Paradoxical QRST integral changes with ventricular repolarization dispersion

Journal of Electrocardiology Vol. 32 Supplement 1999

Paradoxical QRST Integral Changes With Ventricular Repolarization Dispersion Osamu

Okazaki,

MD,

and

Robert

L. L u x , P h D

Abstract: Body surface QRST integral (QRSTI) maps have been shown theoret-

ically to reflect disparity of intrinsic repolarization properties and have been experimentally linked to increased arrhythmia susceptibility. Paradoxically, a lower magnitude of QRSTI in patients with heart disease and at risk for arrhythmias has been reported. We hypothesized that this paradoxical reduction in QRST magnitude is a consequence of increased heterogeneity of repotarization gradients in normal hearts. We generated QRSTI using a previously published heart model to compare QRSTI for aligned and random repolarization gradients. The heart model consisted of 50,000 cubic units in an anatomically correct arrangement that included parameters to simulate anisotropic conduction and inhomogeneous distribution of refractoriness. Body surface potential maps (BSPMs) were generated on a torso surface assuming a homogeneous torso and using the boundary element method for normal alignment of repolarization gradients and spatially reassigned repolarization values that randomized repolarization directions. QT duration was measured by the subtraction of Q onset time from T offset time on the BSPM. T offset was defined as the last potential to be detected at intervals of 3 ms that was above the threshold of 0.1 mV during recovery. The time of T offset showed a consistent tendency to shift to the left posterior and to split. When slow conduction velodties were assigned, BSPMs showed delayed propagation and multiple extrema. QRSTI showed systematic magnitude decrease with increasing randomness of repolarization gradient direction. Ventricular fibrillation (VF) could be induced by successive extrastimuli under the conditions of over 70% deviation and slow conduction of 0.5 m/s for the longitudinal direction. In conclusion, a possible explanation for the paradoxical reduction in QRSTI in the presence of constant repolarization disparity is the change in alignment of repolarization gradients. K e y w o r d s : QT dispersion, QRST integral map, repolarization, inhomogeneity, vulnerability.

B o d y surface QRST integral (QRSTI) m a p p i n g has b e e n reported to reflect intrinsic ventricular recovery

properties. The QRST area should be a m e a s u r e of i n h o m o g e n e i t y of ventricular repolarization properties (1,2). Experimental data have s h o w n that increased refractory dispersion a n d QRST area changes of the electrogram w e r e highly correlated with decreased ventricular fibrillation (VF) threshold (3). The vulnerability of the ventricles to arrhythmias increases in the presence of g r e a t e r - t h a n - n o r m a l disparity of local recovery times by QRSTI m a p p i n g (4). A n

From the Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah, Salt Lake City, Utah.

Reprint requests: Osamu Okazaki, Nora Eccles Harrison CVRTI, University of Utah, Building 500, 95 South 2500 East, Salt Lake City, UT 84112-5000. Copyright © 1999 by Churchill Livingstone ®

0022-0736199/320S-0013510.00/0 60

Paradoxical QRST Integral Changes

Model (Cardiomaster)

.%:

!~ ;

A~dsotropicco~dactio~

x

ar Gradient

.

1o ~ye~ Z~y,,

L,_Z'.

~ ~ ~ _ _ z~yocaratat,an~orropy

io

ratio

-

,

~'t,,,::!:2:::} ~

Ep 1 o0

'~

,~oo

a:b

,,.,.,,,

:~oo

=

3:1

Frontal

~."r.i".-.-~i .

Duration

Homzontal .

' ~'~x"~'1

"~

~,' o

Okazaki and Lux 61

•

.

.

~_"qI~

,~..---.,,

'.k.~z..., Saglttal

Human Torsomodel

F i g . 1. T h r e e - d i m e n s i o n a l m o d e l w i t h s p a t i a l r e s o l u t i o n of 1.5 m m a n d m y o c a r d i a l a n i s o t r o p y i n t o h u m a n t o r s o m o d e l .

