The Mechanism of Pause-Induced Torsade de Pointes in Long QT Syndrome

Jinqiu Liu, M.D.; Kenneth R. Laurita, Ph.D.

Disclosures

J Cardiovasc Electrophysiol. 2005;16(9):981-987. 

In This Article

Results

Shown in Figure 1 are representative tracings of the ECG, the shortest (Epi) and longest (M-cell) transmural action potentials, and DOR under control conditions (left), LQT2 (middle), and LQT3 (right). QT interval was markedly prolonged from 356 msec to 453 msec and 549 msec, respectively in LQT2 and LQT3 models at a steady state baseline pacing cycle length of 1,000 msec. Correspondingly, APD was markedly increased in both models compared to control. Under LQT2 and LQT3 conditions, M-cells had greater APD prolongation than Epi. As a result, transmural dispersion was increased from 18 msec during control to 58 msec in LQT2 and 115 msec in LQT3. Similar results were observed over all experiments (panel B). Compared to control, LQT3 conditions increased QT interval (103%), mean APD (109%), and DOR (404%) to a greater extent than LQT2 conditions (24%, 22%, and 45%, respectively).

Panel A shows the ECG and the shortest (Epi) and longest (M-cell) action potentials under control, LQT2, and LQT3 conditions during steady state pacing. All numbers within action potential are APD in msec. QT interval, APD, and transmural dispersion of repolarization (DOR) increased during LQT2 and LQT3 compared to control. Similar results were observed over all experiments. Compared to control (n = 6), QT interval, mean APD and transmural DOR under conditions of LQT2 (n = 6) and LQT3 (n = 11) significantly increased. LQT3 had a greater effect on QT interval, mean APD, and DOR compared to LQT2. (*P < 0.01).

Figure 2 shows the ECG and transmural action potentials recorded during a short-long cycle length induced pause of 1,700 msec under control (top), LQT2 (middle), and LQT3 (bottom) conditions. After the pause (S3), transmural APD and DOR increased compared to baseline (S1) under LQT2 and (to a greater extent) LQT3 conditions, but not under control conditions. Isochronal maps of APD during a short-long cycle length sequence (Fig. 3) show the effect of a pause on spatial dispersion of repolarization under the three conditions. APD was always shortest near the epicardium. In control conditions (top), the pause had little effect on transmural APD and repolarization gradients. In this representative example, the maximum local repolarization gradient following the pause under control conditions was 3.60 msec/mm. In contrast, during LQT2 (middle) the pause-enhanced transmural APD for M-cells. As a result, transmural repolarization gradients increased. The pause had the greatest effect during LQT3 conditions (bottom), transmural APD and repolarization gradients increased dramatically, compared to LQT2 as evidenced by the intense red color and crowding of isochrone lines. Under LQT3 conditions, the maximum local repolarization gradient increased from 18.85 msec/mm during baseline pacing to 43.92 msec/mm following the pause (S3). Similar results were observed over all experiments at each pause (S2–S3) interval tested ( Table 1 ).

ECG and the shortest (Epi) and longest (M-cell) action potentials under control (top), LQT2 (middle), and LQT3 (bottom) conditions during a short-long cycle length sequence. S1 indicates steady state baseline stimulation (1,000 msec); S2, premature stimulation (short); S3, postmature stimulation (1,700 msec, long). All numbers within action potential are APD in msec. The pause enhanced APD prolongation and transmural DOR during LQT2 and LQT3, but not during control conditions. Compared to LQT2, the pause produced a greater effect in LQT3.

ECG and isochronal maps of transmural APD during baseline pacing and after a pause under control (top), LQT2 (middle), and LQT3 (bottom) conditions. The red boxes on the ECG indicate the beat corresponding to isochrone maps shown below. During control, the pause had very little effect on APD and repolarization gradients (i.e., crowding of isochrone lines). However, during LQT2 and LQT3, the pause increased APD and repolarization gradients dramatically compared to baseline pacing (S1), especially in LQT3.

During control conditions, no EADs, EAD-induced triggered activity, or spontaneous TdP was observed. Under LQT2 conditions, EADs and EAD-induced triggered activity were observed in only 1 of 6 preparations. In contrast, during LQT3, EADs and EAD-induced triggered activity were observed in 9 of 11 preparations (P < 0.02). Figure 4 shows an example of a pause-induced EAD under LQT3 conditions. The EAD occurs during the action potential plateau phase of the S3 beat and is absent during baseline pacing (S1). Over all experiments (panel B), the number of recording sites (as a percentage of the total array) where EAD activity was observed was significantly greater (P < 0.03) after a pause of 2,000 msec (43%± 5%) compared to baseline pacing (32%± 2%). Similar results were observed with a pause of 900 msec during baseline pacing at 600 msec.

