Tachycardia Around a Fixed Obstacle in Anisotropic Myocardium

1307 Differential Effects of Heptanol, Potassium, and Tetrodotoxin on Reentrant Ventricular Tachycardia Around a Fixed Obstacle in Anisotropic Myocardium Josep Brugada, MD; Lluis Mont, MD; Lucas Boersma, MD; Charles Kirchhof, MD; and Maurits A. Allessie, MD Background. The aim of this study was to test the hypothesis that electrical uncoupling and depression of the fast sodium channels have differential effects on propagation of the electrical impulse relative to the fiber orientation. Methods and Results. In a model of reentrant ventricular tachycardia (VT) (mean cycle length, 144±13 msec) around a ring of anisotropic myocardium in 10 Langendorff-perfused rabbit hearts, the effects of extracellular K+ concentration ([K'],) and heptanol were studied. [K']. and heptanol each had a dose-dependent effect on VT cycle length. However, high [K+]. slowed the VT mainly by depressing longitudinal conduction, whereas heptanol preferentially depressed transverse conduction. The ratio between longitudinal and transverse conduction velocities progressively decreased with high [K']. and progressively increased with heptanol. Heptanol terminated VT at a mean concentration of 3.5±0.5 mm. The cycle length before termination was 446+120 msec (p<0.001). In eight of 10 experiments, termination occurred by failure of conduction during transverse propagation. VT terminated at a mean [K']. of 11.6± 1.8 mm. The cycle length before termination was 493+341 msec (p<0.01). In seven of 10 cases, termination occurred by failure of conduction during longitudinal propagation. In the remaining five episodes (two with heptanol and three with high [K'].), termination occurred by collision of the reentrant beat with an antidromic impulse being reflected within the ring. In a separate series of six hearts, tetrodotoxin was administered during VT. Like high [K']., tetrodotoxin prolonged the cycle length of the VT by preferentially slowing longitudinal conduction, and VT was terminated by longitudinal block. Conclusions. During reentrant VT, electrical uncoupling of cells by heptanol or modification of active membrane properties by high [K'],, or tetrodotoxin has a differential depressing effect on propagation of the impulse relative to the fiber orientation. (Circulation 1991;84:1307-1318) Deirectional differences in conduction velocity of impulse propagation relative to fiber orientation is a well known characteristic of ventricular myocardium based on a different effective axial resistance parallel and perpendicular to the fiber orientation.1-4 Spach et a15 demonstrated that From the Department of Physiology, University of Limburg, Maastricht, The Netherlands. Supported by a fellowship (to J.B.) from the Royal Netherlands Academy of Arts and Sciences. This research received the first place in the Young Investigators Award competition of the American College of Cardiology, Atlanta, Ga., 1991. Address for reprints: Josep Brugada, MD, Department of Physiology, Biomedical Center, P.O. Box 616, University of Limburg, 6200 MD Maastricht, The Netherlands. Received July 31, 1990; revision accepted April 23, 1991. the passive anisotropic properties of cardiac muscle play an important role in the excitability and safety factor of conduction. They also showed that increasing the junctional resistance with ouabain preferentially depressed impulse conduction transverse to the fiber orientation.6 Balke et a17 and Delmar et a18 demonstrated that the selective increase in junctional resistance by administration of heptanol9,10 caused a See p 1447 preferential slowing of conduction and conduction block of transverse propagating wave fronts without affecting the amplitude or maximum rate of depolarization of the action potential. There is increasing evidence for the reentrant nature of ventricular tachycardias (VTs) in the

1308 Circulation Vol 84, No 3 September 1991 TABLE 1. Effects of Increasing Doses of Heptanol on Ventricular Tachycardia Heptanol (mm) 0 1 2 3 4 Rabbits (n) 10 10 10 10 6 VTCL 144+13 177±35* 231+49t 357±84t 488±118t (+23%) (+60%) (+148%) (+239%) 6L 69±10 63±12* 52+ 13t 46+12t 38±4t (-9%) (-24%) (-33%) (-45%) 0Z 21+4 17±5t 12±5t 9±4t 7±2t (-19%) (-43%) (-57% ) (-67%) OL/Or 3.4+0.5 3.8±0.7 4.3±1* 5.0±lt 5.3 +0.9t VTCL, ventricular tachycardia cycle length (msec); 0L longitudinal conduction velocity (cm/sec); OT, transverse conduction velocity (cm/sec). *p<0.01, tp<0.001. chronic phase of myocardial infarction.1112 During intraventricular reentry, one might expect the impulse to propagate parallel to the fiber direction in some segments of the circuit and perpendicular in others. In the present study, we tested the hypothesis that during sustained reentrant VT, uncoupling agents such as heptanol would terminate VT by selective blockade of transverse conduction, whereas depression of the upstroke of the action potential by high extracellular K' concentration ([K'].) or tetrodotoxin (TTX) terminates VT by preferential depression of longitudinal propagation.5,13 Methods Sixteen Flemish rabbits of either sex weighing 3.4-5.2 kg were used for this study. The experimental model consisted of a ring of healthy epicardium in the left ventricle of Langendorff-perfused rabbit hearts created by a cryoprocedure.1415 A detailed description of the