Termination of Ventricular Tachycardia in the Human Heart
Insights From Three-dimensional Mapping of Nonsustained and Sustained Ventricular Tachycardias
Background To define the electrophysiological basis for the termination of ventricular tachycardia (VT), three-dimensional cardiac mapping and analysis of the terminal beats of nonsustained VT and beats of sustained VT were performed in six patients with healed myocardial infarcts.
Methods and Results Termination of VT was due to activation from multiple initiation sites that were discordant from those responsible for the maintenance of sustained VT in 45% of cases, to repetitive activation from single sites that were discordant from those responsible for the maintenance of sustained VT in 24% of cases, or to activation from sites concordant with the sites of repetitive activation during sustained VT in 31% of cases. Sustained VT was characterized by occasional shifting of initiation sites, even after the tachycardia entered the stable monomorphic phase. Mapping was of sufficient density to define the mechanisms for 21 terminating beats of VT. In 5 cases, termination was due to intramural reentry, which initiated with the total activation time of the preceding beat of 204±11 milliseconds (ms) but terminated primarily because of a decrease in total activation time (144±23 ms, P=.03) that was associated with the development of intramural conduction block or with significant changes in the activation sequence along the reentrant circuit. In 16 cases, terminal beats were initiated by a focal mechanism on the basis of the absence of intervening electrical activity from the termination of the preceding beat to the initiation of the terminating beat (172±9 ms). Focal activation was associated with less conduction delay of the preceding beat (115±6 ms) than terminating reentrant beats (P<.001) and usually terminated suddenly without oscillations in cycle length or total activation time.
Conclusions Termination of VT is associated with alterations in initiation sites that are most often discordant from those maintaining sustained VT and is due to either reentrant or focal mechanisms.
The electrophysiological basis for the termination of VT remains to be determined.1 Studies in experimental preparations of reentry have shown that nonsustained VTs differ from sustained VTs in that they demonstrate oscillations in both CI and TA time.2 3 Analysis of surface ECGs from patients with sustained and nonsustained VTs has also suggested that oscillations in CI, TA (as reflected by the QRS duration), and repolarization time (as reflected by the QT interval) may contribute in some way.4
Delineation of the electrophysiological mechanisms of VT termination requires a comparison of activation maps of nonsustained and sustained VT. We have performed intraoperative three-dimensional activation mapping in patients with healed myocardial infarcts and refractory VT undergoing subendocardial resection.5 6 We found that sustained monomorphic VT initiated by intramural reentry in one half of cases. In the other half, sustained monomorphic VT arose from a focal mechanism on the basis of the absence of intervening electrical activity from the termination of the preceding beat to the initiation of the next beat despite the presence of multiple intervening intramural recording sites.5 In our companion study of nonsustained VT induced by programmed stimulation in patients with healed myocardial infarcts,6 we found that compared with sustained VT, nonsustained VT initiated with comparable conduction delay of the terminal extrastimulus, initiated primarily at sites discordant from those initiating sustained VT, demonstrated greater oscillations of CI (but not TA) for the first 4 beats of VT, and initiated by reentrant or focal mechanisms.6 In the present study, we delineated the activation sequence of beats responsible for termination of nonsustained VT and how they compare with the activation of beats during sustained VT to determine (1) the basis for oscillations in CI and its contribution to the termination of nonsustained VT, (2) the concordance of sites of earliest activation for terminal beats of nonsustained VT with sites of earliest activation for maintenance beats during sustained VT, and (3) the electrophysiological mechanisms responsible for termination of VT.
We studied six patients with ischemic heart disease and medically refractory sustained monomorphic VT who underwent intraoperative three-dimensional mapping and arrhythmia surgery at Barnes Hospital and in whom nonsustained VT was induced with programmed stimulation. Sustained VT was induced in five of the six patients. The patient characteristics and three-dimensional intraoperative mapping procedures are given in a companion report.6 Transmural and transseptal ventricular mapping was performed with a computer-assisted three-dimensional cardiac mapping system capable of simultaneously recording from 160 transmural sites.5 6 7 As many as 34 plunge-needle electrodes (136 total recording sites), each containing four bipolar electrode pairs separated by 4 mm, were inserted throughout the LV and RV with an interelectrode distance ranging from 1.0 to 3.0 cm. Needle density was concentrated in regions surrounding the infarct zone. Surface ECG leads I, aVF, and V5R were recorded. Electrograms were analyzed with an automated system.5 6 7 8 9 10 11 Computer-assigned activation times were based on a peak criterion. Three-dimensional activation maps were constructed, as previously described,5 6 9 10 11 from data acquired during programmed stimulation and during induced nonsustained and sustained VTs.