Table

1. E l e c t r o p h y s i o l o g i c a l P a r a m e t e r s a n d A b b r e v i a t i o n s

phase I 2 Waveforms -100

Node

Atrium

AV node

30 40 40 145 200 275 -60 -30 20 250 0 1.0 0.5

0 2 20 128 140 190 -60 20 20 0 0 1.0 1.0

10 10 80 175 200 315 -100 20 20 0 0 1.0 0.1

Sinus Parameter

Unit

To ms T 1 ms T2 ms T3 ms ARP ms FRT ms V o mV V 1 mV V2 mV GRD ms/layer DVT % ECF CV m / s

His Bundle & BB

Purkinje

Ventricle

0 0 80 175 200 295 -100 20 20 0 0 1.0 2.5

0 0 80 275 200 395 -100 20 20 0 0 1.0 2.5

0 5 80 175 200 295 -100 20 20 6 0 -- 100 1.0 0.75 (L) 0.25 (T)

APD, Action potential duration; SN, sinus node; A, atrium; AVN, atrioventricular node; HB, His bundle; BB, bundle branch; V, ventricle; To_3, time of APD phase 0-3; ARP, absolute refractory period; FRP, full refractory period; Vo_2, potential (0, resting; 1, action; 2, plateau); GRD, Gradient of action potential distribution; DVT: deviation of APD; ECF, effective conductivity factor; CV, conduction velocity; L, longitudinal; T, transverse.

62

Journal of Electrocardiology Vol. 32 Supplement 1999

II

Measurement of QT duration

mY

RI~

i

•

Q onset

,

T offset - Q onset QTe =

T offset (420 msec)

4

P t.0 .,

t,0

24 msec

RR

T

(~ ST

S.O

177 msec

378 msec

192 msec

399 msec

(sec)

:::::::::::::::::::::::::::::::::::::::::::::

129 msee

159 msec

,

I o-°1

I 1 9 8 rr,..see

[ [

I 201 msec

162 rnsee

liig N Q onset

-

420 msec

R

QT d u r a t i o n

]

°

431 msec

T

offset

Ventrieular Activation and Recovery Process

Iht,~rvaL -" ST p e i n t : T~tc~,~[A: • ~ 5T : • si~ ST: -n 5T : -~it~ 51":

O,LI~| m s ~20 m s O,iOO 'mY 3 CO~tS G,3tO 'm.v G cgunts O.ODO mY

Fig. 2. QT interval defined using sequential isopotential maps.

abnormal dipolar QRSTI map pattern was observed in patients with the idiopathic long QT syndrome w h o are k n o w n to be at risk for ventricular arrhythmias (5). The QRSTI was found to be largely independent

of the activation sequence and to be a measure of the inhomogeneity of ventricular repolarization. Ventricular areas of slow conduction, regionally delayed repolarization or dispersion in repolarization can be

C o n t r o l a n d R a n d o m M o d e l at the s a m e level "42" 42

42 Fig. 3. Random model APD deviation. For the assignment of random repolarization distribution, a percentage from 10% to 100% of normally arranged APDs were reassigned at random according to the table of a random sampling number to the other sites.

Control

ge

10% random 42

42

600/0

42

40% 42

800/o

100%

Paradoxical QRST Integral Changes

•

Okazaki and Lux 63

1~!lO-I:mdECG

t

, •, k i, -I,

I.,,

,

H -

,.. H.r..I~..,[,.-H,_-~HF__~II__It__H~__H~__k.._H......

i .......

, H. '..

+4.-H,..+-H,..HLF-H,-Hk-k-H.q.

, i!' i~:~:i~:i~: ~ i~:~'~i ~:t~i~: ~::;i .:i

Normal Model

..... ....

Interval

QI~'I'I Map

Fr,~

Control

I'o

i

:

mlx~tY

,

i~ll u

,

420 u

EGG

~"N. ~.~

0

] ~000

'.",( 1

" ~ ~..,..._

s"

F i g . 4. 1 6 0 - l e a d ECG, s t a n d a r d ECG, a n d QRSTI m a p of n o r m a l m o d e l . A r e a m a p of QRSTI o b t a i n e d f r o m t h e QT m e a s u r e m e n t d a t a o n BSPMs. T h e r e c t a n g u l a r f r a m e r e p r e s e n t s t h e t o r s o of w h i c h t h e left h a l f r e p r e s e n t s t h e a n t e r i o r c h e s t a n d t h e r i g h t h a l f t h e b a c k w i t h t h e z e r o p o t e n t i a l . M a g n i t u d e s of t h e m a x i m u m a n d m i n i m u m a r e 38 m V / m s a n d - 1 4 m V / m s , respectively.

identified with abnormal dipolar and nondipolar QRSTI maps in patients with idiopathic VF (6). QT intervals were spatially distributed on the torso in a manner predictable from regional cardiac events (7). We hypothesized that this paradoxical reduction in QRST magnitude is a consequence of increased heterogeneity of repolarization gradient directions in diseased hearts in contrast to aligned repolarization gradients in normal hearts. The inhomogeneous properties and other factors may produce the vulnerability to ventricular arrhythmia in correspondence with the recovery process. This study investigated intraventricular repolarization gradient heterogeneity and QT prolongation as an explanation for the paradoxical reduction in QRSTI magnitude in the presence of increased arrhythmia risk.

