Importance of Location and Timing of Electrical Stimuli in Terminating Sustained Functional Reentry in Isolated Swine Ventricular Tissues
Evidence in Support of a Small Reentrant Circuit
Background In excitable chemical media, a spiral wave is formed by reentrant excitation around a core and normal propagation away from the core. Whether or not this applies to cardiac muscle is unknown.
Methods and Results In six isolated swine ventricular slices, we induced sustained episodes of functional reentry with a stationary core. A train of stimuli applied away from the core (7- to 8-mm distance) and near the core (within 1.6 mm) terminated 5 of 24 and 14 of 17 episodes of reentry, respectively (P<.001). When the stimulus was applied away from the core, successful terminations occurred when the line connecting the stimulus and the core was along the myocardial fiber orientation and when the coupling interval was 54±11% of the reentrant cycle length. Stimulation near the core terminated reentry primarily by propagation of the stimulus-induced wave fronts that closed up the excitable gap. However, in two episodes, the application of a stimulus near the core changed the electrogram morphology in only four bipolar pairs. This was sufficient to cause abrupt termination of reentry.
Conclusions (1) A thin layer of activation near the core is responsible for the maintenance of functional reentry. (2) Access to the tissue near the core is essential for the termination of functional reentry by a point stimulus. (3) To terminate reentry with a stimulus away from the core, the stimulus must occur at certain critical coupling intervals, and the line connecting the stimulus and the core must be roughly parallel to the fiber orientation.
Functional reentry, defined by reentrant excitation without an anatomic obstacle, has been demonstrated in cardiac tissues both in vitro1 2 and during fibrillation in vivo.3 The actual size of the reentrant circuit necessary to maintain the reentrant excitation (critical size) was unclear. In excitable media,4 the critical size of a reentrant excitation appears to be limited to a very small area around the central circular region (ie, the core), where the wave fronts travel tangentially. In the remaining portion of the reentrant wave front, the activation propagates away in the direction that is normal to the core and is not essential in sustaining reentry. If this notion is applicable to reentry in cardiac tissue, then the reentrant excitation occurs only in a small area around the core. This small area is the “rotor,” which is defined as reentrant excitation that serves as the source of outward-radiating waves.5 The remaining portion of the cardiac tissue is activated passively by the rotor and does not participate in reentry. A corollary of this hypothesis is that a premature stimulus can terminate reentry only when the stimulus itself or the wave front induced by the stimulus has direct access to the tissue near the core of the reentrant wave to eliminate the rotor. Conversely, electrically induced wave fronts that have no access to the tissue near the core may still change the patterns of activation but will have no effect on the maintenance of reentry. To test this hypothesis, we have developed a model of stationary functional reentry in isolated ventricular tissues.6 7 Unlike the model reported by Davidenko et al,8 in our model multiple pacing stimuli could be given during the same episode of functional reentry without changing the reentrant cycle length or the location of the core. This feature allowed us to systematically determine the effects of stimulus location and timing on a stationary functional reentrant wave front. The purpose of the present study was to use computerized mapping techniques to determine the patterns of activation when pacing stimuli were given at different timings and locations during a stable and stationary episode of functional reentry. We tested the hypothesis that reentrant excitation is controlled only by a thin region around the core that is essential for the maintenance of the functional reentry. The remaining portion of the reentrant wave is not essential for reentrant excitation but is activated passively by the reentrant wave fronts near the core. Access to the tissue near the core by the stimulus itself or by the wave front initiated by the stimulus is important in terminating functional reentry.
The research protocol was approved by the Institutional Animal Care and Use Committee of the Cedars-Sinai Medical Center and followed the guidelines of the American Heart Association.
Five swine (50 to 60 lb) were used in this study. Each was premedicated with an intramuscular injection of ketamine (20 mg/kg), acepromazine (0.5 mg/kg), and atropine (0.05 mg/kg). The swine were then anesthetized with 20 mg/kg IV thiopental sodium, intubated, and ventilated with a Harvard respirator. The chest was opened via a median sternotomy. The heart was quickly removed and placed in cold Tyrode’s solution gassed with 95% O2/5% CO2. Two 30×30-mm sections of myocardium were excised from the right ventricular free wall. The endocardial side of the preparation was removed by gentle dissection with scissors. This resulted in epicardial tissue blocks <2 mm thick. The first tissue was mounted in a tissue bath with the epicardial side on the mapping electrode array and was superfused with Tyrode’s solution gassed with 95% O2/5% CO2. The tissue was held in place by stainless steel insect pins. The second tissue was kept in the 4°C oxygenated Tyrode’s solution to maintain viability. After the first tissue was studied, we attempted to study the second tissue. One second tissue was successfully studied. The composition of the Tyrode’s solution was as follows (mmol/L): NaCl 125.0, KCl 4.5, MgCl2 0.5, CaCl2 2.7, NaH2PO4 1.2, NaHCO3 24.0, and glucose 5.5.9 The size of the tissue bath was 10.7×7.2×2.7 cm. The flow rate of the Tyrode’s solution was 8 to 10 mL/min. The temperature was maintained at 36.0°C and the pH at 7.4±0.1. After the tissue was stabilized with constant pacing for 30 to 60 minutes, stimulation and recording protocols were performed. The total duration of each experiment was ≈3 to 4 hours.