experimental model, recording and stimulation techniques, conduction velocity measurements, and histological examination is given in the companion report.16 Experimental Protocols In 10 experiments, the protocol consisted of two parts executed in the order explained below in five and in the reverse order in the other five. First, the effects of increasing heptanol concentration on reentrant VT were studied. Thirty minutes after induction of sustained reentrant VT by programmed electrical stimulation, heptanol was added to the perfusion fluid in a concentration of 1.0 mm. Every 30 minutes, the concentration of heptanol was increased in steps of 1.0 mm until tachycardia terminated. Continuous on-line recording of activation maps allowed careful monitoring of the time course and dose-dependent effects of heptanol on VT cycle length and longitudinal conduction velocity (0G) measured at the left free wall and transverse conduction velocity (0T) measured at the corridor between the left anterior descending coronary artery (LAD) and the obstacle. The site of VT termination was correlated with the histologically determined fiber orientation. During washout of heptanol, VT was induced again by programmed electrical stimulation; after the cycle length of the VT had returned to its control value, the same protocol was repeated by increasing the [K'], Starting from a control value of 4 mm, the [K+]0 in the perfusion fluid was increased in steps of 2 mm at 5-minute intervals until VT terminated. The effects of increasing [K'], on VT cycle length, 6L and XT, and the site of VT termination were compared with the results obtained during previous heptanol administration. In a separate series of six hearts, TTX was added to the perfusate in increasing doses of 1, 2, 5, 10, 20, and 30,uM until VT terminated. Each dose was given during a period of 5 minutes. Data Analysis Each preparation served as its own control. Results are expressed as mean±+sd. Comparisons between control and administration of heptanol, high [K+], or TTX were performed with Student's t test for paired data. Comparison between the effects of heptanol and the effects of [K+]0 was performed with an analysis of variance. A value of p<0.05 was considered to be statistically significant. Results Characteristics of Ventricular Tachycardia During the control period, epicardial mapping showed that in all cases VT was a result of reentry of the impulse around the obstacle. The cycle length of VT ranged from 130 to 176 msec in different experiments (mean length, 143+13 msec). During the control period of 30 minutes, the cycle length was constant (+3 msec), and spontaneous termination was never observed. During VT, conduction velocity was different in different segments of the ring depending on the angle between the direction of propagation and epicardial fiber orientation. During longitudinal propagation (angle between fiber orientation and direction of propagation, 2±3 ), 0L was 70±9 cm/sec (range, 52-91 cm/sec). During transverse conduction (angle between fiber orientation and direction of propagation, 89+±40), O

Brugada et al Reentry and Safety for Conduction 1309 CONTROL HEPTANOL 1 mm ZT 3 276 l]l31i HEPTANOL 2mM HEPTANOL 3mM FIGURE 1. Activation maps of counterclockwise tachycardia during control period and administration of 1, 2, and 3 mm heptanol. Numbers indicate local activation times in msec. Isochrones are drawn at 10-msec intervals. Large numbers in center of each panel indicate cycle length of tachycardia in msec. Heptanol caused a dose-dependent increase in cycle length of tachycardia because ofpreferential slowing of transverse conduction. LAD, left anterior descending coronary artery. was 21±4 cm/sec (range, 16-26 cm/sec). The anisotropic ratio (0JO/r) was 3.4+0.5 (range, 2.2-4.5). Effects of Heptanol and High [K'], on Ventricular Tachycardia Heptanol had a dose-dependent effect on VT cycle length. The time course of the effect of heptanol showed that steady-state values were reached after approximately 20 minutes. Heptanol (1.0 mm) increased the cycle length by 23% (p<0.002), 2.0 mm increased it by 60% (p<0.001), 3.0 mm increased it by 148% (p<0.001), and 4 mm increased it by 239% (p<0.001, n=6). Heptanol also had a dose-dependent effect on conduction velocity; however, 6G was

1310 Circulation Vol 84, No 3 September 1991 CONTROL 3 mm Heptanol 5? 6 8 12 11 6 1963 19 15 17 19 139 101 1 60 0 42 27 26 27 24 26 22 18 7 47 14 33 35 35 33 33 199 18962 164 34 41 43 41 49 19 2123 5059 ~ ~~ ~ -- 132ms 358 ms A 2 c 62. C 8 9 10 1- ---- 12-13-- 14 -- FIGURE 2. Maps, schematics, and electrograms of impulse propagation through narrow corridor between left anterior descending coronary artery and obstacle during a single tachycardia before and after administration of 3 mm heptanol. Upper panels: Activation times are given in msec, and isochrones are drawn at 10-msec intervals. Distance between electrodes is 2.25 mm. Middle panels: Pathway of impulse propagation is schematically indicated. Encircled letters and numbers refer to sites of recording of electrograms given in lower panels. During control, cycle length of ventricular tachycardia was 132 msec. After heptanol administration, cycle length was prolonged to 358 msec, mainly as a result of extreme slowing of conduction in this area. (See text for further description.)