Data and Statistical Analyses
The mechanism responsible for a particular premature beat was defined as macroreentrant when (1) there was continuous depolarization from the preceding beat, (2) the site of initiation of a premature beat was adjacent to the site of termination of the preceding beat, and (3) the conduction velocity of the activation wave front from the site of termination of the preceding beat to the site of initiation of the ectopic beat was similar to the conduction velocity of the terminal portion of the activation wave front of the preceding beat.5 6 11 A mechanism was defined as focal when the site of initiation of a premature beat demonstrated radial spread of activation and was remote from the site of termination of the preceding beat with no intervening depolarizations despite multiple (four or more) intermediate recording sites.5 6 11 A focal mechanism does not exclude the possibility of microreentry. CI, TA time, and oscillations in CI or TA for a beat of VT have been defined previously.6 Data are presented as mean±SEM unless otherwise stated. Student's t test was used to identify significant differences in CI or TA. For comparison of TA and CI data between runs of sustained and nonsustained VTs, a two-way ANOVA was used.
The initiation and termination of a total of 44 nonsustained VTs (≤39 beats long) and the initiation and/or maintenance of six sustained VTs were recorded from six patients. A total of 151 extrastimuli preceding the initiation of nonsustained VT, 289 beats of nonsustained VT, 16 extrastimuli preceding the initiation of sustained VT, and 130 beats of sustained VT were recorded and analyzed. Analysis of CI and TA was performed for all beats of nonsustained VT and for up to the first 28 beats of sustained VT. Detailed analysis of the site of earliest activation, the three-dimensional activation sequence, and the arrhythmia mechanism were performed for the last 3 to 5 beats of short (<10 beats) runs of nonsustained VT (a total of 131 beats), for the last 22 to 33 beats of the three nonsustained VTs >25 beats long (a total of 86 beats), and for the 11th to the 28th beats of sustained VT (a total of 72 beats); a total of 289 beats were mapped based on the analysis of >24 000 electrograms. Analysis of the activation sequence and mechanism of the preceding extrastimuli and the first 5 beats of nonsustained VT and the first 10 beats of sustained VT was performed in our accompanying study.6
Characteristics of Termination
The TA of the terminal beats of nonsustained VT (140±8 ms) was less than that of maintenance beats (15th to 18th beats) of sustained VT (165±7 ms, P=.02). The TA of the last beat of nonsustained VT exceeded that during sustained VT in only six cases. However, the CI of the last beat of nonsustained VT (298±12 ms) did not differ significantly from that of a comparable beat of sustained VT (276±11 ms, P=.20).
To explore the role of the varying initiation sites in the termination of VT, the site of earliest activation for beats from 30 nonsustained and 6 sustained VTs (from the five patients with both nonsustained and sustained VT) was determined. VTs were classified as to whether termination was due to activation from multiple (≥3) initiation sites discordant from sites responsible for maintenance of sustained VT (group I), repetitive (≥3) activation from a single site discordant from sites responsible for maintenance of sustained VT (group II), or activation from sites concordant with sites of repetitive activation during sustained VT (group III).
In 45% of cases (4 monomorphic and 10 polymorphic VTs), termination of VT was associated with multiple subendocardial or subepicardial initiation sites that were discordant from those responsible for sustained VT. As shown in Fig 1⇓, nonsustained VTs from patient 1 initiated at any of a number of primarily subendocardial sites A through H, all of which were discordant from the site of earliest activation during the maintenance of sustained VT (asterisk). Initiation sites could fire repetitively (eg, repetitive firing from site A in the VT in Fig 1c⇓) or singly. Sequential activation of multiple disparate sites was often associated with oscillations of CI (eg, Fig 1b, 1c, 1e, and 1f⇓), although at times minimal or no oscillation of CI was evident despite changing initiation sites (eg, Fig 1g⇓, beats T2-T4).
Termination of VT was associated with repetitive firing primarily from sites disparate from those responsible for sustained VT in 24% of cases (5 monomorphic and 2 polymorphic VTs) and was most evident in the three long runs of nonsustained monomorphic VT. Three patients demonstrated monomorphic VTs that terminated after 28, 37, and 39 beats, respectively. In each case, the tachycardias initiated at one site but switched to one or several different sites, which then fired repeatedly until the VTs terminated. In all three cases, the sites of repetitive activation were all discordant from the sites of initiation during the maintenance of sustained VT in these same patients with the exception of 2 isolated beats of VT in patient 2.
