blocked (238 ms), ie, 40 ms (Table 1⇑, experiment 1A).
When the full range of coupling intervals of stimulated premature impulses from the LAD margin, measured at the circuit entry electrogram (electrogram 54), is plotted (Fig 4⇓), the curve of premature coupling intervals against return cycle length shows two phases. There is a phase representing the window of reset of the circuit, whereby the sum of the premature coupling interval and the return cycle length is less than twice the ventricular tachycardia cycle length (the segment of the curve that falls below the dashed line in Fig 4⇓). Over this range of coupling intervals, all points in the reentrant circuit were preexcited by the premature impulse. The window of reset of this circuit spans 41 ms between coupling intervals of 250 and 209 ms. The inner boundary of the window is determined by the coupling interval at which the premature impulse was blocked in the circuit at the site with the longest refractory period and the shortest excitable gap (site near electrode 41). Between coupling intervals of 250 ms and the tachycardia cycle length of 278 ms, despite entrance into the circuit of the premature impulse, the entire circuit was not reset because the stimulated wave front propagated prematurely through only part of the circuit (illustrated in Fig 2⇑). The part of the resetting curve during these intervals constitutes a second phase of the curve in Fig 4⇓, the window of compensatory pause, which is indicated by coincidence between the points on the curve and the dashed line. In this experiment, the window of reset was 15% of the reentrant cycle length (Table 1⇑, experiment 1A), and it is the duration of this portion of the curve that shows resetting, which has been interpreted (in studies without detailed mapping2 3 4 ) to indicate the duration of the temporal extent of the excitable gap of the reentrant circuit. In fact, the present mapping study shows that there is no one temporal excitable gap but that the gap differs at different sites. The excitable gap near site 41 was ≈40 ms (the difference between the tachycardia cycle length and the cycle length at which block occurred) (Table 1⇑, experiment 1A), coinciding with the reset portion of the curve, but the gap at circuit entry electrogram 54 was at least 69 ms (25% of the cycle length; Table 1⇑, experiment 1A). That is, a stimulated wave front was able to enter the circuit at electrogram 54 over a range of coupling intervals spanning 69 ms from a premature coupling interval of 209 ms to the tachycardia cycle length of 278 ms. A gap larger than 69 ms may have existed at electrode 54, since we did not determine the refractory period of this site, and a larger gap that we did not measure may have existed at another region in the circuit.
We have also found that anisotropic reentrant circuits that cannot be reset at any premature coupling interval may still have substantial local temporal excitable gaps, also demonstrating that the gap changes in different regions of the circuit (Fig 5⇓). Fig 5E⇓ shows the curve of premature coupling intervals versus the return cycle at the circuit entry electrogram (electrogram 175) for such an experiment on another figure-of-eight circuit that is diagrammed in Fig 5B⇓. In this circuit, segments of the lines of block (thick black lines) were oblique or perpendicular to the long axis of the fibers, a characteristic that sometimes occurs in anisotropic circuits and may be the result of unspecified damage caused by infarction. The LAD stimulation site (pulse symbol) and electrogram recording sites around the circuit to the left are indicated on the diagram in Fig 5B⇓. The data points in Fig 5E⇓ fall on the line, showing that the sum of the premature coupling interval and the return cycle equals twice the ventricular tachycardia cycle, indicating that the circuit was not reset. Yet the electrograms that were recorded around the circuit indicate that the stimulated impulses did enter the circuit. In Fig 5A⇓, 5C⇓, and 5D⇓, the sequence of activation during the native tachycardia begins at electrogram 65 (asterisk at the left on the bottom trace in each panel) and progresses to electrogram 53 (top trace). The vertical lines show the time a premature stimulus was applied, and the long arrow to the right of the lines shows propagation of the premature impulse in the circuit. Despite the failure of the circuit to be reset, part of the circuit was preexcited. Electrode 175 (circled) is the circuit entry electrogram and is preexcited (darkened part of the trace) at each of the three coupling intervals shown (217, 155, and 194), indicating the presence of an excitable gap at the circuit entry site. However, at all coupling intervals, slowing of activation of the premature wave front propagating in the circuit from the circuit entry electrogram caused a loss of prematurity at subsequent sites in the circuit and failure to reset. The amount of slowing was proportional to the prematurity of the reentrant wave front, as shown by the slopes of the long arrows. For coupling intervals of 217 ms (Fig 5A⇓) and 155 ms (Fig 5C⇓) at electrode 175, activation falls back on time when electrode 56 is reached (coupling interval of 227 ms, which is the same as the tachycardia cycle length) because of conduction slowing that increased with increasing prematurity. Therefore, the region around the end of the left line of block could not be preexcited, suggesting that the wave front is propagating in just-recovered myocardium in this region and that the temporal gap immediately before electrode 56 is small. However, despite the inability of premature impulses to reset the circuit, a premature impulse with a coupling interval of 194 ms, midway through the range of coupling intervals, was blocked in the circuit, thereby terminating reentry (Fig 5D⇓). This premature impulse activated electrode 56 five milliseconds earlier than either the later-coupled or the earlier-coupled premature impulse (coupling interval of 222 ms in Fig 5D⇓), causing block just distal to this site. The cause of the earlier excitation of electrode 56 appears to be less delay in conduction around the lower end of the left line of block than with the shorter or longer coupling intervals; the reason is unknown. Therefore, the refractory period at the site of block between electrodes 56 and 55 is ≈222 ms, which is 5 to 7 ms shorter than the tachycardia cycle length, indicating that the excitable gap was ≈5 to 7 ms in this region. In this circuit, therefore, the temporal size of the excitable gap varied from at least 75 ms at the circuit entry electrogram (the inner boundary of the gap at this site was not determined) to ≈7 ms at this site of block (Table 1⇑, experiment 5). The inability to reset this circuit with premature stimuli delivered at 5- to 10-ms decrements was the result of the very small duration of the excitable gap in this region.