Brugada et al Reentry and Safety for Conduction 1311 TABLE 2. Effects of Increasing Extracellular K' Concentration on Ventricular Tachycardia [K+]. (mm) 4 6 8 10 12 14 Rabbits (n) 10 10 10 10 5 4 VTCL 142±13 159+34* 194±54t 259±96t 381±89t 774±378t (+12%) (+37%) (+82%) (+168%) (+445%) OL 71±8 63±7t 53±84 36±84 21±5t 11±3t (-11%) (-25%) (-49%) (-70%) (-85%) OT 21±3 19±3t 18±4: 15±44 10±24 8±3t (-10%) (-14%) (-29%) (-52%) (-62%) k/ot 3.4±0.4 3.4±0.4 3.0±0.7 2.6±0.81 2.0±0.3t 1.5±0.3t [K+]0, extracellular potassium concentration; VTCL, ventricular tachycardia cycle length (msec); 6L, longitudinal conduction velocity (cm/sec); 6r, transverse conduction velocity (cm/sec). *p<0.05, tp<0.01, tp<0.001. more sensitive to heptanol administration than 6L. As a result, the anisotropic ratio progressively increased from 3.4+0.5 during control to 5.3±0.9 (p<0.01) for 4 mm heptanol (see Table 1). In Figure 1, the activation maps during a sustained counterclockwise VT are shown during control and administration of 1, 2, and 3 mm heptanol. During control, the VT had a stable cycle length of 147 msec. At the corridor between the LAD and the obstacle where the impulse propagated perpendicular to the fiber orientation, crowding of isochrones indicates that conduction velocity was much slower (6,, 21 cm/sec) than in the left free wall where the impulse propagated parallel to the fiber orientation (6L, 69 cm/sec). At increasing doses of heptanol, the VT slowed progressively to a cycle length of 345 msec. This increase in cycle length was caused mainly by a preferential slowing of transverse conduction at the corridor between the LAD and the central obstacle, whereas longitudinal conduction was less affected. For 3 mm heptanol, areas of very slow conduction and conduction block (local delays as long as 60 msec between two electrodes) appeared in the segment of transverse conduction. Effective 6, was now reduced to 8 cm/sec, whereas 6L was 54 cm/sec. In all experiments, average O6 and 6L values during administration of 3 mm heptanol were 9±4 and 46±12 cm/sec. In Figure 2, another example of the effects of heptanol on transverse conduction is shown. In this figure, only the transverse conduction at the corridor between the LAD and the obstacle is shown together with some electrograms recorded from that region. During control, the counterclockwise VT had a cycle length of 132 msec. In the segment of transverse conduction, the circulating impulse was propagating slowly but smoothly. The pairs of unipolar electrograms (A-A', B-B',...) recorded at either side of the narrow pathway were almost simultaneously activated, demonstrating that the impulse was transmitted through this area as a uniform, nonfragmented wave front. After the administration of 3 mm heptanol, the cycle length of the VT was considerably prolonged to 358 msec. The activation map from the area of transverse conduction showed that this prolongation of VT cycle length was caused mainly by a dramatic increase in conduction time in this segment of the circuit. During control, this segment of the ring was activated in 49 msec, whereas during heptanol administration (3 mm), the total conduction time prolonged to 259 msec. As shown schematically in Figure 2 (middle panel), the impulse now propagated in a zig-zag pattern, meandering around multiple lines of transverse block (thick lines), making the actual length of the pathway about fourfold longer than that during control. Stepwise increase of [K'], in the same hearts caused a progressive prolongation of the VT cycle length. At each concentration, steady-state values were obtained after approximately 3 minutes. The VT cycle length increased by 12% for 6 mm [K'], (p<0.04), by 37% for 8 mm (p<0.006), by 82% for 10 mm (p<0.003), by 168% for 12 mm (p<0.004, n=5), and by 445% for 14 mm (p<0.04, n=4). Comparison of the effects of [K']. on longitudinal and transverse conductions showed that increasing [K']. depressed longitudinal conduction more than it depressed transverse conduction. As a result, the anisotropic ratio progressively decreased at increasing [K']. concentrations-from 3.4+0.4 during control to 1.5±0.3 for 14 mm [K+], (p<0.001) (see Table 2). In Figure 3, the activation maps of a clockwise VT during 4, 6, 10, and 12 mm [K+], are shown. During control, the VT had a stable cycle length of 134 msec, 6, was 19 cm/sec, and OL was 66 cm/sec (ratio, 3.5). At increasing [K+]0, the VT cycle length progressively prolonged to 467 msec at 12 mm [K+].. During administration of 12 mm [K+]., 6L was 19 cm/sec and 6, was 9 cm/sec (ratio, 2.1). Comparison of the effects of heptanol and high [K+], on the anisotropic ratio is shown in Figure 4. During increasing doses of heptanol, the anisotropic ratio progressively increased from 3.4 to 5.3. In contrast, during high [K+],, the anisotropic ratio progressively decreased from 3.4 to 1.5. Termination of Venticular Tachycardia by Heptanol and High [K+]0 Heptanol administration and high [K+]. terminated all episodes of VT. The mean concentration of