Methods The 3-dimensional model consisted of approximately 50,000 functional cubic units with spatial resolution of 1.5 mm. Anisotropic electrical conductivity was taken into account for calculating surface potentials (Fig. 1). The fiber orientations rotated counter clockwise 90 ° from epicardium to endocardium. The conduction velocity ratio for longitudinal to transverse fibers was set to 3:1. The surface electrocardiographic (ECG) potentials on a torso model were calculated using the boundary element method (7,8). Table i shows the parameters of electrophysiologic properties and action potential

Table 3. S l o w Conduction Control/Random

Table 2. Normal/Random Model (0.75 m/s) Deviation

(%)

T offset

0

10

20

30

40

50

60

70

80

90 I00

420 441 441 441 441 441 450 462 462 483 483

(ms) 258 279 279 279 279 279 288 300 300 321 321 QTc interval

Model (0.5 m/s) Deviation

(%)

0

I0

20

30

40

50

60

70

80

90 100

T offset 474 468 468 477 474 489 486 489 501 507 510 (ms) 312 306 306 315 312 327 324 327 329 335 338 QTc interval

64

Journal of Electrocardiology Vol. 32 Supplement 1999

~T duration

Fig. 5. QT durations were

360

prolonged in proportion to higher percentage of randomness. Conduction velocities of 0.75 m/s and 0.5 m/s were assigned for control and slow conduction states, respectively. VF could be induced by successive extrastimuli with the combination of over 70% inhomogenous ventriculum and slow conduction velocity.

340 321]

360 340 321

301

IgO

28 2| 2'

80 Io

0.5 Condu Vel~

duration (APD) waveforms of this heart model. Ventricular gradients of 6 ms per layer were assigned, and the difference between endocardium and epicardium was 60 ms (9).

I ! , Control l OTc = 258

Standard 12-lead ECG, body surface potential mapping (BSPM), and recovery parameters were obtained for each assignment of repolarization values. QT duration was measured by the subtraction

i............:.~ .......................} ..............

El ................... i .......................... i -* .......................

i:~

i:i :-t',''

, """

......

. . . . . . .

~i~ -'~ ......................

i'a

~,~ ~ ¢

I ':I~IN:

10% random .....................

............................

:;77

<..

L__J

_2~

l .........

................<,.T._ ..........

I 80%! random I.i ................

l

~.........~<

t ......................... i .......................

i............. i)

I OTc =

.

300

i

i

" ~

split T p ~ i ~ ! s ~o~ ~ mod~

100% random 7 " .........~

[ OTc = 321 Fig. 6. Basic cycle length 1,000 ms for 3-second ECG and QT measurement data, The positions of T offset potential showed a consistent tendency to shift to the left posterior and to disperse with the split at 80% random model on BSPMs.

Paradoxical

of Q onset time f r o m T offset time on the BSPM, w h e r e T offset time was defined as the last potential to be detected at intervals of 3 m s above a threshold of 0.1 mV during recovery (Fig. 2). QRSTI area m a p s w e r e obtained f r o m 160 b o d y surface ECGs. QRSTI m a p s a n d the m e a n value of the m a x i m u m and the m i n i m u m on each m a p w e r e c o m p a r e d b e t w e e n control a n d r a n d o m gradient models. QRSTI contours of 5 mV.ms i n c r e m e n t s are s h o w n w i t h the m a x i m u m and the m i n i m u m indicated by + and - signs, respectively. Basic cycle length f r o m sinus n o d e activation was 1000 ms for 3 beats. For r a n d o m repolarization, a percentage (from 10% to 100%) of n o r m a l l y a r r a n g e d APDs w e r e reassigned, at r a n d o m , according to the table of a r a n d o m n u m b e r s . A r a n d o m APD distribution can be defined with p a r a m e t e r DVT (deviation) which specifies the deviation of APD f r o m its defined value. The APD for a m o d e l cell (i, j, k) is determ i n e d by APD (i, j, k) = APDd + DVT x NR, w h e r e NR is a r a n d o m n u m b e r in the range ( - 1 , 1). DVT