As shown in Fig 2⇓ (top), the tachycardia from patient 2 initiated at subendocardial site A for T1 and switched to subendocardial site B for T2-T39, except for initiation at subendocardial site C for beat T8 and at subendocardial site D for beats T13 and T37. Sites A, B, and C were all distant from the sites of repetitive firing during the maintenance of the two sustained VTs (asterisk and site D). Oscillations in CI and TA were evident for beats T1-T3 due to the change in sites of initiation. The tachycardia suddenly terminated after beat T39, 2 beats after a change in initiation site for beat T37 to site D, which was concordant with sites responsible for maintenance of one of the two sustained VTs in this patient. There was no change in either CI or TA. VT beats that initiated at site D demonstrated activation sequences similar to beats of sustained VT initiating at that same site.
The VT in patient 4 (Fig 2⇑, middle) initiated at subendocardial site A for beat T1 and then switched to subendocardial site B for T2, subepicardial site C for T3, and midmyocardial site D for beats T4-T28. All of these sites were distant from that of repetitive initiation during the maintenance of the sustained VT (asterisk). The VT terminated after 28 beats without oscillation in CI or TA for the last 6 beats. Oscillations in CI and TA were evident for beats T1-T4 and were most likely due to changing sites of initiation. However, for subsequent beats T5-T8, T16, T19, and T20, all of which initiated at the same site (D), oscillations (increases or decreases >20 ms) in CI (shown in black) were usually associated with changes in TA of the preceding beat, which occurred in the same direction.
The third VT (Fig 2⇑, bottom) from patient 5 initiated at site A in the subendocardium of the interventricular septum for the first beat T1 and then at midmyocardial site B in the posterior LV for the subsequent 36 beats. Termination after 37 beats was preceded by an increased CI in the terminal beat from 219 ms for beat T36 to 257 ms for beat T37 (Fig 3⇓). In this case, the increase in CI for the beat T37 was preceded by an increase in TA from 120 ms (for T34) to 163 ms for T35 (due to late activation at site D in the posterobasal epicardium, Fig 3⇓) and may have contributed to the alteration in CI 1 beat later. There also was an increase in the TA for beat T37 of the tachycardia (from 111 to 132 ms), after which the VT terminated.
Thus, prolonged runs of nonsustained VT involved repetitive firing primarily from sites disparate from those responsible for repetitive firing during sustained VT; in one of the three cases, termination was associated with a sudden increase in CI at the site of repetitive activation, possibly because of changes in conduction of preceding beats.
Repetitive firing from sites concordant with those underlying sustained VT occurred in 31% of cases (6 monomorphic and 3 polymorphic VTs). Termination of VT was due to one of two causes. For some, such as that shown in Fig 4⇓, the first beat of the nonsustained VT initiated at the same sites as the first beat of sustained VT. The sustained VT (Fig 4⇓, right) initiated at midmyocardial site A in the lateral LV for the first beat and then switched to the adjacent subendocardial site A′ (T2-T3) and then to endocardial sites B (T4) and C (T5 and beyond). The nonsustained VT (Fig 4⇓, left) initiated at midmyocardial site A for the first beat and then subendocardial site A′; for the next 3 beats (T2-T4), however, it never switched to site B or C before terminating. Thus, although the nonsustained VT was initiated and maintained at sites concordant with the initiation of sustained VT, the tachycardia terminated because it failed to “switch” to the concordant site that was responsible for maintenance of sustained VT.
In other cases, multiple runs of nonsustained VT were initiated and maintained at the same sites as those that maintained sustained VT. As shown in Fig 5⇓, repetitive activation from a concordant posterobasal subepicardial site either terminated suddenly after 4 beats with no oscillation of CI or TA (Fig 5A⇓ and 5C⇓) or was followed by late coupled ectopic beats arising from distant and discordant sites with marked CI oscillation similar to that noted in group I (Fig 5B⇓ and 5D⇓).
Variation in Initiation Sites During Sustained VT
We previously demonstrated variation in sites of earliest activation during the initiation of sustained VT (up to the first 10 beats).6 In the present study, we analyzed three-dimensional activation patterns of the subsequent 10 to 18 beats of sustained VT. Surprisingly, even during the stable phase of sustained VT when monomorphic VT was evident in all three monitored surface leads (I, aVL, V5R), there could be considerable variation in the sites of earliest activation of beats during the maintenance of sustained VT. As shown in Fig 6⇓ (and Fig 5E⇑), after initiation at midmyocardial site A (T1), beats T2-T10 of the sustained VT from patient 2 involved repetitive activation from subepicardial site B in the posterobasal LV. There was an increase in TA for beat T7 (227 versus 189 ms for beat T6). Two beats later, the CI for T10 increased to 295 ms (from 242 ms for beat T9). Beat T11 initiated with an even greater CI (349 ms) but did so from subendocardial site C in the midposterior LV. After activation of T11, site C failed to activate; instead of terminating, however, the tachycardia continued with initiation of beat T12 and the subsequent ≥10 beats occurring back at site B. The CI for beat T12 (309 ms) was less than that for T11, and those for beats T13-T14 were even less (251 and 285 ms, respectively). The CIs of subsequent beats T14-T24 were very stable (296±1 ms), ranging from 285 to 299 ms, slightly greater than those for beats T2-T9 of sustained VT (231±4 ms) or for the beats of nonsustained VT that initiated at site B (Fig 5A to 5D⇑; 259±4 ms). Of note, site C, the initiation site that appeared to “save” the tachycardia from terminating, was also the site of repetitive initiation for a second run of sustained VT in patient 4. Thus, alterations in the initiation site contribute to the development of sustained VT, and there appears to be a dynamic interaction between sites that are capable of initiating activation even during sustained monomorphic VT.