A 3.8×3.2-cm plaque electrode array made of 509 bipolar electrodes in 21 columns and 25 rows was placed at the bottom of the tissue bath. A schematic of this electrode array has been published.9 Recording electrodes were connected to a computerized mapping system (EMAP, Uniservices).10 The electrograms were filtered from 0.5 Hz to 3 MHz and were acquired at 1000 samples per second with 16 bits of accuracy.11
Two platinum pacing electrodes with a diameter of 0.3 mm (Grass Instruments) were hooked onto the edge of the tissue. A pair of silver stimulation electrodes was built in the center of the mapping plaque. Pacing threshold was determined with constant pacing at 300- or 500-ms cycle length. The lowest current output (in milliamperes) that resulted in stable capture was the pacing threshold. The output was adjusted to twice diastolic threshold current for baseline pacing and for giving train stimuli during reentry. The effective refractory period was then determined with the extrastimulus technique using twice diastolic threshold current given at the same pacing site, with a baseline driving cycle length of 500 ms.
Patterns of activation were mapped during pacing at a 300- or 400-ms cycle length from both the edge and the center of the tissues. Afterward, the hook electrodes at the edge of the tissue were used to deliver baseline pacing (S1) with 5-ms pulse width at twice diastolic threshold current. After eight S1 stimuli at a cycle length of 300 ms, a strong S2 stimulus was delivered to the center of the pacing plaque to induce reentry.3 6 An advantage of this model is that the core of reentry is usually at a predicable location, ie, near the site of S2. After the induction of reentry, stimuli can then be given through the same pacing electrode to observe the effects of pacing near the core. If sustained reentry was not induced, cromakalim (5 μmol/L) was added to the superfusate. Cromakalim, a KATP channel opener, is known to significantly shorten the action potential duration and facilitate the induction of reentry.7 Twenty minutes after stable cromakalim superfusion, the same pacing protocol was performed to induce sustained reentry.
Once sustained reentry was induced, trains of stimuli at twice diastolic threshold current were given at fixed cycle lengths (300 to 400 ms) to either the edge (S1 site) or the center (S2 site) of the tissue. The patterns of activation during pacing and at the time of termination were acquired for off-line analysis. Eight seconds of data were acquired each time. The tissues were then fixed with 10% buffered formalin for at least 48 hours. All tissues were then embedded in paraffin and were stained with hematoxylin and eosin to determine the myocardial fiber orientation and to rule out the presence of scar or necrotic tissues that might serve as anatomic barriers to electrical impulse propagation.
The methods of data analysis have been reported in detail elsewhere.11 Briefly, the times of activation were taken as the time of the fastest slope (dV/dt) of each electrogram by the computer. Manual editing was then performed for each activation for each episode of reentry. A dynamic display of the activation patterns of each episode of reentry was visualized on a computer screen in which each electrode site is represented by a dot. For each activation registered, a software program directed the corresponding dot on the computer monitor to illuminate. The dot initially illuminated red, then yellow, followed by green, light blue, and finally dark blue. Each color persisted for 20 ms.
The core is defined as an area that is excitable but remains unexcited and around which reentry occurs.9 Because of the effects of cromakalim, the core was small and an unexcited area cannot be visualized. For this study, we define core as an area encircled by the innermost edge of the reentrant wave front. Myocardial fiber orientation was determined by histological examination. The line connecting the stimulus site and the core of reentry was along the fiber orientation if the angle formed by this line and the long axis of the fiber was between −15° and +15°. It was across the fiber orientation if the angle was between 75° and 105°. Otherwise, the line was considered to be oblique to myocardial fiber orientation.
Statistical analyses were performed with GB-Stat.12 The data are presented as mean±SD. Student’s t tests were used to compare means. χ2 analysis was used to test the frequency distribution of success or failure to terminate reentry by stimuli given at the edge of the tissue, with the line connecting the stimulus site and the core of reentry along, across, or oblique to the myocardial fiber orientation. A value of P≤.05 was considered significant.
Reentrant excitations were inducible in six tissues. The reentry was nonsustained in five tissues at baseline. In these tissues, superfusion with cromakalim was necessary to sustain reentry. The S1S2 interval used to induce reentry ranged from 130 to 270 ms. The S2 strength needed to induce sustained reentry was 10 to 15 mA. In all tissues, fixed-rate pacing from the edge of the tissue and from the center showed no evidence of conduction block. Fig 1A⇓ shows an example of pacing from the edge of the tissue, which showed orderly conduction of the activation from the pacing site to the other edge of the tissue. Fig 1B⇓ shows pacing from the center. Conduction was anisotropic and was faster parallel to the fiber orientation than across the fiber orientation.
Electrophysiological Effects of Cromakalim
Cromakalim resulted in significant shortening of the effective refractory period measured at 500-ms pacing cycle lengths (184±13 ms at baseline versus 100±24 ms during cromakalim infusion, P<.001). The pacing threshold, the conduction velocity along the fiber orientation, and the conduction velocity across the fiber orientation were 0.37±0.13 mA, 0.36±0.05 m/s, and 0.20±0.05 m/s, respectively, at baseline. These values did not change significantly during cromakalim administration (0.64±0.25 mA, 0.44±0.09 m/s, and 0.19±0.05 m/s, respectively). These conduction velocities are similar to those reported by other investigators using normal canine13 14 and swine15 tissues.