1312 Circulation Vol 84, No 3 September 1991 K+ 4 mm 53 56 59 60 63\70 76 334546053 55 565616 996 222 3623 4 24162 67 675 98 41464554 65 68 72 35 42555564 68928 1 3313\ 661 64 67 68 7 86 06 0132 7 126 125 126 65 66 67 7481 124 12 1 7 71 C 12 127 123 120 6114 72 72 74 76 127125 121 119116 134 ms 78 7S 76 71 A2 125122 19 116 82 80 82 84 128 129115 89 84 84 126 126 a1 87 8 Ii 123 123 124 107 in>4 126 122 122 117 91 12S 114 11 1312113 94 94 115 114 11 94 111 115 112 94 1111121 e 8 apex K+6 mm 52 54 56 160 67 \72 78 3343 47054567 63 6 737680 21235334541 51 6163 707 8 i2 19\2326w 3 4648553466&0 74758 3 5 3843\565964673t185 4;iF=18 60 64 68 2 7 8 95 137 134 133 132 131 131 66 67 6 75 80 5 134 131 6 125 123 73 70 71 7 133 131 28 124 120 118 73 73 75 77 1331 2612412 142 ms 132 131 28 125 120 84 82 83 86 131 129 125 120 94 8 36 125 129 128 126124 U 8 86 13 122 123 125 126 12 111 128 124 120 125 125 123 1 115 93 91 126 121 121 127 12 11 95 116 117 125 12 55 113 113 1 13 a9 110 114118 113 (2 111 111 52 58 616 70 1t 15 C?42 545 36372786 24 21 s.\m5 38 567 6 72 7q 37 5 1211 619\39<\ 37A40 451 76 8553 3 24 6 414 264 7191 91 jjj1991 11 3 8 68 71 7 83362 192 193 194 32032 77 79 83 8894 18118t 5319 80 84 8 L761783192 190 9 82 161751 187 50199 215mns 11 s 1 1740 0158 1 t71 11 1 167 176 163 78 17 72 159 167 1 7 193 193 162 54 15 7117 90 1 1 4 163 158 61 15 5 198138 1" 51S 1 13 38 /3 K+ 10 mm K+ 12mM FIGURE 3. Activation maps of a single clockwise circulating tachycardia during increasing concentrations of extracellular K' (4, 6, 10, and 12 mm). Cycle length of tachycardia (large numbers in msec in center of map) increased with increasing concentrations of K+. Numbers indicate local activation times in msec. Isochrones are drawn at 10-msec intervals. LAD, left anterior descending coronary artery. heptanol at which VT terminated was 3.6±0.5 mm. Before termination, the VT cycle length was 446±+120 msec (p<0.001 versus control). Mean [K'], at which VT terminated was 11.6±+1.8 mm. The VT cycle length before termination was 493+±341 msec (p<0.01 versus control). During heptanol administration in eight of 10 experiments, termination occurred by failure of the wave front to propagate through the corridor between the LAD and the obstacle. During high [KJ]o in seven of 10 experiments, termination occurred by longitudinal conduction block. In Figure 5, the termination of a clockwise

6 5 >4 I3 2 1 o Heptanol,mM Heptanol 0 1 2 3 4 K+ mm K' 4 6 8 10 12 14 FIGURE 4. Plots of anisotropic ratios of longitudinal conduction velocity (LCV) and transverse conduction velocity (TCV) during tachycardia to increasing doses ofheptanol and extracellular K' concentrations. *p<0.01 compared with control. VT during heptanol administration (upper panels) and during high [K'], (lower panels) is shown with the histological determination of fiber orientation at the site of termination. The tachycardias, which were initiated in the same heart, were identical during control (cycle length, 165 msec). During heptanol administration, shortly before termination (left upper panel), the VT had a cycle length of 390 msec because of severe depression of transverse conduction. Several areas of large local conduction delays and crowding of isochrones were found in this area. Termination of VT (middle upper panel) occurred as a result of failure of the impulse to propagate across one of these areas. As shown by the histological section in the right upper panel, conduction was perpendicular to the epicardial fiber orientation in the area of block. During high [K'],, the last reentrant beat (left lower panel) had a cycle length of 386 msec. Although transverse conduction in the corridor between the LAD and the obstacle now remained uniform, local delays in conduction and crowding of isochrones appeared in the free wall of the left ventricle during longitudinal conduction of the circulating wave front. Termination of VT by high [K+]0 (middle lower panel) occurred by failure of the impulse to propagate across these areas of conduction delay. Histological determination of the fiber orientation showed that conduction block occurred while the impulse was propagating parallel to the fiber orientation (right lower panel). In a minority of experiments (two of 10 during heptanol and three of 10 during high [K'],), termination of VT was not caused by primary conduction block in a certain segment of the circuitous pathway but rather by collision with another wave front, which was reflected in the ring. In Figure 6, an example of termination of VT by collision with a returning wave during heptanol administration is given. Fifteen unipolar electrograms spaced at equal intervals around Brugada et al Reentry and Safety for Conduction 1313 the obstacle are shown. During sustained VT, the circulating