QRST

Integral Changes

•

Okazaki

and Lux

65

is a changing range of APD. The a b o v e formula implies that a r a n d o m APD varies f r o m (APDd DVT) to (APDd + DVT), w h e r e APDd is a defined value setting on this m o d e l of r a n d o m n u m b e r varies b e t w e e n - 1 a n d 1. R a n d o m distribution patterns w e r e d e m o n s t r a t e d as s h o w n in Figure 3.

Results A v e n t r i c u l a r c o n d u c t i o n velocity of 0.75 m / s w a s assigned for the l o n g i t u d i n a l direction for the n o r m a l control ECG. Figure 4 d e m o n s t r a t e s 160lead ECG, s t a n d a r d 12-lead ECG, a n d QRSTI. As the d e v i a t i o n of APD w a s spatially r a n d o m i z e d , t h e r e w a s n o significant c h a n g e of the a c t i v a t i o n process on QRS w a v e f o r m s a n d the BSPM, b u t significant ST-T c h a n g e s w e r e o b s e r v e d b y increased p e r c e n t a g e of d e v i a t i o n of APD on the r e c o v e r y process. QT d u r a t i o n was p r o l o n g e d b y

$8 QT duration

QRSTI Conduction Velocity 0.75 m/sec

7ox

79

,F'V-

?' }7 .........

300

I ..... .F.-:.l:::f:::

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::I 0

~ll!lt

/

60%

~!, t

"F'I'-<"I -"

........ i.,.
~ li,,eiltt l l n m i '

: ~:---; zBe

::":':"<:<;...>,.7 ......

.... ~ . 1 ~

.....~!_...~ .

l~lll-t,v, i l

,"~-

' ~> .~+:~-,

.

.

.

.

.

.

.

.......;

::::~. :t~.::,.

~.oo~

• ~,~

...........:...;. • !~.......... ~

i~ I :1 0

::"- : :~:~

321

>,~. ~-<....................~...........

~ ..............

........

lI:l:i

tq2J t

Fig. 7. QRSTI maps of control and randomized states in proportion to the intraventricular deviation of random. QRSTI maps showed muhipeaks with max 18 mV.ms and abnormal dipolar pattern on the 80% random model (right middle panel). Plus-in-minus phenomenon with abnormal polarity was obtained from area map analysis on the 100% random model (right lower panel).

66

Journal of Electrocardiology Vol. 32 Supplement 1999 |i

Control~~'/~'~

~I~,312

QRSTI

QT duration

Conduction Velocity 0.5 mlsec 20 •

~ . ,

~.-~'~.r'~ ~+

+~=~~==306 o

40 ~

o

........

~++'

+ o .,m

+;:............ 2 ::, ::, :::,:,

,,= 312

...........

,,,,,

80

I................. ::i:=+:~ ......i+ ...i.......

60~

.......

t,~

,

+=6.,324

100X

iiiii

~:~

'

+:o=. 338

0

Fig. 8. QRSTI maps under the slow conduction velocity, 0.5 m/s.

the r a n d o m models compared to the normal model in Table 2. W h e n a conduction velocity of 0.5 rn/s was assigned to the ventricular ceils, QT durations showed greater dispersion for the random models compared to the control in Table 3. The more the distribution of APD deviated, the greater QT duration prolonged as shown in Figure 5. At 100% deviation, the QT interval prolonged 24.4% compared to control. W h e n slow conduction velocity was assigned, QT interval prolonged 31% and an isopotential map demonstrated delayed propagation and a rnultipeak pattern.

Isopotential Maps Figure 6 shows isopotential maps of normal and random models with no changes of the activation process. T offset intervals were prolonged with the progress of intraventricular heterogeneity. The positions of T offset potential showed a consistent tendency to shift to the left posterior. Split T poten-

tials were observed at 462 ms for the 80% random model.