Effects of Stimulation Away From the Core of Functional Reentry
A total of 303 stimuli were applied to the edge of ventricular tissues in 24 episodes of sustained reentry. The cycle length averaged 134±31 ms. In all episodes, the core was stationary, was within 1.6 mm of the S2 site, and was at least 5 mm from the S1 site. The probability of local capture was a function of the coupling interval between the preceding activation and the time of the stimulus. Fig 2A⇓ shows the relationship between probability of capture and the coupling interval as a percentage of the cycle length of reentry when stimulus is applied at the S1 site, away from the core. Fig 2B⇓ shows the same relationship when the external stimulus is applied at the S2 site and will be discussed later. In Fig 2A, the probability of local capture was nearly 100% when the coupling interval was >60% of the cycle length and was 0% when the coupling interval was <30% of the cycle length. Between 30% and 60% of the cycle length, the probability of capture progressively increased. These findings are compatible with the notion that refractory period of cardiac tissue is a probability function.16
Only 5 of the 303 stimuli (1.65%) delivered during 24 episodes of reentry successfully terminated reentry. Analyzing the patterns of activation associated with all 303 stimuli showed the following mechanisms of failure to terminate reentry: (1) The stimulus did not result in local capture (n=134). The maximum time interval between the preceding local activation and the time of the stimulus was 39.4±12.1 ms (from 0 to 74 ms, which corresponds to 0% to 59.8% of the reentrant cycle length). (2) The stimulus resulted in local capture but failed to activate distal sites, including the core (n=164). The minimum coupling interval between the preceding local activation and the time of the stimulus averaged 49.3±11.6 ms. The inability of the induced front to reach the distal core was caused by the presence of a layer of depolarized tissue between the induced front and the core that protected the core. The absence of the core access is reflected by the full compensatory pause after the premature depolarization, indicating that the reentry was undisturbed. In no episodes did the stimulus advance (reset) the reentrant cycle length without terminating the reentry.
Successful termination (n=5) was always associated with the wave fronts that reached the core region. The following are some of the examples of the effects of pacing from the edge of the tissue.
Failure to Terminate Functional Reentry by Pacing From the Edge of the Tissue
The vast majority of the pacing stimuli given to the edge of the tissue failed to terminate reentry. Fig 3⇓ shows an example. The reentry was counterclockwise and had a cycle length that alternated between 126 and 115 ms. Start of data acquisition was taken as the zero reference time. The first four frames of Fig 3A⇓ (from t=2310 to t=2430 ms) correspond to one complete rotation of the functional reentry. The core was very small, and its size could not be measured accurately. A solid white line is used to indicate the location of the core, which is near the center of the tissue. The length and width of the line do not necessarily indicate the exact size of the core. The core was stretched along the fiber orientation. The core is shown only in the first four frames of Fig 3A⇓. At time t=2440 ms, a premature stimulus was applied to the lower edge of the tissue (marked by an asterisk in the t=2460 ms frame). As the arrows indicate, the stimulus resulted in local capture and started to propagate in two directions (arrows pointing left and right). However, the area between the core and the stimulus was already activated by the reentrant wave front (the red dots below the curved arrow in the 2460-ms frame). This layer of depolarized tissue effectively prevented the core from being activated by the stimulus-initiated wave front. The frame t=2475 ms shows that the sustained reentrant wave front collided with the leftward wave generated by the external stimulus. The collision occurred away from the core and did not perturb the reentrant circuit near the core. The t=2510 ms frame in Fig 3A⇓ shows that the reentrant front continued its original path.
Fig 3B⇑ shows the map of the V1V2 interval. V1 is the activation immediately before the stimulus, and V2 is the activation immediately after the stimulus. In this map, we show the time interval between the two activations that bracket the time of stimulation. On the map, the location of the core is represented by the solid ellipse. The dashed triangular line shows the area where the stimulus was applied and where the wave collision took place. The area between the wave collision and the core (roughly a 3.2-mm-wide pathway) is large enough to sustain the reentrant activation. Although the electrodes in the triangular area showed shorter cycle length, the activation times in other parts of the tissue were largely unchanged. Note that the cycle length of the reentry alternated between roughly 115 and 126 ms. The stimulus was given while the cycle length was short (115 ms). Therefore, the V1V2 interval of 113 to 118 ms seen in most of the electrode sites represents the intrinsic (native) cycle length that remains undisturbed by the stimulus. The V1V2 intervals of 78 to 98 ms near the site of stimulus represent intervals that were actually shortened by the stimulus. In some channels, the cycle lengths were >120 ms. These channels were also activated by the intrinsic (native) rhythm, which alternated to a longer cycle length. These activations were also not perturbed by the premature stimulus.
Some representative electrograms from the wave collision area are shown in Fig 3C⇑. In these channels, the wave collision can be seen as a change in electrogram morphology of V2 in all channels. Electrograms of channels 369, 370, and 391 show that even the earliest premature beats were followed by a fully compensatory pause (160 ms). The electrogram morphology of the subsequent activations (V3) was the same as before the stimulus in all channels. Therefore, the reentrant circuit was not reset by the stimulus. This example illustrates that if an induced wave front does not have access to the tissue near the core, it fails to terminate or change the characteristics of the reentry.