In four other experiments, we determined local temporal excitable gaps at different entry sites into the circuit, three entry sites in two experiments (Table 1⇑, experiments 3 and 4) and two entry sites in one experiment (Table 1⇑, experiment 2), by initiating premature impulses at different sites outside and within the circuit (see “Materials and Methods”). We also related the local temporal gaps to the resetting properties of the circuits in these experiments. Figs 6⇓ and 7⇓ show determination of the temporal gap at two different entry sites in the same circuit. The native reentrant circuit closely resembled the one shown in Fig 1B⇑. Although Fig 6A⇓ shows the circuit during premature stimulation (described below), the two parallel lines of functional block (long, thick, black lines) during the native circuit were located in the same region. The native reentrant wave front (not shown) moved around the left line of functional block in a counterclockwise direction, and the right line of block in a clockwise direction. The activation sequence around the right line of block during the native tachycardia is shown in Fig 6B⇓ by electrograms recorded from selected electrodes indicated on the map; beginning at electrode 26 (asterisk in bottom trace), the sequence of activation progresses in a clockwise direction to electrode 41 (top trace), which completes the circuit. Fig 6A⇓ shows the pattern of activation during stimulation at the basal electrodes (pulse symbol). Although this earliest stimulated wave front activated electrode sites 14 to 46 nearly simultaneously (Fig 6B⇓, darkened electrograms; vertical line marks the stimulus artifact) when the head of the native wave front was near electrode 12 (no change in electrogram 12 cycle length or morphology in Fig 6B⇓), the most distal and prematurely activated site that constitutes the circuit entry point is electrode 46 (circled in Fig 6B⇓), which has a premature coupling interval of 186 ms (53 ms early). From the entry point at electrode 46, the premature impulse propagated around the circuit with delay but still resulted in resetting of the circuit by 22 ms (Fig 6A⇓ and 6B⇓). Although the local temporal gap at this circuit entry site (electrode 46) was 53 ms, the circuit was reset over only the shortest 45 ms of this range (Table 1⇑, experiment 4, base) with no resetting over the remaining 8 ms of the local gap, similar to Fig 4⇑.
Fig 7⇑ shows determination of the temporal gap at another entry point in this circuit, at electrode 41 in the central common pathway. The illustration is also of the earliest effective stimulated impulse. From the entry point at electrode 41 (circled in Fig 7B⇑, coupling interval of 156 ms), the stimulated impulse propagated antidromically to collide with the native wave front (unfilled arrows) between electrode 43 and electrode 59; electrode 59 has a normal cycle length and morphology (Fig 7B⇑). The approximate region of collision is indicated by the interface of the shaded and unshaded areas on the map in Fig 7A⇑. Slowing of activation of the premature impulse occurred around the circuit in the orthodromic (clockwise) direction (arrows in Fig 7A⇑) with preexcitation of the local site 41 and the resetting of the circuit over a range of coupling intervals of 88 ms.
The excitable gap was also determined in this experiment from stimulating at the LAD electrodes (not shown). The circuit entry electrode was electrode 12. From maps of premature impulses over the entire cycle length, we determined that the local temporal excitable gap at electrode 12 was 47 ms (this region was excited over a 47-ms range of coupling intervals), and the circuit was reset over this entire range (Table 1⇑, experiment 4, LAD). Therefore, there was variability in the local temporal excitable gap at three different locations in this circuit: 53 ms at base electrode 46 (Fig 6⇑), 88 ms at center electrode 41 (Fig 7⇑), and 47 ms at electrode 12 (Table 1⇑, experiment 4, EGAP column).