QRSTI Maps A normal dipolar pattern is shown in the lower panel of Figure 4 and the upper panel of Figure 7 with extrema of 38 mV.ms and - 1 4 mV.Ins. As shown in the lower panel of Figure 7, abnormal dipolar QRSTI maps were obtained by the assignm ent of r a n d o m gradient distributions of 60% in contrast to the normal control model with the paradoxical magnitude changes. Multipeaked distributions were observed at 80% random deviation as shown in 80% r a n d o m model of Figure 7. The "plus-in-minus" pattern for the 100% random model is shown in Figure 7. These patterns reflected the i n h o m o g e n e i t y of the ventriclum. Without the slow ventricular conduction velocity assigned, sustained VF could not be induced by 5 successive extrastimuli with 20-ms square pulses under the condition of 0.75 m/s. Thus we as-

Paradoxical QRST Integral Changes • 1 6 Z m.tec

T r a c e s of E C G s a n d B S P M s p r o c e s s

Okazaki and Lux 67

t(6"l ~ e c

[I

vr/ ~

r~e

Oce)

F ~ k 207 ~ •

1~

ms~c

177

~ c

360 m e ~ a

II

L.:].............

.....

..~

-~

~

c

. . . . . . . . . . . . . . . .

......

l.~iL, m~ee

m~c

._

0',*)

C.V. 0.5 m / ~ ~o~re r a n d o m

model

[I

i!I15:m

Toffsa li

70% random

~ _ ~ .... I...................

o

~,,t.~,=, s.ooo .=,.v To

i0

4~

QRSTI

Fig. 9. (A) ECG and BSPMs on the control, 60% and 70% random model (0.5 m/s). (B) Abnormal dipolar pattern on QRSTI with slow conduction on 70% random model. signed 0.5 m / s as a s l o w c o n d u c t i o n velocity to the ventriculum. Figure 8 s h o w s QRSTI maps of n o r m a l dipolar pattern for control w i t h m a x and min of 34 mV-ms and - 1 5 mV-ms, respectively.

In proportion to r a n d o m i z e d gradients, paradoxical m a g n i t u d e changes and abnormal dipolar pattern o n QRSTI were obtained from QT interval m e a s u r e m e n t s by BSPMs. Figures 9A and B s h o w

68

Journal of Electrocardiology Vol. 32 Supplement 1999

VF induction i~i i i::)i L2>.L:i

=::i]]7"~-...............

~. . . . . ~.

2 3 i :~e<: ......

~ iiiiili!!~iii!iiili~%

i

~..~

.{ti# ~.ltl

..

• /,':i[i ili ~ I

]

7!~:T

I

Fr

lii~[~iiii,i]i£iii£]ii!!!~iii]~]iii!]i!~i][]]~ ' .....

ii1:77X,I

~, ~

.......

[ii

,.............7~

i:1

Fig. 10. ECGs and BSPMs on the critical zone of VF induction between 60% and 70% random model.

ECGs, BSPMs, a n d QRSTI m a p s of the a b n o r m a l dipolar p a t t e r n for the 7 0 % r a n d o m m o d e l w i t h e x t r e m a of 36 m V . m s ( m a x ) a n d - 1 5 m V . m s (min).

VF Induction VF could be induced by successive extrastimuli in cases of o v e r 7 0 % dispersed models and 0.5 m / s ventricular conduction velocity. Figure 10 shows ECG traces a n d BSPMs of VF n e a r the critical zone of VF induction b e t w e e n the 6 0 % a n d 7 0 % r a n d o m models. Vulnerability to VF was d e m o n s t r a t e d with the h e t e r o g e n e o u s conditions and a b n o r m a l dipolar pattern.

Discussion In this simulation study, QRSTI m a p s having a b n o r m a l dipolar patterns w i t h paradoxical magnitude changes and increased vulnerability to VF w e r e observed for the i n h o m o g e n e o u s m o d e l in contrast to normal. R a n d o m i z e d repolarization gradients w e r e used as a m o d e l of the h e t e r o g e n e o u s heart in reference to idiopathic VF and long QT s y n d r o m e w i t h o u t structural heart disease (5,6). Originally QT intervals w e r e m e a s u r e d by Bazett's formula, but it was difficult to detect the Q onset