Termination of Functional Reentry by Pacing From the Edge of the Tissue
Fig 4A⇓ shows a series of frames during reentry with a cycle length of 127 ms. The gray area in the top left frame represents the size of the tissue relative to the electrode plaque, and the double-headed red arrow indicates the fiber orientation of the tissue. This tissue is smaller than the plaque; therefore, no area in the tissue was left unmapped. The circulating wave propagated in a clockwise direction around a region of functional block. This area of functional block is shown as a solid white line (center of tissue) in the next four frames of Fig 4A⇓. The frames t=890 to t=1015 ms represent one rotation of the functional reentry. The core was stationary and was close to the site of the S2 stimulus. The site of the application of the stimulus (the S1 site) is marked by an asterisk (t=1025 ms). Fig 4B⇓ shows bipolar recordings during the same episode of reentry. The arrows point to premature stimuli that were given within 1.6 mm of the bipolar recording electrode 387. The first premature stimulus was given at a coupling interval of 37 ms and failed to result in local capture. There was no change of cycle length. The second premature stimulus with a coupling interval of 89 ms shortened the cycle length to 96 ms. The horizontal axis in these recordings represents the time scale, with the initiation of data acquisition as the zero reference time. As the upward arrows indicate in the t=1025 ms frame of Fig 4A⇓, the application of the external stimulus initiated an antidromic and an orthodromic wave front. The antidromic wave front collided with the original arm of the reentrant front (downward arrow, t=1035 ms) on the right side of the region of functional block. The result of the collision of these two fronts is the annihilation of functional reentry (right side of frame t=1060 ms). This collision is evident from the change in morphology of the electrograms from the boxed region of the t=1060 ms frame shown in Fig 4C⇓. In the meantime, the frames t=1060 ms through t=1120 ms indicate that the paced orthodromic wave front encountered refractory tissue and terminated. The double white lines in the t=1120 ms frame show the area of conduction block. Selected electrograms from this region are shown in Fig 4D⇓. The arrow points to the low-amplitude deflection recorded on channels 179, 200, and 220. These deflections were of lower amplitude than the bipolar electrograms during reentry registered by the same channels. Reentry terminated after these deflections.
Factors That Determine the Success or Failure of Stimulation From the Edge of the Tissue
Two factors appear to be related to the success or failure of a stimulus from the edge of the tissue to terminate functional reentry. One factor is the coupling interval. The coupling interval (the time interval between local bipolar electrogram and time of stimulus) for successful episodes averaged 87±23 ms, with a wide range of 69 to 126 ms. However, because the cycle length of reentry of these episodes also varied from 127 to 185 ms, these coupling intervals were actually clustered within a narrow range (54±11%) of the reentrant cycle length. This clustering indicates that the timing of the stimulus is critically important in terminating a functional reentrant wave front, a finding compatible with the “protective zone” concept.10 17
A second factor that determines the success or failure of the stimulus is the fiber orientation. Among the 169 stimuli from the edge of the tissue that resulted in local capture, five were given along the fiber orientation. All these five episodes resulted in successful termination of reentry. When the line connecting the stimulus site and the core was across (n=79) or oblique to (n=85) myocardial fiber orientation, none terminated reentry (P<.001).
Effects of Stimulation Near the Core of Reentry
We gave 157 stimuli near the core (S2 area) during 17 episodes of sustained reentry. The cycle length averaged 129±25 ms. In all episodes, the core was stationary and was close to the S2 site in the center of the tissue. The ability to achieve local capture was dependent on the coupling interval (Fig 2B⇑). However, there were significant differences compared with stimuli applied away from the core at the S1 site (Fig 2A⇑). In Fig 2A⇑, almost all stimuli with a coupling interval >60% of the cycle length of reentry resulted in capture. In Fig 2B⇑, however, the slope was shallower. Except for the coupling intervals >90% of the cycle length of reentry, stimuli with shorter coupling intervals often failed to capture. Although the probability of capture progressively increased when the coupling interval lengthened, it did not reach 90% until the coupling interval was almost as long as the cycle length of reentry.
Among these 157 stimuli delivered during 17 episodes of reentry, 14 (8.92%) terminated the reentry (P<.001 versus the efficacy of pacing at the S1 site in terminating reentry). The remaining stimuli failed to terminate reentry. The mechanism of failure included the following: (1) The stimulus did not result in local capture (n=89). The interval between immediately preceding activation and the time of stimulus in these episodes varied between 0 and 233 ms or 0% and 89.6% of the cycle length. (2) The stimulus resulted in local capture but did not advance the cycle length of reentry (n=38); ie, a full compensatory pause was present. (3) The stimulus advanced the cycle length of reentry, indicating that the wave front initiated by the stimulus had penetrated the reentrant circuit and closed up a portion of the excitable gap. However, the cycle length reduction was insufficient to terminate the reentry (n=12). (4) The pattern of reentry was changed into a figure-eight pattern (n=4). The following are some of the examples.