Table 1⇑ summarizes the temporal properties of the excitable gap for the entire series of eight experiments. A local temporal excitable gap permitting entry of prematurely stimulated wave fronts existed in all of the circuits (EGAP column). However, not all sites of premature stimulation produced wave fronts that were able to enter the circuit. For example, in experiment 1A, stimuli at the base or lateral electrodes initiated wave fronts that did not enter the circuit at any coupling interval. In the experiments in which the local gap was determined at two to three sites in the circuit, either only at circuit entry sites (experiments 2, 3, and 4) or at an entry site and a site of block (experiments 1A and 5), the local gap was different at each site. Five of the circuits could be reset (experiments 1A, 1B, 2, 3, and 4), indicating the existence of a temporal excitable gap throughout the circuit at least as long as the zone of reset, permitting the entire circuit to be preexcited by the propagating premature impulse. In three of these five circuits (experiments 1A, 2, and 4), we measured local temporal excitable gaps that exceeded the window of reset (reset/EGAP<1), and in the other two, it coincided with the window of reset (reset/EGAP=1). In three experiments, the reentrant circuit could not be reset (reset window=0 for experiments 5, 6, and 7). However, Table 1⇑ shows that a local temporal gap existed in each of these circuits at the circuit entry site, ranging from 3% to 33% of the reentrant cycle length. The local temporal excitable gap that we determined in each of these circuits is, therefore, greater than the minimal excitable gap in the circuit that limited reset, again demonstrating variability of the gap in different regions of the circuit.
Conduction Characteristics During the Temporal Excitable Gap
Conduction properties of the premature impulse in the circuit can provide information on excitability characteristics of the temporal gap. We located the regions of delay in activation and quantified the slowing of activation by dividing the circuits into four regions: the two turning points of the wave front through 180° around the ends of the line of functional block, designated the “entry pathway” and “exit pathway” with respect to the central common pathway, and the two conduction pathways between the turning points, one within the central common pathway and one parallel to the lines of block outside the central common pathway, designated the outer pathway. Around the turning points (entry and exit pathways), the reentrant wave front is moving mostly transverse to the long axis of the myocardial fiber bundles, whereas in the central common pathway and outer pathway, it is moving mostly parallel to the long axis in our eight experiments.11 13 As expected, during the native tachycardia, the velocity of activation (calculated from the distance between the recording sites and the time required for the wave front to move this distance), parallel with the lines of functional block in the central common pathway and outer pathway (and therefore along the long fiber axis), was significantly greater for all experiments (61.4±22 cm/s) than the velocity around the turning points (entry and exit pathways) of the lines of functional block for all experiments (24.6±12 cm/s, P<.01). Activation was more rapid through the outer pathway (75.5±26.1 cm/s) than through the central common pathway (47.2±22.3, P<.01). Activation velocities in both the longitudinal and transverse directions were within the range of normal conduction velocities for ventricular muscle,18 suggesting that the reentrant impulse may be conducting in fully excitable myocardium. Table 2⇓ shows the ratio of the activation velocities of the native reentrant impulse and most prematurely stimulated impulse (N/P ratio) in each of the four regions of the circuit for the eight circuits and 13 sites of stimulation. A ratio of 1, therefore signifies that the velocity of the premature wave front is equal to that of the native wave front over that portion of the circuit, whereas a ratio of >1 indicates that the velocity of the premature wave front is slower than the native wave front. Over the range of coupling intervals of premature impulses that entered the circuit, the shortest coupling interval resulted in the slowest activation through the entire circuit. For 12 of the 13 sites of stimulation, the greatest degree of activation slowing of the premature wave front occurred in the segment immediately after entry into the circuit (asterisks) (mean N/P ratio, 1.42; n=12). Five of the eight circuits studied could be reset, indicating that despite this slowing in the entrance region, there was premature activation of the remainder of the circuit. This occurred because in the other regions of the circuit, the premature wave front propagated with a velocity similar to that of the native wave front with a mean N/P ratio of 1.05. For the most premature wave front in the three nonentry portions of the circuit, a ratio of 1 occurred in one or more portions of each of the circuits, indicating an activation velocity for the most premature wave front in these regions that was equal to the velocity of the native reentrant wave front (Table 2⇓). Thus, even though slowing of activation exists in part of the circuit, indicating that the premature impulse conducts in relatively refractory tissue and that there is only a partially excitable gap in these regions, other regions of the circuit may have a fully or nearly fully excitable gap, since activation by premature wave fronts occurs at normal velocities, which are the same as the native wave fronts in these regions. Therefore, the temporal gap is composed of partially and fully excitable regions.
The Spatial Extent of Excitable Gap in Anisotropic Reentrant Circuits
To measure the spatial extent of the excitable gap (the length of the circuit that is excitable), we made the initial assumption that all the myocardium in the circuit in the orthodromic direction between the head of the circulating reentrant wave front and the end of its refractory tail is excitable and constitutes the spatial excitable gap. Since the head of the circulating wave front can be identified from the activation maps, it was then necessary to identify the end of the refractory tail to determine the spatial gap. This was done by identifying the site that was the most orthodromic from the head of the reentrant wave front and at which the premature wave front could enter the circuit, ie, the circuit entry electrogram, as previously defined. Propagation of the most premature wave front into the circuit, therefore, identified the largest extent of the spatial gap in the circuit entry regions, because this represents the time that the tail has just passed the entrance point.