impulse propagated from electrode 1 to electrode 15 with fast conduction between electrodes 1 and 8 and slow conduction between electrodes 10 and 14, which were located at the corridor between the LAD and the obstacle. During the last beat of the VT, the sequence of activation was suddenly reversed in part of the ring (electrodes 1-9) and propagated as a returning impulse against the direction of the original circulating impulse. The two wave fronts propagating in opposite directions around the ring collided at electrode 1, resulting in sudden termination of the VT. During heptanol administration, the point of reflection in both cases was located in the area of severely depressed transverse conduction, whereas during high [K+]0, in all three cases the returning wave originated from the area of depressed longitudinal conduction. The isochrone maps from the area of reflection showed that the returning wave was not caused by "true" reflection.17 A returning wave originated when the circulating impulse encountered arcs of conduction block not extending along the entire width of the ring. As a result, the impulse could continue its normal orthodromic course while simultaneously starting an antidromic wave by microreentry through the incomplete arc of unidirectional block. Effects of Tetrodotoxin on Ventricular Tachycardia In a different series of six hearts, the effects of YTX on reentrant VT were studied. TTX was administered at increasing doses (1, 2, 5, 10, 20, and 30,uM) at 5-minute intervals until VT terminated. The mean concentration of TTX at which VT terminated was 25+5,uM. The VT cycle length increased from 167+17 msec during control to 793 ± 122 msec before termination (p<0.001). 0L and 0T values during control were 67±7 and 23+2 cm/sec (anisotropic ratio, 2.9 0.3) compared with 17+6 and 8±2 cm/sec (ratio, 2.1±0.3), respectively, during TTX just before termination of VT. In four of six hearts, VT terminated by longitudinal conduction block. In the remaining two hearts, VT terminated by collision of the circulating impulse with an antidromically returning wave front. An example of the effects of TTX on VT is given in Figure 7. During control, the counterclockwise VT had a cycle length of 171 msec. Administration of 5 and 10,uM TTX prolonged the cycle lengths to 333 and 610 msec, respectively. A further increase in the dose of TTX to 20,uM prolonged the cycle length to 752 msec, and VT was terminated by longitudinal conduction block at a segment of the ring where conduction velocity was 62 cm/sec during control. Discussion Our experimental model consisted of a ring of normal perfused epicardium with uniform anisotropic properties. Because of the natural epicardial fiber orientation, during reentrant VT the circulating wave front travels parallel to the fiber orientation in some segnents and perpendicular to the fiber axis in other

1314 Circulation Vol 84, No 3 September 1991 Heptanol 4 mm 500,u1m K+ 12 mm - apex FIGURE 5. Activation maps during termination of tachycardia by heptanol (upper panels) and high extracellular K' (lower panels). Numbers indicate local activation times in msec. Isochrones are drawn at 10-msec intervals. Thick line and double bars indicate sites of termination of tachycardias. As can be seen from histological examinations (right panels), conduction block of circulating impulse during heptanol administration occurred while wave front propagated transverse to fiber orientation and during high K' while wave front propagated parallel to fiber orientation. LAD, left anterior descending coronary artery. segments of the ring. During control, conduction velocity was about threefold faster during longitudinal than during transverse propagation. The importance of tissue anisotropy for reentrant circuits has been emphasized by several authors.15'18-20 It has been suggested that areas of very slow conduction with multiple-component low-amplitude electrograms during reentrant tachycardia represent intrinsic asymmetry of cardiac activation because of fiber orientation.18 Enhanced tissue anisotropy could be a cause of VT because activation transverse to myocardial fibers would be sufficiently slow to permit reentry.20 If anisotropy plays an important role in initiation and perpetuation of reentrant tachycardias, the question arises of whether a new class of antiarrhythmic agents acting primarily on the passive electrical properties of myocardial fibers could be of value in prevention and treatment of reentrant tachyarrhythmias.21 Our study was designed to compare the effects of an uncoupling agent (heptanol) on reentrant VT with the effects of other substances that depress conduction velocity by inactivation or blockade of the fast sodium channels (high [K+],) and TTX). Role of Anisotropy In the early phase of ischemia, impulse conduction is slowed by a depression of the active membrane properties plus an increase in extracellular and axial resistance.22 In the chronic phase of myocardial infarction, Ursell et a123 demonstrated that although