The pause associated with a short-long cycle length sequence also facilitated the induction of EADs in LQT3. Shown are the ECG and action potential recording during a short-long cycle length sequence (panel A). The arrow indicates an EAD. EADs were observed following a pause (2,000 msec) but not during baseline pacing (1,000 msec). Panel B shows summary data. After a pause of 2,000 msec, significantly more sites in the mapping field (as a percent of the entire mapping field) showed EADs compared to baseline pacing at 1,000 msec. (*P < 0.03)

Figure 5 illustrates an EAD-induced triggered beat following a pause of 1,700 msec under LQT3 conditions (panel A). The contour map on the left (panel B) shows transmural APD of the S3 beat. Action potential traces from the M-cell region (site a) and the epicardial region (site b) demonstrate the regional prolongation of APD following the pause. The maximum local repolarization gradient (43.92 msec/mm) occurred near the bottom of the mapping field (diamond). The contour map on the right shows activation of the triggered beat (V). Interestingly, the breakthrough site (asterisk) corresponds exactly to the location of the maximum repolarization gradient rather than where APD is longest. In addition, the relatively small area of the first contour (dark blue) indicates active propagation in the plane of the mapping field and, thus, an origin near the transmural surface. No evidence of conduction block was observed in the mapping field.

Pause-dependent triggered activity. Shown is an EAD-induced triggered beat following a pause of 1,700 msec at baseline pacing 1,000 msec. Action potential traces are from the M-cell region (site a) and the epicardial region (site b). The M-cell action potential demonstrates that the triggered beat occurred before full repolarization of S3. The contour map on the left (panel B) shows APD of an S3 beat from an identical recording in the same wedge preparation, but in the absence of an ectopic beat (V). Blue color represents shortest APD, red color represents longest APD. The maximum local repolarization gradient (43.92 msec/mm) occurred near the bottom of the mapping field (diamond). The contour map on the right shows activation of the triggered beat. The site of breakthrough (asterisk) corresponded to the location of the maximum local repolarization gradient. No evidence of conduction block was observed in the mapping field.

Shown in Figure 6 is spontaneous pause-induced TdP that lasted for ~3 seconds. The first contour map on the left shows APD after the pause. The maximum local repolarization gradient (42.96 msec/mm) occurred near the bottom of the mapping field (diamond). A premature beat (V2) occurred following V1 and, as observed in the previous example (Fig. 5), breakthrough (asterisk) occurred exactly at the site where the local repolarization gradient was largest. However, unlike the previous example, conduction block occurred in the direction of longest APD (M-cells) and the impulse propagated off the mapping field (arrow). After a pause, V3 and a second premature beat (V4) occurred. The activation map of V4 shows the same site of breakthrough and block as V2. Then, after a delay of approximately 200 msec a spontaneous beat reentered the mapping field (V5) and blocked in the midmyocardial region (thick black line). However, the impulse propagated toward the epicardium through a small isthmus, creating a pattern suggesting figure of eight type reentry. On subsequent beats (data not shown) the region of conduction block persisted, but its exact location varied slightly from beat to beat.

Shown is spontaneous pause-dependent TdP induced by a short-long cycle length sequence. The ECG shows a spontaneous beat (V1) followed closely by a spontaneous premature beat (V2) that did not induce TdP. Following the pause, a second spontaneous (V3) and premature (V4) beat initiate TdP. The first contour map on the top (left) shows APD after a pause (1,700 msec) from a similar recording in the same wedge preparation corresponding to V3, but without an ectopic beat. The maximum repolarization gradient (diamond) occurred near the bottom of the mapping field. The contour map on top (right) shows activation of V2. The breakthrough site (asterisk) occurred exactly at the site where the local repolarization gradient was largest, and conduction block (thick black line) occurred in the direction of longest APD. The contour maps on the bottom show breakthrough activation of V4 and the first beat of TdP entering from outside the mapping field (V5). See text for details.

The incidence of EAD-induced triggered activity and TdP following a pause was greater compared to no pause. Over all experiments, 31 pause-dependent episodes of EAD-induced triggered activity and 12 episodes of TdP were documented. In contrast, only 16 episodes of nonpause-dependent triggered activity and 7 episodes of TdP occurred (Fig. 7). Statistical significance was reached between pause and nonpause-dependent EAD-induced triggered activity (P < 0.02) but not for TdP. These data suggest that the pause associated with a short-long cycle length sequence increases the likelihood of EAD-induced triggered activity and, possibly, TdP.

The incidence of EAD-induced triggered activity and TdP following a pause (pause dependent) compared to steady state rhythm (nonpause dependent). A pause significantly increased the number of EAD-induced triggered beats (*P < 0.02). A pause increased the number of TdP episodes, but this did not reach statistical significance.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....