By these criteria, the experiments already illustrated have revealed information about the spatial extent of the excitable gap. In Fig 3⇑, for a short coupled premature impulse, the distance between the circuit entry site (electrode 54), which identifies the end of the tail of refractoriness, and the head of the wave front, which is at electrode 39 (identified by normal morphology and cycle length of the native tachycardia), is the minimal spatial extent of the gap in this region in the orthodromic direction of the circuit, a distance of 35 mm. (We use the descriptor “minimal” since the spatial gap might extend even further orthodromically than electrode 54, but the prematurely stimulated wave front only reached more orthodromic sites after entering the circuit and propagating in this direction.) Nearly simultaneous activation of this region, spanning electrograms 39 to 54 in the circuit by the premature impulse and illustrated by the darkened electrograms in Fig 3B⇑, indicates that the myocardium between these electrodes is in fact excitable; therefore, all the electrodes lie within the spatial excitable gap. The spatial gap is 35% of the circuit length, which was estimated to be 101 mm by measuring the distance as close as possible to the central line of block (Table 3⇓, experiment 1A). Since in this experiment (and the one illustrated in Fig 5⇑; see Table 3⇓, experiment 5) there were no other sites of premature stimulation and no other circuit entry points, we were not able to determine the spatial extent of the gap in other regions and, thus, whether it changed size.
However, in the experiment previously described in Figs 6⇑ and 7⇑, premature impulses, initiated from three stimulation sites, enabled us to determine the spatial extent of the gap at the different circuit entry points (Table 3⇑, experiment 4). Stimulation at the basal site revealed a very large spatial extent of the gap. In Fig 6⇑, we described that the greatest degree of preexcitation of this reentrant circuit from the basal stimulation site occurred at electrogram 46, which is the circuit entry electrogram. At this time, the head of the native reentrant wave front was in the vicinity of electrode 12. Therefore, in Fig 6⇑ the distance between electrode 46 and electrode 12 is the spatial extent of the excitable gap in this region, a distance of 42 mm (42% of the circuit length of 101 mm). Basal stimulation at this short coupling interval also resulted in a broad wave front that simultaneously preexcited the entire basal side of the right-hand line of block between electrodes 14 and 46 (Fig 6B⇑, darkened electrograms), confirming that this entire region was excitable when the head of the reentrant wave front was in the vicinity of electrode 12.
Central stimulation at the shortest coupling interval within the common central pathway of this circuit preexcited the most local electrogram (electrogram 41) at a coupling interval of 156 ms (Fig 7⇑), when the head of the native wave front was in the vicinity of electrogram 59 (Fig 7B⇑). Therefore, at this moment, the spatial excitable gap extended at a minimum from just distal to electrode 41 to the head of the advancing wave front near electrode 59. Antidromic conduction of the premature impulse occurred from the stimulus site to the electrodes between it and the advancing native wave front. Thus, the region of antidromic propagation was excitable and represents the minimal spatial extent of the excitable gap in this direction, which measures ≈26 mm.
However, even at the shortest coupling interval of premature impulses delivered from the LAD stimulation site in this circuit (not shown), which should expose the excitable gap in its most counterclockwise position, electrogram 24 could not be preexcited when the circuit entry electrogram was electrode 12 (electrode locations in the circuit shown in Figs 6⇑ and 7⇑). This implies that when the head of the native wave front was at electrogram 24, the excitable gap at electrode 12 had a very small spatial extent, which we estimate to be ≈5 mm in a circuit with a length of 101 mm (Table 3⇑, experiment 4, LAD). Therefore, as shown by the extent of the spatial gap at the circuit entry points in this experiment, the spatial extent of the excitable gap changed markedly in different locations of the circuit.
The minimum spatial extent of the excitable gaps in one or more regions of all circuits studied are given in Table 3⇑. The size of the gap ranged from 5 to 57 mm (mean value, 27.8±1.69 mm), which is 5% to 55% of the circuit length. In some cases, this is likely to be a significant underestimation of the extent of the gap because of the methodological shortcomings described above. In the three experiments in which the spatial extent of the gap was determined at multiple sites, (experiments 2, 3, and 4), there is clearly variation in the size of the gap in one (experiment 4) but no clear variation in the others (experiments 2 and 3).