Brugada et al Reentry and Safety for Conduction 1315 395 msec FIGURE 6. Fifteen unipolar electrograms spaced at equal intervals around obstacle are displayed during termination of tachycardia. During stable tachycardia, activation sequence was ordered from electrograms 1 to 15 with a cycle length of395 msec. Sudden change in sequence of activation produced two different wave fronts circulating in opposite directions that collided at electrogram 1, suddenly terminating tachycardia. Arrows indicate direction of activation. the action potentials had become normal again, conduction of the electrical impulse was still impaired because of the development of fibrotic septa disrupting the electrical coupling between cells in a transverse direction. Several investigators used uncoupling agents to study the role of intercellular resistance on anisotropic conduction.7'0'24 These studies showed that uncoupling had a preferential effect on conduction perpendicular to the fiber axis and in regions with a high coupling resistance in and around a myocardial infarction.1024 It has been suggested that by producing complete conduction block in these areas, the abnormalities in propagation in infarcted myocardium might be eliminated.24 In the present series of experiments, the administration of heptanol during reentrant VT affected transverse conduction to a greater extent than longitudinal conduction. This was the opposite of the effects of high [K], or TTX, which mainly affected longitudinal conduction. As a result, the anisotropic ratio in conduction velocity was increased by heptanol and decreased by high [K], or TTX. Termination of reentrant VT by heptanol occurred because of failure of transverse conduction. In contrast, termination of VT by high [K'], or 1TX was caused by failure of longitudinal conduction. However, large doses of these substances were needed to terminate VT; before termination, the VT cycle length was threefold to fourfold that of the control VT interval. This suggests that in our model, the safety factor for conduction was high both parallel and perpendicular to the fiber orientation. Thus, because of the high degree of safety factor uniformity in a circuit of normal anisotropic myocardium, a general depression of either passive or active membrane properties did not terminate VT at a suitable therapeutic range. On the other hand, in diseased myocardium, one might expect circuitous pathways that exhibit large safety factor differences. In such a substrate of reentrant VT, a moderate increase in intercellular resistance or decrease in generated excitatory current might specifically block conduction in a segment of the circuit with a relatively low safety factor. If in this segment the fibers are oriented perpendicular to the direction of the circulating impulse, an uncoupling agent might be most effective; in cases in which the impulse propagates longitudinal to the fiber axis, a sodium channel blocker might be more successful in terminating the tachycardia. Termination of Tachycardia by a Retuming Impulse in the Ring In seven of 26 VTs (two with heptanol, three with high [K'],, and two with TTX), termination of VT occurred by collision of the circulating wave front with a spontaneous antidromic returning impulse. A common feature of these returning impulses was that they originated from areas in the ring where conduction was very slow and marked local conduction delays were present. In these areas, the wavelength of the impulse was markedly shortened, allowing microreentry within the width of the ring. During heptanol administration, the returning waves always originated from the area of depressed transverse conduction, whereas during high [K+]0 and TTX administration, the site of a spontaneously returning

1316 Circulation Vol 84, No 3 September 1991 CONTROL TTX 5yM 0 aj 74 66154147 138 36 36 37 38 84 8 680 5 46 5 34 33 36 35 36 35 11 7 is 233 31 32 34 34 35 1 3878276 534 33313734 211 06101\ 9 79 23 25 25 26 29 28 114112 0610 21 22 21 21 23 23 12 111511 06 9 19116 19 17 11 12 13 137136 13325122 171 15 16 10 11 10 3 12 107 6 153 154 156 7 7 6 6 151 155 158 5 4 154 157 2 4 4 157 171 1 5 158 161 166 167 171 4 159 166 165 166 167 5 164 161 165 165 166 161 163 Apex TTX 10 pm TTX 20,M FIGURE 7. Activation maps during increasing doses of tetrodotoxin (TTX). During control, impulse propagated counterclockwise with a cycle length of 171 msec. For 5 and 10 AM T7X, cycle lengths prolonged to 333 and 610 msec, respectively. For 20,M TTX, tachycardia terminated by failure of conduction at left free wall. Cycle length before termination was 752 msec. Numbers indicate local activation time in msec. Isochrones are drawn at 10-msec intervals. Double bars indicate conduction block. LAD, left anterior descending coronary artery.