Even after a sustained VT enters a stable monomorphic phase, the initiation site could vary considerably from beat to beat (Fig 7⇓). In the VT illustrated, the initiation site changed from subendocardial site A to B to C to B to C and then back to B over the course of the first 21 beats. These changes in initiation sites at T10, T14, T18, T20, and T21 were associated with changes (>20 ms) in CI for beats T10, T14-T16, T18, T19, and T21 and in TA for beats T14, T18, and T21 but no changes in the three surface ECG leads monitored. Thus, oscillations in CI, and at times TA, even during sustained VT, can result from changes in initiation site despite minimal if any change in QRS morphology.
Basis for Oscillation of CIs
Oscillation in CI occurred with 93 beats from 44 nonsustained VTs. In 53 cases (57%), CI oscillation preceding termination of VT was associated with changes in the initiation site from beat to beat, as shown in Fig 1⇑.
In 40 cases (43%), oscillations in CI occurred despite repetitive firing from the same site (eg, Fig 1d⇑). The basis for this is unknown, but it was characteristic of several nonsustained VTs for up to the first 3 beats after a shift in the initiation sites in 33 of 40 cases. Thus, in only 7 cases did oscillation of CI occur more than 3 beats (range, 4 to 35) after a shift in initiation site.
Only four VTs terminated with no evidence of oscillations in CI (or TA). This finding was due to changes in initiation sites, which occurred with comparable CIs in two cases, and to repetitive initiation sites in two cases. As shown in Fig 1g⇑, termination of nonsustained VT in patient 1 occurs without CI oscillation despite the fact that the site of initiation for the last 3 beats shifted from site H to C to B before termination.
Oscillation in CI occurred with 21 beats from six sustained VTs. In 10 cases (48%), it was related to changes in initiation site; in the other 11 cases (52%), it was related to repetitive firing from the same site for up to 2 beats after a shift in initiation site.
Electrophysiological mechanisms could be defined for 21 terminating beats (last and/or penultimate beats) of 13 VTs from four patients. In five cases, termination involved intramural reentry that failed to propagate further. These reentrant beats were associated with a TA of the preceding beat of 204±11 ms. However, these reentrant beats conducted with less delay (144±23 ms, P=.035), which may have contributed to termination of reentry and ultimately termination of VT.
In some cases, intramural reentry was evident during sustained VT, but nonsustained VT terminated without the development of reentry because of insufficient conduction delay. As shown in Fig 8C⇓, sustained VT from patient 1 demonstrated a decrease in TA from 252 ms for beat T1 to 120 and 119 ms for beats T2 and T3, respectively. Yet, there was progressive conduction delay during beats T4 and T5 (169 and 213 ms, respectively) that contributed to the development of intramural reentry during sustained VT that was evident in beats T7 and beyond (see Fig 3 in Reference 55 ). Nonsustained VT in Fig 8A⇓ initiated at the same site as that initiating sustained VT and demonstrated a similar decrease in TA from 196 ms for T1 to 123 ms for T2. However, in contrast to the sustained VT, TA never rose beyond 159 ms (for T3), and the tachycardia terminated after 5 beats (TA of T5 was 126 ms) without the development of intramural reentry.
In other cases, intramural reentry failed to develop despite a considerable degree of conduction delay. As shown in Fig 8B⇑, the 3-beat nonsustained VT from patient 1 demonstrated considerable conduction delay for the second (T2) and third VT beats (T3) (198 to 212 ms, respectively). However, as shown in Fig 9⇓, the site of marked delayed conduction for T3 was located in the apex (site O), distant from the site of marked conduction delay during sustained VT (site F) in the basal lateral wall and distant from the site of reentrant initiation during sustained VT in the most basal lateral wall (site A). In fact, during this last beat of nonsustained VT, there is no significant conduction delay at site F, which activates 16 ms after the initiation of T3 at site G. Thus, the presence of marked conduction delay appears to be a necessary but not a sufficient condition for the development of reentry, and the site of delayed activation appears to be an important factor.