Failure to Terminate a Functional Reentry by Pacing the Center
Fig 5⇓ shows an example of a premature stimulus that is delivered very close to the core of the reentrant circuit and failed to terminate functional reentry. The first four frames of Fig 5A⇓ show one complete rotation of functional reentry, which was counterclockwise (in the direction of the arrows) and had a cycle length of 112 ms. The core of the reentry was narrow and stretched from left to right along the fiber orientation of the tissue. The core is shown by a solid white line in the center of the tissue in the first four frames of Fig 5A⇓. A stimulus was applied to the center of the tissue (marked by a white asterisk) in the frame t=3005 ms. This premature stimulus initiated a new wave front that interacted with the sustained front in an antidromic and orthodromic manner. The boxed region in the frame t=3030 ms represents the area that was captured earliest. As the arrows in the t=3030 ms frame show, this new wave front propagated toward the top edge of the tissue, while the sustained reentrant front tried to make the turn in the right side of the frame. In the t=3050 ms frame, the original wave front collided with this antidromic front and was prevented from propagating any farther (t=3050 ms). The boxed region of the t=3060 ms frame shows the continuation of the orthodromic wave front, which was clearly initiated from the site of external stimulation and was not the continuation of the original reentrant wave front. The electrograms of Fig 5B⇓ taken from this boxed region show early captured activations in channels 264 to 286. The last two frames (t=3125 ms and t=3160 ms) show that the reentrant front has formed again. The electroexternal stimulus has shortened the cycle length (V1V2 interval) to 89 ms. The sum of the V1V2 and V2V3 intervals (201 ms) was less than twice the original cycle length of 112 ms, and the original reentrant circuit was reset by 23 ms. The difference between this case and the case in which the application of the premature stimulus caused termination is that in this case, the orthodromic wave front did not close up the excitable gap completely, and reentry continued. Fig 5C⇓ shows the map of the V1V2 interval. Almost all electrodes registered shorter V1V2 intervals than the reentrant cycle length of 112 ms. Reentry has been reset by the stimulus without termination.
Termination of Functional Reentry by Pacing the Center
Fig 6⇓ shows an example of termination of a functional reentry by a premature stimulus that was delivered very close to the core of the reentrant circuit. Fig 6A⇓ is a sequence of frames that shows one rotation of the functional reentry as well as the events that occurred after the stimulus was applied. The first four frames of Fig 6A⇓ show one rotation of the functional reentry. As the arrows indicate, the reentry was counterclockwise and had a cycle length of 112 ms. The core of the reentry is represented by a solid white line in the first four frames of this panel. At t=5000 ms, a stimulus was applied to the center of the tissue. The point of application of the stimulus is marked with an asterisk. The earliest areas captured by the stimulus are represented by the boxed region in the t=5020 ms frame. The application of the premature stimulus started a new front that propagated in two directions. One propagated very rapidly along the fiber orientation to the right. The other new front propagated toward the bottom of the frame at a much slower rate (t=5040 ms). Selected electrograms from the boxed region of the t=5050 ms frame indicate the presence of activations with new morphology (channels 263 and 285 in Fig 6B⇓). The interaction between the stimulus-initiated front and the sustained reentrant front resulted in the annihilation of the sustained front. This termination is due to the direct collision of the sustained reentrant front with the refractory tail of the stimulus-initiated front. The boxed region of the frame t=5065 ms shows the newly initiated front at the far right side of the tissue. The electrograms recorded in this box were activated 38 ms earlier than they would have been activated without the stimulus. The arrow in the t=5080 ms frame indicates that the new front propagated from the right edge of the tissue toward the left edge of the tissue. However, as indicated by the double solid line in the t=5095 ms frame, this propagation failed. The existence of a depolarized region near the core may be the reason for this failure of propagation. The electrograms in Fig 6C⇓ are selected from around the core area. The activations after stimulus artifact in these tracings show a decremental conduction in this region (small electrogram in channel 241). The refractoriness of the tissue in this region is probably the cause of the termination of the new front.
The gray area of the frame on top of Fig 6D⇑ shows the size of the tissue relative to the electrode plaque, and the double-headed arrow represents the direction of fiber orientation. The same electrode that was used to apply the external stimulus during reentry is used to pace the tissue immediately after the termination of reentry. The lower three frames of Fig 6D⇑ show that the activated front propagated in the horizontal direction much faster than the vertical direction. This observation is consistent with the fact that the captured stimulus during reentry propagated much faster across the lower region of the core and resulted in the shortening of the reentrant cycle length. Fig 6E⇑ shows the map of the V1V2 interval. The numbers show the time difference between the activations that bracketed the stimulus artifact. The core is the solid ellipse in the center of the tissue. The asterisk is the site of stimulation. The dashed line shows the area of propagation of the newly induced front. The double line segments at left center show the area of block. The block and termination of reentry may be related to hitting the refractory tissue because of the ≈20 ms premature arrival of this orthodromic wave front.
Termination of Functional Reentry Without Inducing Antidromic or Orthodromic Wave Fronts
Fig 7⇓ demonstrates the abrupt termination of a single reentrant wave front after the application of an external stimulus near the reentrant core. In Fig 7A⇓, the arrows indicate the direction of the rotation of the reentrant wave front, which was induced with an S1S2 protocol at baseline (without cromakalim) and had a cycle length of 215 ms. The core of reentry was elliptical and stretched along the fiber orientation. During the first cycle of the reentry (frames t=4270 ms to t=4445 ms), the area marked by “A” in the 4365-ms frame remained inactivated. This area showed a 2-to-1 conduction block (electrograms of Fig 7B⇓). Frames t=4505 ms to t=4685 ms show that during the subsequent cycle of the reentry, this region was activated (frame 4630 ms) by the reentrant wave front. The distance between the closest site of block in the “A” region and the core was 5 to 7 mm. This indicates that the actual reentrant circuit (the rotor) should have a thickness (radius) ≤5 to 7 mm in that direction. At t=4725 ms, a stimulus was applied to the tissue at the site marked by a white asterisk. The last activation in Fig 7C⇓ shows that, after the application of the stimulus, only four electrodes (184, 185, 204, and 205) showed changes in electrogram morphology. These channels were in a rectangular area in the t=4775 ms frame of Fig 7A⇓. Conversely, electrodes 159, 160, and 164 did not register a change in signal morphology. Afterward, the activation propagated toward the top of the tissue and terminated. In this episode, the reentry terminated without creating a new antidromic or orthodromic wave front.