The Excitable Gap in Reentrant Circuits
Traditionally, the properties of the excitable gap have been thought to be different in reentrant circuits caused by anatomic and functional mechanisms.19 20 Reentry caused by the leading circle mechanism, the first mechanism described in detail in heart muscle causing functional reentry, has a very small partially excitable gap,19 whereas some anatomic circuits have a large fully excitable gap.8 21 22 23 24 However, more recently, studies on spiral waves in cardiac muscle have shown that this mechanism of functional reentry is associated with a fully excitable gap,9 25 so such a clear distinction between functional and anatomic reentry can no longer be made. An excitable gap has been shown to occur during atrial5 26 and ventricular6 fibrillation and atrial flutter,27 examples of arrhythmias caused by functional reentrant mechanisms, possibly including spiral waves.28 29 In the present study, we have characterized the properties of the excitable gap of functional reentrant circuits that occur in nonuniform anisotropic myocardium of the infarct epicardial border zone and have found that these circuits also have an excitable gap with a fully excitable component, although the fully excitable component did not exist in all regions of the circuit. In those regions of the circuit where there was a fully excitable gap, propagation of the most premature impulses occurred with conduction velocities that were equal to the conduction velocities of the native impulse (native wave front velocity/premature velocity=1 in Table 2⇑). Although it is possible that the native reentrant wave front itself conducts in relatively refractory myocardium, we do not think that this is the case, since conduction velocities transverse and longitudinal to fiber bundle orientation were similar to velocities in fully excitable normal ventricular muscle.18 However, we did not find a fully excitable gap in all parts of any of the circuits that we studied, unlike some anatomic circuits.7 8 23 24
The temporal excitable gap of the circuits that we studied changed in different parts of the circuit. Although it has been recognized that the gap can change in reentrant circuits,7 21 27 30 in many previous studies, the temporal excitable gap measured at one site has often been treated as a unique value that is considered to be the same in all parts of the circuit; therefore, measurement of the gap at a single site in the circuit has provided the value for the excitable gap of the entire circuit.2 3 4 31 32 This is particularly relevant for studies in which resetting response curves have been used to characterize the properties of the excitable gap in clinical studies of reentrant ventricular tachycardia in which the zone of reset is taken as a measurement of the excitable gap.2 3 4 33 34 However, circuits that we studied that could not be reset still had excitable gaps in part of the circuit, as also shown in the study of Boersma et al.7 Reentry could even be terminated by a blocked premature impulse in a circuit that could not be reset (Fig 5⇑).
The present study is the first to distinguish the spatial gap from the temporal gap and measure the spatial gap in the in situ heart. Pertsov et al9 have previously described a spatial and temporal gap in spiral wave reentry in isolated superfused epimyocardium, and Girouard et al10 have demonstrated the spatial extent of the gap in reentrant circuits in isolated guinea pig hearts. The spatial gap in both these studies changed in different parts of the circuit; we also noted this change in an experiment in which we were able to measure the spatial gap at three different sites.
Possible Mechanisms for Slow Activation, Functional Lines of Block, and the Excitable Gap
The characteristics of the temporal and spatial excitable gap in the present study provide new information concerning the nature of slow activation and the mechanism for the formation of the lines of block associated with reentry in the nonuniformly anisotropic epicardial border zone. Five of the eight circuits that we studied could be reset over a wide range of coupling intervals, and as described above, in some regions the premature impulses conducted at the same velocities as the native reentrant impulse. This property shows that propagation in just-recovered myocardium of the refractory tail is not a mechanism for reentry, as is found in the leading circle mechanism of functional reentry.19 If the head of the reentrant wave front was conducting in just-recovered, partially excitable myocardium in any part of the circuit, premature activation would not be able to preexcite locally or advance activation throughout the circuit, since premature impulses would encroach on the effective refractory period of myocardium in this region and lead to conduction block. In our original description of this model of reentry, we attributed the slow activation that enables reentry to occur to slow and discontinuous transverse propagation in the nonuniformly anisotropic myocardium.11 In fact, slow activation in these anisotropic circuits does occur mainly in the transverse direction, whereas activation in longitudinally oriented segments of the circuit is more rapid. However, the recent studies on spiral wave reentry in cardiac muscle9 25 have shown the importance of a convex wave-front curvature as a cause of slow activation in reentrant circuits, with the greater the curvature, the slower the conduction.35 36 These concepts originated in earlier studies of spiral waves in excitable media.37 38 Wave-front curvature should also play a role in slow conduction in the circuits in the epicardial border zone. In the elongated figure-of-eight reentrant circuits in an anisotropic matrix, the curvature is expected to vary in different segments of the circuit, with the greatest curvature being around the pivot points at the ends of the lines of block as the reentrant wave front emerges from the relatively narrow central common pathway into a larger mass of myocardium.35 Therefore, changes in curvature can also explain the differences in conduction velocity in different parts of the circuits that we studied, and the relative roles of curvature and anisotropy are uncertain.