In some cases, reentry terminated because of the development of conduction block at critical sites in the reentrant circuit. As shown in Fig 10⇓, beat T2 initiated at site A and conducted over the pathway A-B-C-D-E-F-G and then back to A to initiate beat T3 by intramural reentry. After activation at A, beat T3 conducted to sites B-C-D-E but failed to conduct to sites F or G because of the development of conduction block at these sites.
In other cases, block developing within a reentrant pathway was followed by termination despite marked conduction delay because of alteration in the activation sequence along the reentrant circuit. As shown in Fig 10⇑, the terminal beat of the tachycardia, T4, initiated at epicardial site A in the posterobasal LV but by an indeterminate mechanism and propagated to sites B-C-D-E. However, unlike beat T3, activation during T4 spread from site E to midmyocardial site G and then from G to subendocardial site F. Delayed activation at F (234 ms) was greater than that at site G during beat T2, but because site F was not adjacent to site A, the activation wave front could not propagate further, and the tachycardia terminated. Again, conduction delay appears to be a necessary but not a sufficient condition for the development of intramural reentry, and the location of the site of delayed activation can be critical to the maintenance of reentrant excitation. The TA of beats T2-T4 (211, 125, and 234 ms, respectively) were not always reflected by the QRS duration of these beats (132, 119, and 148 ms, respectively).
In 16 cases, the penultimate or terminal beats initiated by a focal mechanism with no evidence of reentry. The TA of beats preceding focal activation (115±6 ms) was less than those for beats preceding penultimate or terminal beats initiating by intramural reentry (P<.001). The TA of the focal beats (116±7 ms) was comparable to that of terminal beats initiating by reentry (P=.19). An example of termination by a focal mechanism is shown in Fig 11⇓. Beats T2-T4 all initiate at a subendocardial site in level II by focal mechanisms and conduct with terminal activation in the posterobasal zone of the LV with a TA of 128 to 130 ms. As shown in Fig 12⇓, after focal activation from site A, conduction during T3 and T4 is rapid to adjacent sites B to J. After termination of T3 at a posterior epicardial site in level II (TA of 130 ms), there is no intervening electrical activity at any intermediate electrode sites, including sites B to J for 129 ms, until T4 initiates at site A by a focal mechanism. After initiation of T4 at site A, there was no further focal activation, and the tachycardia suddenly terminated with no oscillation in CI or TA.
Termination Involving Both Reentrant and Focal Mechanisms
In three cases, termination involved both reentrant and nonreentrant mechanisms. In each case, reentrant activity for one or two cycles was followed by a focal mechanism for ≤3 beats, after which the tachycardia terminated. The reentrant beats were associated with a TA of the preceding beat of 212±10 ms. However, the TAs of the reentrant beats were less (132±25 ms, P=.04) and were not sufficient to maintain reentry, and focal activation from distant subendocardium sites arose for 1 to 3 beats before the VTs terminated. The focal beats arose with a CI of 300±30 ms and conducted with a TA of 118±13 ms.
Fig 13⇓ shows the termination of the 7-beat nonsustained VT from patient 6 (the initiation of which is shown in Fig 11 of Reference 66 ). Beats T5 and T6 initiate at anterolateral sites in levels III and II, respectively, by intramural reentry (arrows). For the initiation of beat T4, the conduction velocity of terminal activation from level IV to level III (35 cm/s) was comparable to the conduction velocity from the termination site of T5 to the initiation of T6 (33 cm/s). The extent of marked conduction delay in level III for beat T5 (197 ms) is greater than that during beat T4 (167 ms) and thus may account in part for the increase in CI of beat T6 (273 ms versus 202 ms for beat T5). Beat T6 propagates apically but exhibits less conduction delay (109 ms) than T4 or T5 and terminates at an apical posterior site in level IV (120-ms isochrone). Beat T7 initiates at the same subendocardial site in level II as beat T6 and with a comparable CI (281 ms) but by a focal mechanism. There is no intervening electrical activity between the site of termination of T6 and the initiation of T7 (172 ms) despite the presence of multiple intervening electrical recording sites. Beat T7 conducts rapidly, with a TA of 87 ms. The focal site of initiation, and indeed all sites, failed to activate and initiate any further ectopic beats; the tachycardia therefore terminated.