Site of Stimulus Relative to the Core
In one tissue, we induced a total of seven episodes of reentry, with the cores located at six different sites all around the site of stimulus (Fig 8⇓). The lines connecting the cores and the site of stimulus could be either along or across the myocardial fiber orientation. All seven episodes were successfully terminated by trains of 7 to 18 stimuli applied at that stimulus site. These findings show that, in contrast to applying a stimulus distant from the core, the point stimuli applied at or near the core can terminate reentry regardless of the relative relationship with myocardial fiber orientation.
Conversion of a Single Reentrant Wave Front to a Figure⇑-Eight Reentry by Pacing the Center
Fig 9⇓ demonstrates the formation of a figure-eight reentry from a single reentrant wave front after the application of an external stimulus delivered to the center of the tissue. Fig 9A⇓ shows a series of frames of this episode. The frames t=4075 ms through t=4225 ms represent one complete rotation of the reentrant front. The reentry was counterclockwise, with a cycle length of 159 ms. An external stimulus was applied at t=4225 ms and is marked by an asterisk in the t=4225 ms frame. The wave front generated by this stimulus propagated counterclockwise and reached site “a” of the t=4290 ms frame 20 ms prematurely. The interaction between this new front and the tail of the sustained reentrant front resulted in the formation of a wave break that is represented as double lines in the t=4330 ms frame. The frames at t=4370 ms and t=4390 ms show that this wave break formed a new wave front that coexisted with the original sustained front to produce sustained figure-eight reentry. Fig 9B⇓ shows electrograms recorded from the area where the wave break forms. The bottom four channels show the reversal of activation sequences, corresponding to the formation of the second loop of reentry that was caused by the application of an external stimulus. These electrograms indicate that the wave break occurs near channel 278.
Importance of Timing and Location of the Stimulus on Terminating Functional Reentry
In these experiments, we demonstrated that functional reentry can be terminated by application of a point stimulus of twice diastolic threshold current. We used a model of functional reentry that maintains a stable cycle length and the location of the core during train stimulation at 300 to 400 ms cycle lengths if the core was not invaded. This model allowed us to test the effects of different timing and location of the stimuli on functional reentrant wave fronts. The major finding is that the location of the stimulus is critically important in determining the outcome of pacing. We found that a stimulus applied near the core has a higher probability of terminating reentry than a stimulus given at the edge of the tissue. These findings imply that the critical size that is necessary for maintaining reentry (ie, the size of a rotor) may indeed be very small. Accurate positioning of the stimulating electrode with respect to the reentrant circuit is important for successful termination of reentry. Recently, it was proposed that meandering functional reentrant waves may underlie the generation and maintenance of polymorphic ventricular tachycardia and ventricular fibrillation.18 If our results are applicable to meandering functional reentrant wave fronts, then it may be possible to terminate these arrhythmias by giving an appropriately timed stimulus to the core when it visits the site of the stimulating electrode. For example, the appearance of double potentials in the electrogram of the stimulating electrode may indicate that the core is nearby.
A second finding is that reentry terminated when a stimulus was given at 32 to 116 ms after the previous activation (roughly 23.4% to 72.5% of the reentrant cycle length). This finding is similar to that reported by Bonometti et al,10 who used much stronger current to terminate functional reentrant wave fronts and found that the best timing to terminate figure-eight reentry is ≈40 ms after the central common pathway is excited. Because of the existence of this critical timing window, a protective zone is present in the functional reentrant circuit during which an electrical stimulus can terminate reentry.
Critical Size of the Functional Reentrant Circuit
Although reentrant excitation in this study appeared to activate the entire tissue in the form of a spiral-shaped wave, the results of the present study indicate that only a small area close to the core is crucial to the maintenance of the reentrant activity. The following observations support this contention: (1) A stimulus without any access to that small area around the core could capture a significant portion of the mapped region, but reentry would still continue at the same cycle length. (2) Occasionally, 2-to-1 conduction block occurred in the periphery of the tissue while reentry continued uninterrupted at the same cycle length. (3) To terminate functional reentry, the stimulus had to be applied either close to the core or such that the stimulus initiated a propagated wave front that invaded the core region. In two episodes, the stimulus changed the electrogram morphology in only four recording electrodes. Although no antidromic or orthodromic wave fronts were initiated, reentry was abruptly terminated by the stimulus. These findings imply that the maintenance of functional reentry in isolated swine ventricular tissue occurs within a small area near the core. The width of this area extends only 2 to 3 electrode rows (3.2 to 4.8 mm) from the core. The tissue outside this area is passively activated by the reentrant wave front and does not participate in the reentrant excitation.
Presence of an Excitable Gap in Functional Reentry
The leading-circle hypothesis19 of functional reentry proposes that there is little or no excitable gap in the functional reentrant wave front. Accordingly, “the head of the circulating wave front is continuously biting in its own tail of refractoriness.”19 Many investigators subsequently documented the presence of significant excitable gap during functional reentry,2 20 21 which has been attributed to the anisotropic nature of the tissue.22 In this study, we demonstrated that the critical portion of the reentrant circuit was within a few millimeters of the core. A premature stimulus given at that location often could capture the local tissue, if the coupling interval was >20% of the reentrant cycle length. The probability of capture increased with increasing coupling interval but did not reach 100% until the coupling interval was almost the same as the cycle length. These findings indicate that an excitable gap is present even near the core of functional reentry. The magnitude of the excitable gap, however, is not fixed.