antidromic wave was located in a segment of depressed longitudinal conduction. In some cases, such as the one presented in Figure 2, electrical uncoupling changed uniform transverse conduction into a zig-zag pattern around multiple arcs of transverse block. This zig-zag pattern often preceded microreentry, resulting in a returning antidromic wave that terminated VT. Interestingly, in human atrial bundles, Spach et a125 observed a similar phenomenon during longitudinal propagation when sodium conductance was impaired. They showed that the induction of premature action potentials changed uniform longitudinal conduction into a dissociated zig-zag type of conduction around arcs of longitudinal block. The phenomenon of zigzag conduction appears to be of importance for the characteristics of reentrant VT by creating an area of very slow apparent conduction velocity. In the example shown in Figure 2, 0T during control was 19 cm/sec. After heptanol administration, the apparent conduction velocity decreased to 4 cm/sec. However, because of the zig-zag pattern of conduction during heptanol, the actual path length was fourfold longer than that during control. If this prolongation in path length is taken into account, a true conduction velocity of 18 cm/sec becomes evident. Thus, an area of microscopic zig-zag conduction might provide an area of extremely slow conduction in part of the circuit, whereas the actual conduction velocity is still within the physiological range. Study Limitations The use of pharmacological blocking agents to study basic electrophysiological properties is limited by other possible nonspecific drug effects. Our assumptions that heptanol affects only passive membrane properties and that high [Kt]0 selectively depresses active membrane properties are probably not entirely true. The question has been raised as to what extent the effects of heptanol on conduction are results of changes in the sodium conductance (gna).26 In dog cardiac Purkinje fibers and squid giant axons, it has been shown that apart from its main effect on internal resistance (R1), heptanol also reduces gna.2728 However, Jalife et a129 concluded that the depression of conduction by heptanol could not be explained by such a reduction in gna. They showed Brugada et al Reentry and Safety for Conduction 1317 that during complete blockade of conduction by heptanol in sheep Purkinje fibers, the cells were still able to generate normal action potentials, suggesting that the effect of heptanol on conduction was caused mainly by an increase in Ri. Our observation that longitudinal conduction was only moderately depressed during seriously depressed transverse conduction by heptanol supports this conclusion. The effects of increased [K']. on impulse propagation are thought to be caused primarily by a decrease in the membrane resting potential, resulting in decreases in threshold potential and gna and prolongation of the time constant of the foot of the action potential and the reactivation kinetics of the sodium channel system.30 However, at high [K'], values, an increase in K' conductance and a concomitant decrease in membrane resistance have also been reported.31 No effect on axial (myoplasmic and gap junction resistance) or extracellular resistance has been found.31,32 Gettes et a133 demonstrated that the slowing of conduction produced by extracellular K' concentrations of more than 9 mm were caused mainly by a depression of the fast sodium channels. Because of the possible effects of K' on membrane resistance, we studied the effects of a selective sodium channel blocker (TTX) in an additional series of six hearts. The effects of TTX on VT were comparable to those resulting from an increase in [K'],. As during high [Kt]0 administration, longitudinal propagation was preferentially depressed and VT was terminated by longitudinal conduction block. This suggests that the effects of high [K'], on VT were caused mainly by a decrease in the active membrane properties. References 1. Sano T, Takamaya N, Shimamoto T: Directional difference of conduction velocity in cardiac ventricular syncytium studied by microelectrodes. Circ Res 1959;7:262-267 2. Clerc L: Directional differences of impulse spread in trabecular muscle from mammalian heart. JPhysiol (Lond) 1976;255: 335-346 3. Spach MS, Miller WIT, Geselowitz D, Barr RC, Kootsey JM, Johnson EA: The discontinuous nature of propagation in normal canine muscle: Evidence for recurrent discontinuities of intracellular resistance that affect the membrane currents. Circ Res 1981;48:39-54 4. Spach MS, Miller WT III, Miller-Jones E, Warren RB, Barr RC: Extracellular potentials related to intracellular action potentials during impulse conduction in anisotropic canine cardiac muscle. Circ Res 1979;45:188-204 5. Spach MS, Kootsey JM: The nature of electrical propagation in cardiac muscle. Am J Physiol 1983;244:H3-H22 6. Spach MS, Kootsey JM, Sloan JD: Active modulation of electrical coupling between cardiac cells of the dog: A mechanism for transient and steady state variation in conduction velocity. Circ Res 1982;51:347-362 7. Balke CW, Lesh MD, Spear JF, Kadish A, Levine