a n d the T offset. QT dispersion on the 12-lead ECG was d e t e r m i n e d for each individual by subtracting the shortest QT interval f r o m the longest QT interval. Values shorter t h a n the m e a n value plus 2 standard deviations c o m p a r e d to control w e r e considered abnormal, but w e did not use this m e t h o d . On the other hand, BSPM can be used to detect the Q onset a n d the T offset; in addition, QRSTI m a p s h a v e value in detecting patients at risk of developing ventricular arrhythmias. QT dispersion relates to i n h o m g e n e i t y of ventricular repolarization and increased vulnerability to cardiac arrhythmias. Electrophysiological properties of QT dispersion are multifactorial and complex (10-13). Our results suggest that VF is not caused by a single specific underlying m e c h a n i s m . Spatial i n h o m o g e n e i t y of refractoriness m a y occur as a c o n s e q u e n c e of collision of excitation fronts a n d facilitate reentry. Simulation studies can verify effects of constituent factors separately. APDs w e r e fixed, and the reduced conduction velocities w e r e assigned. Dyn a m i c changes w o u l d be expected to affect the w h o l e process. Specifically, heart rate variability and conduction velocity are highly i m p o r t a n t in contributing to QT dispersion. I n h o m o g e n e o u s repolarization representing local dispersion of refractoriness m a y be responsible for a local slow conduction (14). The reference voltage is effective for the d e t e r m i n a t i o n of Q onset a n d T offset potential. Zero potential is the s u m of surface potentials at all

Paradoxical QRST Integral Changes points. Surface potential should be very near to the zero potential. The area of QRST is a m e a s u r e of the electrical effects p r o d u c e d by intraventricular heterogeneity in the recovery process.

Conclusions QT prolongation was induced by increasing deviation of APD distributions, showing the i n h o m o g enous recovery process of the heart w i t h o u t changing of activation sequence. The r a n d o m i z e d distribution contributed to paradoxical QRSTI magnitude changes and QT dispersion with a b n o r m a l dipolar patterns. The combination of slow conduction and increased deviation of APD elicited QT dispersion in the ventricle and predicted vulnerability to life-threatening arrhythmia. These results suggest that dispersion of intraventricular recovery properties reflects QT interval o n the QRSTI maps.

References 1. Wilson FN, MacLeod AG, Barker PS, Johnston FD: The determination and significance of the areas of the ventricular deflections of the electrocardiogram. Am Heart J 10:46, 1934 2. Abildskov JA, Evans AK, Lux RL, Burgess M J: Ventricular recovery properties and QRST deflection area in cardiac electrograms. Am J Physiol 239:H227, 1980 3. Kubota I, Lux RL, Burgess M J, Abildskov JA: Relation of cardiac surface QRST distributions to ventricular fibrillation threshold in dogs. Circulation 78:171, 1988

•

Okazaki and Lux

69

4. Gardner MJ, Montague TJ, Armstrong CS, et ah Vulnerability to ventricular arrhythmia: assessment by mapping of body surface potential. Circulation 73:684, 1986 5. De Ambroggi L, Bertoni T, Locati E, et ah Mapping of body surface potentials in patients with the idiopathic long QT syndrome. Circulation 74:1334, 1986 ~. Peeters HA, Sippensgroenewegen A, Wever EF, et ah Electrocardiographic identification of abnormal ventricular depolarization and repolarization in patients with idiopathic ventricular fibrillation. J Am Coll Cardiol 31:1406, 1998 7. Mirvis DM: Spatial variation of QT intervals in normal persons and in patients with acute myocardial infarction. J Am Coll Cardiol 5:625, 1985 8. Wei D, Okazaki O, Harumi K, et al: Comparative simulation of excitation and body surface electrocardiogram with isotropic and anisotropic computer heart models. IEEE Trans Biomed Eng 42:343, 1995 9. Wei D: Whole-heart modeling: progress, principles and applications. Prog Biophys Molec Biol 67:17, 1997 10. Sekiya S, Ichikawa S, Tsutsumi T, Harumi K: Nonuniform action potential durations at different sites in canine left ventricle. Jpn Heart J 24:935, 1983 11. Lux RL, Fuller MS, MacLeod RS, et al: QT interval dispersion: dispersion of ventricular repolarization or dispersion of QT interval? J Electrocardiol 30(suppl): 176, i998 12. Abildskov JA, Lux RL: Distribution of QRST deflection areas in relation to repolarization and arrhythmias. J Electrocardiol 24:197, 1991 13. Lux RL, Green LS, MacLeod RS, Taccardi B: Assessment of spatial and temporal characteristics of ventricular repolarization. J Electrocardiol 27(suppl): 100, 1994 14. de Bakker JMT, van Capelle JL, Janse MJ, et al: Slow conduction in the infarcted human heart: "zigzag" course of activation. Circulation 88:915, 1993