The relation between the probability of capture and the reentrant cycle length is much different when the stimulus is given at the edge of the tissue (Fig 2A⇑) rather than near the core (Fig 2B⇑) of functional reentry. The probability of capture increases over a larger range of coupling intervals when the stimulus is applied near the core (Fig 2⇑). This difference suggests that the electrical activation and the excitability characteristics near the core and away from the core may be significantly different. Simulation studies23 showed that the action potentials in the center of a vortex are characterized by the presence of an organizing center (phase singularity) within which the activity is not well defined (ambiguous), in an otherwise rhythmic process. The timing in this region becomes ambiguous because the activation front and the repolarization wake come close to each other near the core. Even slight meandering of the tip of the reentrant wave front can result in large differences of the action potential duration and excitability characteristics at the site of stimulus. Therefore, the relation between the probability of capture and the coupling interval varies greatly near the core but less so when the site of stimulus is distant from the core.
Generation of Spiral Morphology
Reentrant excitation in the form of spiral-shaped waves can be found in many different excitable media and has been clearly documented in cardiac tissue in vitro.2 The mechanisms by which spiral morphology is generated are unclear. Fig 10⇓ illustrates our hypothesis on the generation of spiral morphology. In this figure, a generic core is present around which reentrant excitation occurs. This core can be either anatomic or functional. In Fig 10A⇓, the red dot indicates that reentry started at the 12 o’clock position before rotating in the counterclockwise direction (B through F). While the reentrant wave front circles the core, the remaining tissue is activated by the wave front propagating outward from the core. As time progresses, the activation wave front moves farther out from the core region. In Fig 10F⇓, a curved wave front can be visualized by connecting the outer red dots of the figure. When displayed in a dynamic fashion, this curved wave front would rotate counterclockwise, creating the image of a spiral-shaped wave.
Fig 10⇑ has the following implications: (1) The only portion of the reentrant wave front that is critical for the maintenance of reentry is the layer of activation immediately next to the core. The remaining portion of the reentrant wave front is activated passively by propagation of the wave fronts away from the core. The results of a previous study20 and the present study support this concept because breaking the reentrant wave front near the core was effective in terminating reentry. In the present study, the critical portion of the reentrant wave occupied a thickness of ≈4.8 mm around the core. Abrupt termination of reentry could be achieved by interference with this inner layer of reentry. This concept also explains the mechanisms by which pacing from the edge of the tissue may capture a significant portion of the mapped tissue and yet fail to alter the cycle length of reentry. A 2-to-1 conduction block that occurred outside the core also did not alter the reentrant cycle length. (2) The spiral morphology is not helpful in differentiating the mechanisms of reentry. Fig 10⇑ illustrates how an anatomic reentry could exhibit itself as a spiral-shaped wave. Ikeda et al24 recently mapped reentrant excitation in isolated atrial tissue with a hole in the center. The morphology of the wave front had an apparent curvature, which was not different from the morphology of functional reentry without holes in the same atrial tissue preparation.9 (3) The apparent curvature of the spiral wave front is determined by the velocity of the outwardly propagating wave front. If the propagation in the outward direction is slow, the wave front in Fig 10F⇑ will look more curved than if the outward propagation is fast. Therefore, the presence of apparent spiral shape in some models of functional reentry but not in others can be explained by the differential velocity of outward propagation (radial component) relative to the velocity at which the activation front is propagating around the perimeter (tangential component) of the core. If the velocity of outward propagation is fast, spiral waves may not be demonstrable unless a larger area is mapped. (4) To terminate reentry, the timing of the stimulus is important. The stimulus has to interact with the reentrant circuit during the excitable gap to create antidromic and orthodromic wave fronts to terminate reentry. This hypothesis also predicts that, if the stimulus is given during the phase 3 repolarization of the reentrant circuit, it may result in graded responses6 25 that prolong the action potential duration. This mechanism can terminate reentry without initiating a propagated wave front by prolonging refractoriness near the core. In the present study, both phenomena have been observed.
A major limitation of the study is that only two-dimensional patterns of activation were mapped, whereas the actual heart is three-dimensional. The electrical structure of the functional reentrant wave fronts in three dimensions for the intact heart cannot be predicted by the data reported in this manuscript. It remains unclear whether or not a twice-threshold pacing stimulus can terminate functional reentry in the in vivo heart. A second limitation is that transmembrane potentials were not simultaneously recorded during the study. It is therefore unclear whether our hypothesis that graded responses terminated reentry in some of the episodes by prolonging action potential duration is valid.