The results of this study demonstrate that termination of VT involves a number of factors. First, terminal beats of VT tend to arise from sites that are discordant from sites of earliest activation during sustained VT. Ectopic beats can arise from any of multiple sites throughout the heart in response to programmed stimulation, but there are certain sites that are responsible for repetitive activation during sustained VT that are infrequently activated during the termination (as well as initiation and maintenance6 ) of nonsustained VT. Second, termination often involved shifting of initiation sites (with up to four sites in a run of nonsustained VT), which contributes to the oscillation in CI that is characteristic of many nonsustained VTs.6 Although we observed that CI oscillation was often associated with alteration in the site of initiation, in several cases CI oscillation continued for the subsequent 1 to 3 beats after the shift despite repetitive firing from the new site. Whether the preceding CI oscillations lead to heterogeneities of recovery of excitability, especially at the site of initiation, that continue to be manifest for several subsequent ectopic beats remains to be clarified. This phenomenon was observed during sustained as well as nonsustained VT, suggesting that it may reflect an intrinsic property of the myocardium that could be amenable to pharmacological modulation. In fact, a shift in the initiation site was associated with CI oscillations either immediately or within the subsequent 1 to 3 beats in 92% of the cases. In contrast, oscillations of CI were associated with oscillations in TA in a minority of these cases, providing a potential explanation for the lack of statistical difference in oscillations in TA for nonsustained versus sustained VT that we noted.6
At times CI oscillation was not associated with changes in initiation site but rather with oscillations in TA of the beat either immediately preceding or 2 beats before the CI oscillation. In these cases, conduction delay could occur at sites in proximity to the initiation site and potentially modulate the initiation site directly (as in Fig 3⇑, in which delayed activation at site C during beat T35 led to initiation of T36 by intramural reentry). However, in many cases, the site of delay responsible for oscillations in TA that preceded those of CI, and at times termination of the tachycardia, were at sites distant from the initiation sites, suggesting indirect effects such as alteration of the degree of heterogeneity of repolarization.
Duff et al4 reported that termination of VT is associated with oscillations of both CI and TA as assessed by measurements obtained from the surface ECG. However, we found in humans6 that the TA is not accurately reflected in measurements of QRS duration, in accord with results of three-dimensional activation mapping in the feline heart,9 most likely because the areas of marked conduction delay often occur in small localized regions of the heart that are not represented on the surface ECG.
An interesting finding in the present study was that sustained VTs exhibit alteration in initiation sites even beyond the 10th beat of the tachycardia. This was usually not associated with changes in the QRS morphology but could have been related to oscillations in CI (and at times with oscillation in TA). Another interesting finding was that repetitive activation during sustained VT could be associated with CI oscillation, a failure of repetitive activation, and activation from a different initiation site, which could maintain the tachycardia sufficiently so that activation from the original site of repetitive firing could again take over and the sustained VT continues to propagate. Exactly how this dynamic interaction of activation from different sites leads to continuation of sustained VT in some cases (Figs 6⇑ and 7⇑) and to termination of VT in other cases (eg, Fig 1⇑) remains unknown. However, these findings imply a dynamic interaction between different sites of initiation and their response to changes in CI and TA.
Some VTs, especially those >25 beats long, terminated with no evidence of oscillations of either CI or TA. In these cases, the sites of repetitive initiation were discordant from those responsible for maintaining sustained VT, suggesting that characteristics of certain initiation sites may impart lesser or greater stability and the potential for repetitive activation without sudden termination. Identification of these critical sites could lead to more effective techniques for ablation of sites of origin of VT in patients with healed infarcts.
Termination by Intramural Reentry
Termination of VT could occur by intramural reentry, which we have shown to underlie initiation and maintenance of both nonsustained and sustained VTs.5 6 Reentry during termination of nonsustained VT was associated with marked conduction delay (usually >200 ms) of the preceding beat. However, once initiated, the terminating reentrant beats propagated with too little conduction delay, so continued reentrant excitation could not be maintained. Furthermore, even when reentrant pathways were established during nonsustained VT at sites discordant from those responsible for maintaining sustained VT, they seldom lasted more than 1 to 2 beats because of a decrease in the extent of conduction delay. These findings suggest that only certain regions may have a predilection for the development of stable degrees of conduction delay and a stable reentrant circuit.
Our finding that conduction block or alterations in exit sites may contribute to alterations in reentrant excitation and at times lead to termination of VT is consistent with the results of Downar et al12 in endocardial mapping studies in humans. However, we found that intramural reentry could arise in the subepicardium or midmyocardium as well,5 6 suggesting that ablative techniques limited to the endocardium alone may be unsuccessful. The presence of both anatomic and functional conduction block in midmyocardial regions, which would not be detected by mapping of the endocardial and/or epicardial surfaces, appears to play a critical role and suggests the need for interrogation of midmyocardial regions, whether by three-dimensional mapping techniques5 6 or other approaches.13
In some cases, termination of VT occurred despite a conduction delay of the terminal beat that was greater than that noted during sustained VT. In these cases (eg, Fig 9⇑), the site of conduction delay was found to be different from that during sustained VT and to be distant from sites of conduction delay critical to intramural reentry during sustained VT. Thus, conduction delay is a necessary but not a sufficient condition for the development of intramural reentry, a finding we noted in our previous mapping studies in experimental animals9 10 and in the human heart.5 6 Delineation of the site of marked conduction delay manifest during nonsustained VT alone may not therefore identify sites that are critical to the development of reentry. It is likely that the predilection of certain sites for repetitive activation by reentry involves the combination of marked conduction delay at an adjacent site, anatomic and/or functional conduction block, and heterogeneity of repolarization.