The characteristics of conduction of premature impulses in anisotropic reentrant circuits also suggest that the functional lines of block that occur during reentry are not solely the result of a refractory barrier, as has been proposed previously.39 A reentrant wave front pivoting around the end of a functional line of block caused solely by a refractory barrier would not be expected to be advanced by a premature impulse without extension of the line of block.21 However, resetting of the anisotropic circuits demonstrates that all points in the circuit, including the pivoting points around the ends of the lines of block, could be activated earlier by premature impulses than by the native tachycardia impulse. The distal end of the line of block and the pivoting point around the line of block remained in a similar region for these premature impulses. Therefore, the distal ends of the lines of block were not the result of refractoriness. The electrophysiological mechanism for the lines of block in this reentrant model still remains unknown. Given that wave-front curvature is a likely important cause of slow activation, what appears to be lines of functional block might be related to the unexcited core caused by spiral waves, which in an anisotropic matrix would be elongated.9 25 35 36 37 38 If the wavelength is significantly larger than the pivoting radius, the trajectory of the tip of the wave front is expected to assume a linear shape similar to the region of functional block observed during reentrant excitation in the epicardial border zone.36 Although we know that the regions where the lines of block form are normally excitable in the absence of ventricular tachycardia,11 13 we do not know whether they remain unexcited during reentry, similar to the core of a spiral wave, because of the lack of the necessary spatial resolution of our mapping electrode array. However, there are likely to be additional factors contributing to the development of lines of block, including an abnormality of distribution of connexin43 in regions correlating with the location of the common central pathway. Lines of block in anisotropic reentrant circuits in this model may be a consequence of alterations in the spatial organization of gap junctions, which impede conduction of the reentrant wave front transversely.40 The region of nonuniformity provided by gap junctional disarray may also serve to “anchor” the reentrant circuit in a fixed location in a medium (epicardial border zone) where there are significant spatial gradients in electrophysiological properties such as refractoriness.9 25 41
Does the nonuniformity of anisotropy in the epicardial border zone contribute to the properties of the excitable gap? If wave-front curvature is an important cause of slow activation, then it is most likely an important determinant of the temporal and spatial excitable gap that we found. Spiral wave reentry has been shown to have a fully excitable temporal and spatial gap that is not dependent on structural discontinuities, because the critical curvature of the wave front prohibits an abrupt turn leaving an unexcited area.9 25 36 A microanatomic factor contributing to defining the ends of the lines of block (gap junctional disarray40 or some other discontinuity in structures) also may contribute to the occurrence of a significant excitable gap, since it can prevent the reentrant wave front from making an abrupt turn and taking a shorter pathway.7 This may also explain why the anisotropic circuits in nonuniform anisotropic myocardium in infarcts have a larger excitable gap than anisotropic circuits in uniform, normal myocardium,7 42 where there is no gap junctional disarray or anatomic discontinuities, and why the excitable gap in the nonuniform anisotropic circuits may have a fully excitable component, whereas it does not in the uniform anisotropic circuits.7 The question of whether the localized anatomic discontinuities (gap junctional disarray), more generalized nonuniform anisotropy, or a combination of the two is crucial for determining properties of the excitable gap remains to be answered.
In considering the cause for changes in the excitable gap in different parts of anisotropic reentrant circuits, both the temporal and spatial characteristics of the gap should change if there are changes in conduction velocity and/or refractory periods in different parts of the circuit. Changes in the gap are a consequence of changes in the wavelength of the reentrant impulse (conduction velocity×refractory period) in different parts of the circuit. In a computer model of spiral wave reentry, adding anisotropy to an isotropic matrix expands the wavelength in the longitudinal direction relative to the transverse direction.9 Likewise, conduction velocity in the anisotropic epicardial border zone of the infarcted heart is much more rapid in the longitudinal direction than in the transverse direction.11 Therefore, if there were no differences in refractory periods throughout the circuit, an expansion of the wavelength during rapid conduction9 10 42 43 should lead to a decrease in the spatial excitable gap, which would reach a minimum as the wave front finished its period of rapid conduction in the longitudinal segments of the circuit. On the other hand, after propagating slowly around the pivoting point at the ends of the line of block, the wavelength should reach a minimum,9 10 43 leaving a large part of the circuit time to recover and a large spatial gap. Because of the limited number of sites in each of the circuits at which we could measure the excitable gap, we could not precisely test this hypothesis, although some of our results conform to its prediction. In the experiment described in Figs 6⇑ and 7⇑, both the temporal and spatial gaps that occurred on the basal side of the line of block (outer pathway) as the reentrant impulse finished a period of slow conduction around the upper end of the line of block were quite large (Fig 6⇑), whereas the gap at the LAD entrance site (electrode 12) measured after a period of rapid conduction of the reentrant impulse in the central common pathway was much smaller. It is also likely that the refractory periods vary around anisotropic circuits,39 41 contributing to changes in both the spatial and temporal properties of the excitable gap,30 although this was not found to be an important contribution to variations of the excitable gap in reentrant circuits in normal ventricular myocardium.8 23
The properties of the excitable gap influence the characteristics of arrhythmias caused by reentry. Arrhythmias caused by leading circle reentry, in which the wave front propagates in the just-recovered myocardium of the refractory tail and in which there is only a very small partially excitable gap, are inherently unstable and often terminate after a short period of time.19 20 On the other hand, we have shown that the reentrant wave front in nonuniform anisotropic reentrant circuits is, in general, not propagating in myocardium that has just recovered excitability and that the gap may be quite large. This property may contribute to the stability of these reentrant circuits, which enables anisotropic reentry to cause sustained monomorphic ventricular tachycardia. In addition, the presence of a significant temporal and spatial excitable gap in some anisotropic reentrant circuits, despite the functional nature of the circuits, enables reentry to be terminated by single premature stimuli or by overdrive stimulation.13 44 Termination of arrhythmias by stimulation would be expected to be much more difficult when the reentrant circuit has only a small partially excitable gap.