We conclude that both the timing and the location of a pacing stimulus are important in terminating functional reentrant wave fronts in isolated swine right ventricular tissues. The maintenance of reentrant excitation depends on a small area around the core (the “rotor”). The remaining portion of the tissue is activated passively by the outward propagation of wave fronts away from the core, similar to that which occurs in the excitable chemical media.4
This work was done during the tenure of a Cedars-Sinai Burns and Allen Research Institute Research Fellowship (Dr Ikeda) and an AHA/Wyeth-Ayerst Established Investigatorship Award (Dr Chen) and was supported in part by a Specialized Center of Research (SCOR) grant in sudden death (P50-HL-52319) and a FIRST award (R29-HL-50259) from the National Institutes of Health, the Electrocardiographic Heartbeat Organization, and the Ralph M. Parsons Foundation. The authors wish to thank Peter Hunter, PhD, David Bullivant, PhD, Sylvain Martel, and Serge LaFontaine for constructing the mapping system and Chun Hwang, MD, for making the tissue chamber used in this study. We also thank Avile McCullen, Meiling Yuan, and Dustan Hough for technical assistance and Elaine Lebowitz for secretarial assistance.
- Received February 4, 1997.
- Revision received April 2, 1997.
- Accepted April 4, 1997.
- Copyright © 1997 by American Heart Association
Allessie MA, Bonke FIM, Schopman FJG. Circus movement in rabbit atrial muscle as a mechanism of tachycardia. Circ Res. 1973;33:54-62.
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.
Chen P-S, Wolf P, Dixon EG, Danieley ND, Frazier DW, Smith WM, Ideker RE. Mechanism of ventricular vulnerability to single premature stimuli in open chest dogs. Circ Res. 1988;62:1191-1209.
Winfree AT. Spiral waves of chemical activity. Science. 1972;175:634-636.
Winfree AT. Stably rotating patterns of reaction and diffusion. Prog Theor Chem. 1978;4:1-51.
Gotoh M, Uchida T, Mandel WJ, Fishbein MC, Chen P-S, Karagueuzian HS. Cellular graded responses and ventricular vulnerability to reentry by a premature stimulus in isolated canine ventricle. Circulation. 1997;95:2141-2154.
Uchida T, Ikeda T, Kamjoo K, Hough D, Mandel WJ, Weiss JN, Chen P, Karagueuzian HS. Mechanism of acceleration of functional reentry in the ventricle. Pacing Clin Electrophysiol. 1996;19:II-665. Abstract.
Davidenko JM, Salomonsz R, Pertsov AM, Baxter WT, Jalife J. Effects of pacing on stationary reentrant activity: theoretical and experimental study. Circ Res. 1995;77:1166-1179.
Ikeda T, Uchida T, Hough D, Lee JJ, Fishbein MC, Mandel WJ, Chen P-S, Karagueuzian HS. Mechanism of spontaneous termination of functional reentry in isolated canine right atrium: evidence for the presence of an excitable but nonexcited core. Circulation. 1996;94:1962-1973.
Bonometti C, Hwang C, Hough D, Lee JJ, Fishbein MC, Karagueuzian HS, Chen P-S. Interaction between strong electrical stimulation and reentrant wavefronts in canine ventricular fibrillation. Circ Res. 1995;77:407-416.
Lee JJ, Kamjoo K, Hough D, Hwang C, Fan W, Fishbein MC, Bonometti C, Ikeda T, Karagueuzian HS, Chen P-S. Reentrant wave fronts in Wiggers’ stage II ventricular fibrillation: characteristics, and mechanisms of termination and spontaneous regeneration. Circ Res. 1996;78:660-675.
Friedman P. GB-Stat. Silver Spring, Md: Dynamic Microsystems Inc; 1995.
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.
Spach MS, Miller WT, Geselowitz DB, Barr RC, Kootsey JM, Johnson EA. The discontinuous nature of propagation in normal canine cardiac muscle: evidence for recurrent discontinuities of intracellular resistance that affect the membrane currents. Circ Res. 1981;48:39-54.
Gotoh M, Uchida T, Fan W, Fishbein MC, Karagueuzian HS, Chen P-S. Anisotropic repolarization in ventricular tissue. Am J Physiol. 1997;272:H107-H113.
Fan W, Gotoh M, McCullen A, Karagueuzian HS, Chen P-S. Reappraisal of effective refractory period testing. Am J Physiol. 1994;267:H406-H410.
Verrier RL, Brooks WW, Lown B. Protective zone and the determination of vulnerability to ventricular fibrillation. Am J Physiol. 1978;234:H592-H596.
Gray RA, Jalife J, Panfilov AV, Baxter WT, Cabo C, Davidenko JM, Pertsov AM, Hogeweg P. Mechanisms of cardiac fibrillation. Science. 1995;270:1222-1223.
Allessie MA, Bonke FI, Schopman FJ. 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.
Cha Y-M, Birgersdotter-Green U, Wolf PL, Peters BB, Chen P-S. The mechanisms of termination of reentrant activity in ventricular fibrillation. Circ Res. 1994;74:495-506.
Davidenko JM, Pertsov AM, Salomonsz R, Baxter W, Jalife J. Stationary and drifting spiral waves of excitation in isolated cardiac tissue. Nature. 1991;355:349-351.
Allessie MA, Schalij MJ, Kirchhof CJHJ, Boersma L, Huybers M, Hollen J. Experimental electrophysiology and arrhythmogenicity, anisotropy and ventricular tachycardia. Eur Heart J. 1989;10(suppl E):2-8.
Ikeda T, Uchida T, Mandel WJ, Chen P, Karagueuzian HS. Conversion of non-stationary to stationary reentrant wave front with a critically-sized anatomical obstacle in the atrium. J Am Coll Cardiol. 1996;27:60A. Abstract.
Kao CY, Hoffman BF. Graded and decremental response in heart muscle fibers. Am J Physiol. 1958;194:187-196.