Termination of VT can occur by a focal mechanism. With the resolution of the mapping in the present study, it is not possible to rule out microreentry. However, we believe this is very unlikely for several reasons. First, in all cases, the termination site of the preceding beat was distant from the site of initial activation, with multiple intermediate intramural electrodes failing to record any evidence of intervening electrical activity from the termination of the preceding beat to the initiation of the terminating focal beat. Second, activation around the sites of focal initiation was very rapid, with spread to adjacent intramural sites located on all sides of the focal site occurring within 30 to 50 ms. Third, microreentrant circuits as small as those demonstrated in infarcted canine myocardium with a high-density grid electrode array with an interelectrical distance of 350 μm14 were consistently associated with conduction delay of 100 ms between sites 1 to 2 cm apart. However, this was not the case for any of the focal beats of VT that were found to terminate VT in the present study or for any focal beats of nonsustained or sustained VT in our previous studies in patients with prior infarction.5 6 Termination of VT by a focal mechanism could be due to sequential focal activation from a number of different sites (often associated with CI oscillation) or to repetitive focal activation that terminates suddenly, without significant oscillation of CI or TA.
The nature of the focal mechanism for termination of VT remains unknown but may be triggered activity arising from delayed15 or early16 afterdepolarizations. Each of our patients had ischemic cardiomyopathy with at least moderate-to-severe depression of LV function. Myocardium from patients with ischemic cardiomyopathy has been shown to demonstrate delayed afterdepolarizations in vitro when exposed to catecholamines,17 suggesting one potential mechanism of focal activation. Furthermore, myocardium obtained at the time of heart transplantation from patients with dilated cardiomyopathy exhibits altered intracellular handling with increased diastolic levels of intracellular calcium18 that could activate a transient inward current19 (which may be Na+-Ca2+ exchange or a nonspecific cationic current20 ). Early afterdepolarizations are another potential mechanism for late coupled ectopic beats arising by a focal mechanism since myopathic human heart tissue has been shown to exhibit a prolonged action potential duration associated with decreases in both the transient outward current21 and delayed rectifier potassium currents.22 The finding that focal activation arises only from the subendocardium suggests that its source may be subendocardial myocytes and/or Purkinje fibers.
Focal initiation is associated with conduction delay of the preceding beat that is significantly less than that preceding reentrant initiation, suggesting that focal initiation could occur when conduction delay is inadequate to maintain intramural reentry. In several cases, termination involved both reentrant and focal mechanisms. In each case, it involved beats of VT initiating by intramural reentry that were followed by beats of VT arising by a focal mechanism when the TA of the reentrant beats was inadequate to maintain an intramural reentrant circuit. In fact, focal termination of VT was found to be more common than termination by reentry, perhaps because focal activation is not dependent on the development of a critical degree of conduction delay and can be activated by rapid cardiac activity.6 Our previous study in patients with healed myocardial infarcts demonstrated that a focal mechanism underlies the development and maintenance of sustained monomorphic VT induced by programmed stimulation in nearly half of cases.5
The main limitation of this study was the spatial resolution of the recording electrodes. Although a greater density of electrodes would have enabled delineation of mechanisms for a greater number of beats and would have better ruled out the occurrence of slowing that was not measured because of a limited number of electrodes, the resolution obtained in this present study was sufficient to (1) define the mechanisms for 21 terminating beats of VT, (2) define intramural activations that were critical to reentrant excitation, even within midmyocardial regions, and (3) successfully localize the sites of initiation of multiple morphologies of VT to allow surgical ablation and cryoablation with resultant cure of the VTs in all six patients. In previous mapping studies in the human as well as in the canine, feline, and rabbit heart,5 6 7 8 9 10 11 we consistently found that insertion of plunge needle electrodes does not lead to induction of VT by programmed stimulation when VT was not inducible before electrode placement. Furthermore, we found that mapping in patients with healed myocardial infarcts under normothermic bypass allows induction of clinical VT in nearly all cases.5 6
The termination of VT involves alterations in sites of initiation that are often associated with oscillation in CI.2 3 ,6 The dynamic interaction between sites of initiation, CI, and TA in the propagation and termination of VT suggests that pharmacological and ablative manipulation of these parameters could provide effective therapy for VT. Finally, termination and maintenance of sustained VT can result from reentrant and focal mechanisms; approaches to treatment of VT should be directed at both.