This study was supported by grant R37 HL-31393 and program project grant HL-30557 from the National Heart, Lung, and Blood Institute, National Institutes of Health.
Previously published as a preliminary report in abstract form (Circulation. 1993;88[pt 2, suppl I]:I-117).
- Received June 12, 1997.
- Accepted October 23, 1997.
- © 1998 American Heart Association, Inc.
Mines GR. On dynamic equilibrium in the heart. J Physiol (Lond).. 1913;46:349–383.
Almendral JM, Stamato NJ, Rosenthal ME, Marchlinski FE, Miller JM, Josephson ME. Resetting response patterns during sustained ventricular tachycardia: relationship to the excitable gap. Circulation. 1986;74:722–730.
Allessie MA, Kirchhof C, Schefter GJ, Chorro F, Brugada J. Regional control of atrial fibrillation by rapid pacing in conscious dogs. Circulation. 1991;84:1689–1697.
KenKnight BH, Bayly PV, Gerstle RJ, Rollins DL, Wolf PD, Smith WM, Ideker RE. Regional capture of fibrillating ventricular myocardium: evidence of an excitable gap. Circ Res. 1995;77:849–855.
Boersma L, Brugada J, Kirchoff C, Allessie M. Mapping of reset of anatomical and functional reentry in anisotropic ventricular myocardium. Circulation. 1994;89:852–862.
Reiter MJ, Zetalaki Z, Kirchof CJ, Boersma L, Allessie MA. Interaction of acute ventricular dilatation and d-sotalol during sustained reentrant ventricular tachycardia around a fixed obstacle. Circulation. 1994;89:423–431.
Pertsov AM, Davidenko JM, Salomonsz R, Baxter WT, Jalife J. Spiral waves of excitation underlie reentrant activity in isolated cardiac muscle. Circ Res. 1993;72:631–650.
Girouard SD, Pastore JM, Laurita KR, Gregory KW, Rosenbaum DS. Optical mapping in a guinea pig model of ventricular tachycardia reveals mechanisms for multiple wavelengths in a single reentrant circuit. Circulation. 1996;93:603–613.
Dillon SM, Allessie MA, Ursell PC, Wit AL. Influence of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res. 1988;63:182–206.
Wit AL, Dillon SM, Coromilas J, Saltman AE, Waldecker B. Anisotropic reentry in the epicardial border zone of myocardial infarcts. In: Jalife J, ed. Mathematical Approaches to Cardiac Arrhythmias. New York, NY: Annals of the New York Academy of Sciences; 1990;591:86–108.
Waldecker B, Coromilas J, Saltman AE, Dillon SM, Wit AL. Overdrive stimulation of functional reentrant circuits causing ventricular tachycardia in the infarcted canine heart: resetting and entrainment. Circulation. 1993;87:1286–1305.
Callans DJ, Hook BG, Josephson ME. Comparison of resetting and entrainment of uniform sustained ventricular tachycardia: further insights into the characteristics of the excitable gap. Circulation. 1993;87:1229–1238.
Peters NS, Coromilas J, Costeas CA, Deruyter B, Wit AL. Variation of the excitable gap permits preexcitation without reset of reentrant ventricular tachycardia circuits. Circulation. 1993;88(pt 2, suppl I):I-117. Abstract.
Coromilas J, Saltman AE, Waldecker B, Dillon SM, Wit AL. Electrophysiological effects of flecainide on anisotropic conduction and reentry in infarcted canine hearts. Circulation. 1995;91:2245–2263.
El-Sherif N. The figure 8 model of reentrant excitation in the canine postinfarction heart. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology and Arrhythmias. New York, NY: Grune & Stratton; 1985:363–378.
Roberts DE, Hersh LT, Scher AM. Influence of cardiac fiber orientation on wavefront voltage, conduction velocity and tissue resistivity in the dog. Circ Res. 1979;44:701–712.
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.
Allessie MA, Lammers WJEP, Bonke FIM, Hollen J. Intraatrial reentry as a mechanism for atrial flutter induced by acetylcholine and rapid pacing in the dog. Circulation. 1984;70:123–135.
Bernstein RC, Frame LH. Ventricular reentry around a fixed barrier with advancement in an in vitro model. Circulation. 1990;81:267–280.