J, Moore EN: Effects of cellular uncoupling on conduction in anisotropic canine ventricular myocardium. Circ Res 1988;63:879-892 8. Delmar M, Michaels DC, Johnson T, Jalife J: Effects of increasing intercellular resistance on transverse and longitudinal propagation in sheep epicardial muscle. Circ Res 1987; 60:780-785 9. Johnston MF, Simon SA, Ramon F: Interaction of anesthetics with electrical synapses. Nature 1980;286:498-500 10. Deleze J, Herve JC: Effect of several uncouplers of cell-to-cell communication on gap junction morphology in mammalian heart. J Membr Biol 1983;74:203-215 11. Josephson ME, Horowitz LN, Farshidi A, Spear JF, Kastor JA: Recurrent sustained ventricular tachycardia: I. Mechanisms. Circulation 1978;57:431-439 12. De Bakker JMT, Van Capelle FJL, Janse MJ, Wilde AAM, Coronel R, Becker AE, Dingemens KP, Van Hemel NM, Hauer RNW: Reentry as a cause of ventricular tachycardia in patients with chronic ischemic heart disease: Electrophysiologic and anatomic correlation. Circulation 1988;77:589-606 13. Tsuboi N, Kodama I, Toyama J, Yamada K: Anisotropic conduction properties of canine ventricular muscles: Influence of high extracellular K' concentration and stimulation frequency. Jpn Circ J 1985;49:487-498 14. Brugada J, Boersma L, Kirchhof CJHJ, Brugada P, Havenith M, Wellens HJJ, Allessie MA: Double wave reentry as a

1318 Circulation Vol 84, No 3 September 1991 mechanism of acceleration of ventricular tachycardia. Circulation 1990;81:1633-1643 15. Allessie MA, Schalij MJ, Kirchhof CJHJ, Boersma L, Huyberts M, Hollen J: Experimental electrophysiology and arrhythmogenicity: Anisotropy and ventricular tachycardia. Eur Heart J 1989;10:E8-E14 16. Brugada J, Boersma L, Kirchhof C, Heynen V, Allessie M: Reentrant excitation around a fixed obstacle in uniform anisotropic ventricular myocardium. Circulation 1991;84:1296-1306 17. Antzelevitch C, Jalife J, Moe GK: Characteristics of reflection as a mechanism of reentrant arrhythmias and its relationship to parasystole. Circulation 1980;61:182-191 18. Cardinal R, Vermeulen M, Shenasa M, Roberge F, Page P, Helie F, Savard P: Anisotropic conduction and functional dissociation of ischemic tissue during reentrant ventricular tachycardia in canine myocardial infarction. Circulation 1988; 77:1162-1176 19. Richards DA, Blake GJ, Spear JF, Moore EN: Electrophysiologic substrate for ventricular tachycardia: Correlation of properties in vivo and in vitro. Circulation 1984;69:369-381 20. Dillon SM, Allessie MA, Ursell PC, Wit AL: Influences of anisotropic tissue structure on reentrant circuits in the epicardial border zone of subacute canine infarcts. Circ Res 1988;63: 182-206 21. Wit AL: Anisotropic reentry: A model of arrhythmias that may necessitate a new approach to antiarrhythmic drug development, in Rosen MR, Palti Y (eds): LethalArrhythmias Resulting From Myocardial Ischemia and Infarction. Boston, Kluwer Academic Publishers, 1989, pp 199-213 22. Kleber AG, Riegger CB, Janse MJ: Electrical uncoupling and increase of extracellular resistance after induction of ischemia in isolated, arterially perfused rabbit papillary muscle. Circ Res 1987;61:271-279 23. Ursell PC, Gardner PI, Albela A, Fenoglio JJ Jr, Wit AL: Structural and electrophysiological changes in the epicardial border zone of canine myocardial infarcts during infarct healing. Circ Res 1985;56:436-451 24. Spear JF, Balke CW, Lesh MD, Kadish AH, Levine JL, Moore EN: Effect of cellular uncoupling by heptanol on conduction in infarcted myocardium. Circ Res 1990;66:202-217 25. Spach MS, Dolber PC, Heidlage JF: Influence of the passive anisotropic properties on directional differences in propagation following modification of the sodium conductance in human atrial muscle. Circ Res 1988;62:811-832 26. Mackielski JC, Nelson WL: Effects of cellular uncoupling by heptanol on conduction in infarcted myocardium (letter to the editor). Circ Res 1990;67:1299-1300 27. Nelson WL, Makielski JC: Block of cardiac sodium current by heptanol and octanol (abstract). Biophys J 1990;57:299a 28. Oxford GS, Swenson RP: n-alkanols potentiate sodium channel inactivation in squid giant axons. Biophys J 1979;26: 585-590 29. Jalife J, Sicouri S, Delmar M, Michaels DC: Electrical uncoupling and impulse propagation in isolated sheep Purkinje fibers. Am J Physiol 1989;257:H179-H189 30. Gettes LS: Effects of ionic changes in impulse propagation, in Rosen MR, Janse MJ, Wit AL (eds): Cardiac Electrophysiology. Mount Kisco, NY, Futura Publishing Co, 1990, pp 459-479 31. Dominguez G, Fozzard HA: Influence of extracellular K' concentration on cable properties and excitability of sheep cardiac Purkinje fibers. Circ Res 1970;26:565-574 32. Buchanan JW Jr, Oshita S, Fujino T: A method for measurement of internal longitudinal resistance in papillary muscle. Am J Physiol 1986;251:H210-H217 33. Gettes LS, Buchanan JW Jr, Saito T, Kagiyama Y, Oshita S, Fujino T: Studies concerned with slow conduction, in Zipes DP, Jalife J (eds): Cardiac Electrophysiology and Arrhvthmias. Orlando, Fla, Grune & Stratton, 1985, pp 81-87 KEY WORDS. anisotropy * heptanol * tetrodotoxin. ventricular tachycardia * potassium