Selected Abbreviations and Acronyms
|LV||=||left ventricular; left ventricle|
|RV||=||right ventricular; right ventricle|
This work was supported in part by NIH grants HL-46929 (Dr Pogwizd) and HL-50295 (Dr Cain) and the Michael Bilitch Fellowship in Cardiac Pacing and Electrophysiology, North American Society of Pacing and Electrophysiology (Dr Chung). The authors gratefully acknowledge the expert technical assistance of H. Dieter Ambos and Jerome Peirick, surgical mapping by Drs James Cox and T. Bruce Ferguson, and secretarial assistance from Barbara Donnelly.
- Received November 20, 1995.
- Revision received November 25, 1996.
- Accepted December 16, 1996.
- Copyright © 1997 by American Heart Association
Frame LH, Simson MB. Oscillations of conduction, action potential duration, and refractoriness: a mechanism for spontaneous termination of reentrant tachycardias. Circulation. 1988;78:1277-1287.
Frame LH, Rhee EK. Spontaneous termination of reentry after one cycle or short nonsustained runs: role of oscillations and excess dispersion of refractoriness. Circ Res. 1991;68:493-502.
Duff HJ, Mitchell LB, Gillis AM, Sheldon RS, Chudleigh L, Cassidy P, Chiamvimonvat N, Wyse DG. Electrocardiographic correlates of spontaneous termination of ventricular tachycardia in patients with coronary artery disease. Circulation. 1993;88:1054-1062.
Pogwizd SM, Hoyt RH, Saffitz JE, Corr PB, Cox JL, Cain ML. Reentrant and focal mechanisms underlying ventricular tachycardia in the human heart. Circulation. 1992;86:1872-1887.
Chung MK, Pogwizd SM, Miller DP, Cain ME. Three-dimensional mapping of the initiation of nonsustained ventricular tachycardia in the human heart. Circulation. 1997;95:2517-2527.
Kramer JB, Saffitz JE, Witkowski FX, Corr PB. Intramural reentry as a mechanism of ventricular tachycardia during evolving canine myocardial infarction. Circ Res. 1985;56:736-754.
Pogwizd SM, Corr PB. Reentrant and nonreentrant mechanisms contribute to arrhythmogenesis during early myocardial ischemia: results using three-dimensional mapping. Circ Res. 1987;61:352-371.
Pogwizd SM, Corr PB. Electrophysiologic mechanisms underlying arrhythmias due to reperfusion of ischemic myocardium. Circulation. 1987;76:404-426.
Pogwizd SM. Focal mechanisms underlying ventricular tachycardia during prolonged ischemic cardiomyopathy. Circulation. 1994;90:1441-1458.
Downar E, Saito J, Doig JC, Chen TCK, Sevaptsidis E, Masse S, Kimber S, Mickleborough L, Harris L. Endocardial mapping of ventricular tachycardia in the intact human ventricle, III: evidence of multiuse reentry with spontaneous and induced block in portions of reentrant path complex. J Am Coll Cardiol. 1995;25:1591-1600.
Khoury DS, Taccardi B, Lux RL, Ershler PR, Rudy Y. Reconstruction of endocardial potentials and activation sequences from intracavitary probe measurements: localization of pacing sites and effects of myocardial structure. Circulation. 1995;91:845-863.
Zuanetti G, Hoyt RH, Corr PB. β-Adrenergic-mediated influences on microscopic conduction in epicardial regions overlying infarcted myocardium. Circ Res. 1990;67:284-302.
Wit AL, Rosen MR. Afterdepolarizations and triggered activity: distinction from automaticity as an arrhythmogenic mechanism. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System: Scientific Foundations. 2nd ed. New York, NY: Raven Press; 1993:2113-2163.
January CT, Riddle JM, Salata JJ. A model for early afterdepolarizations: induction with the Ca2+ channel agonist Bay K 8644. Circ Res. 1988;62:563-571.
Gwathmey JK, Copelas L, MacKinnon R, Schoen FJ, Feldman MD, Grossman W, Morgan JP. Abnormal intracellular calcium handling in myocardium from patients with end-stage heart failure. Circ Res. 1987;61:70-76.
Berlin JR, Cannell MB, Lederer WJ. Cellular origins of the transient inward current in cardiac myocytes: role of fluctuations and waves of elevated intracellular calcium. Circ Res. 1989;65:115-126.
Näbauer M, Beuckelmann DJ, Erdmann E. Characteristics of transient outward current in human ventricular myocytes from patients with terminal heart failure. Circ Res. 1993;73:386-394.
Beuckelmann DJ, Näbauer M, Erdmann E. Alterations of K+ currents in isolated human ventricular myocytes from patients with terminal heart failure. Circ Res. 1993;73:379-385.