Spinelli W, Hoffman BF. Mechanisms of termination of reentrant atrial arrhythmias by class I and class III antiarrhythmic agents. Circ Res. 1989;65:1565–1579.
Boersma L, Brugada J, Abdollah H, Kirchof C, Allessie M. Effects of heptanol, class 1c, and class III drugs on reentrant ventricular tachycardia: importance of the excitable gap for the inducibility of double wave reentry. Circulation. 1994;90:1012–1022.
Brugada J, Boersma L, Kirchof C, Heynen V, Allessie M. Reentrant excitation around a fixed obstacle in uniform anisotropic ventricular myocardium. Circulation. 1991;84:1296–1306.
Davidenko JM, Salmonsz R, Pertsov AM, Baxter WT, Jalife J. Effects of pacing on stationary reentrant activity: theoretical and experimental study. Circ Res. 1995;77:1166–1179.
Kirchoff C, Chorro F, Scheffer GJ, Brugada J, Komings K, Zetelaki Z, Allessie M. Regional entrainment of atrial fibrillation studied by high resolution mapping in open-chest dogs. Circulation. 1993;88:736–749.
Niwano S, Ortiz J, Abe H, Gonzalez X, Rusdy Y, Waldo AL. Characterization of the excitable gap in a functionally determined reentrant circuit: studies in the sterile pericarditis model of atrial flutter. Circulation. 1994;90:1997–2014.
Chen Peng-Sheng, Wolf PD, Dixon EG, Daniely ND, Frazier DW, Smith WM, Iedeker RE. Mechanism of ventricular vulnerability to single premature stimuli in open-chest dogs. Circ Res. 1988;62:1191–1209.
Frazier DW, Wolf PD, Wharton JM, Tang ASL, Smith WM, Iedeker RE. Stimulus induced critical point: mechanism for electrical initiation of reentry in normal canine myocardium. J Clin Invest. 1989;83:1039–1052.
Frame LH, Hoffman BF. Mechanisms of tachycardia. In: Surawicz B, Reddy CP, Prystowsky EN, eds. Tachycardias. The Hague, the Netherlands: Martinus Nijhoff; 1984:276–301.
Smeets JLRM, Allessie MA, Lammers WJEP, Bonke FIM, Hollen J. The wavelength of the cardiac impulse and reentrant arrhythmias in isolated rabbit atrium: the role of heart rate, autonomic transmitters, temperature and potassium. Circ Res. 1986;58:96–108.
Rensma PL, Allessie MA, Lammers WJEP, Bonke FIM, Schalij MJ. Length of excitation wave and susceptibility to reentrant atrial arrhythmias in normal conscious dogs. Circ Res. 1988;62:395–410.
Rosenthal ME, Stamato NJ, Almendral JM, Gottlieb CD, Josephson ME. Resetting of ventricular tachycardia with electrocardiographic fusion: incidence and significance. Circulation. 1988;77:581–588.
Gottlieb CD, Rosenthal ME, Stamato NJ, Frame LH, Lesh MD, Miller JM, Josephson ME. A quantitative evaluation of refractoriness within a reentrant circuit during ventricular tachycardia: relation to termination. Circulation. 1990;82:1289–1295.
Cabo C, Pertsov AM, Baxter WT, Davidenko JM, Gray RA, Jalife J. Wave-front curvature as a cause of slow conduction and block in isolated cardiac muscle. Circ Res. 1994;75:1014–1028.
Fast VG, Kleber AG. Role of wavefront curvature in propagation of the cardiac impulse. Cardiovasc Res. 1997;33:258–271.
Winfree AT. Spiral waves of chemical activity. Science. 1972;175:634–636.
Restivo M, Gough WB, El-Sherif N. Ventricular arrhythmias in the subacute myocardial infarction period: high-resolution activation and refractory patterns of reentrant rhythms. Circ Res. 1990;66:1310–1327.
Peters NS, Severs NJ, Coromilas J, Wit AL. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation. 1997;95:988–996.
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;57:432–442.
Schalij MJ, Lammers WJEP, Rensma PL, Allessie MA. Anisotropic conduction and reentry in perfused epicardium of rabbit left ventricle. Am J Physiol. 1992;263:H1466–H1478.
Lammers WJEP, Wit AL, Allessie MA. Effects of anisotropy on functional reentrant circuits: preliminary studies of computer simulation studies. In: Sideman S, Bayer R, eds. Activation, Metabolism and Perfusion of the Heart. Dordrecht, the Netherlands: Martinus Nijhoff; 1987:133–150.
El-Sherif N, Gough WB, Restivo M. Reentrant ventricular arrhythmias in the late myocardial infarction period, 14: mechanisms of resetting, entrainment, acceleration or termination of reentrant tachycardia by programmed electrical stimulation. Pacing Clin Electrophysiol. 1987;10:341–371.