Download Relation of Cardiac Surface QRST Distributions to Ventricular

171
Relation of Cardiac Surface QRST
Distributions to Ventricular Fibrillation
Threshold in Dogs
Isao Kubota, MD, Robert L. Lux, PhD, Mary Jo Burgess, MD, and J.A. Abildskov, MD
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
The relation between ventricular fibrillation threshold (VFT) and cardiac surface QRST area
distributions was studied in eight pentobarbital-anesthetized dogs. Unipolar epicardial electrograms were recorded from 64 sites evenly distributed on the right and left ventricles. Localized
areas of short repolarization properties were produced by directing five intensities of light onto
the surface of the anterior right ventricle through apertures of three sizes. VFT, measured at
the center of the lesion, decreased during warming and had a high negative correlation to the
change (warming-control) in QRST area (AQRST1) in the electrogram recorded from the
center of the lesion. This correlation was independent of lesion size. For the six experiments, the
correlation coefficients for 400-, 800-, and 1,600-mm2 lesions averaged -0.95, -0.94, and
-0.96, respectively. The correlation between VFT and AQRST1 without regard to lesion size
averaged -0.88. VFT also had a negative correlation to root mean square (RMS)AQRST
because of warming. RMSAQRST was calculated from the change in QRST areas
(warming - control) in all 64 electrograms. The correlation between VFT and RMSAQRST was
dependent on lesion size. For all experiments, the correlation between VFT and RMSAQRST
averaged -0.97, -0.93, and -0.93 for 400-, 800-, and 1,600-mm2 lesions, respectively. The
correlation between VFT and RMSAQRST without regard to lesion size, however, was
considerably lower, -0.59. The results of this study provide the first direct evidence that VFT
is correlated with cardiac surface QRST area distributions. (Circulation, 1988;78:171-177)
S everal studies have been reported in which
the relation of QRST area distributions of
body surface maps to arrhythmia vulnerability was evaluated.1-4 The rationale for these studies
was the role of disparity of ventricular repolarization properties in vulnerability to arrhythmias5-10
and the ventricular gradient concept.11-17 Wilson et
all' developed a theoretical analysis supporting
their insight that the QRST area should be a measure of inhomogeneity of ventricular repolarization
properties. The theoretical basis of this hypothesis
was developed further by Burger12 and more recently
by Plonsey,13 Geselowitz,14 and Cohn et al.'5 Direct
evidence that QRST areas of cardiac surface electrograms are related to inhomogeneity of repolariFrom the Nora Eccles Harrison Cardiovascular Research and
Training Institute, University of Utah School of Medicine, Salt
Lake City, Utah.
Supported by awards from the Nora Eccles Treadwell Foundation, the Richard A. and Nora Eccles Harrison Fund for
Cardiovascular Research, and by grants HL-35204 and HL34288 from the National Institutes of Health.
Address for reprints: J.A. Abildskov, MD, Nora Eccles Harrison Cardiovascular Research and Training Institute, University
of Utah, Building 100, Salt Lake City, UT 84112.
Received November 23, 1987; revision accepted March 3, 1988.
zation properties has been obtained from animal
experiments.16,17
Although there are theoretical reasons why QRST
area distributions should be related to arrhythmia
vulnerability18 and although clinicall-3 and experimental4 studies support this relation, there have
been no studies in which arrhythmia vulnerability
was directly measured and related to QRST area
distributions. The purpose of this investigation was
to obtain direct evidence that changes in cardiac
surface QRST area distributions are related to fibrillation threshold.
Materials and Methods
Experimental Preparation
Experiments were performed on eight dogs that
were anesthetized with intravenous sodium pentobarbital 30 mg/kg. Before recording cardiac surface
electrograms and measuring ventricular fibrillation
threshold (VFT), an additional dose of intravenous
sodium pentobarbital, 15 mg/kg, was given. An
intravenous infusion of 500 ml 0.9% NaCI containing 240 mg pentobarbital was also given slowly over
the 3-4 hour duration of each experiment. Large
doses of this anesthetic have been shown previ-
172
Circulation Vol 78, No 1. July 1988
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
ously to reduce spontaneous variations of ventricular repolarization properties, probably by reducing
fluctuations of autonomic nervous tone.16,17
Under artificial ventilation, the heart was exposed
through a midline thoracotomy and supported in a
pericardial cradle. The sinus node was crushed, and
bipolar stimulating electrodes used for delivering
basic driving stimuli were attached to the right
atrium and right ventricle near the pulmonary conus.
Two additional bipolar electrodes were placed on
the anterior right ventricle for delivering VFT test
stimuli. The heart was paced at a cycle length of 400
msec with 2-msec duration square-wave stimuli at
twice diastolic threshold intensity. The driving stimuli were applied to both the right atrium and right
ventricle simultaneously. An array of 64 fine silver
wire unipolar electrodes, insulated, except at their
points of attachment, was mounted with uniform
electrode spacing on a nylon sock that was stretched
over the heart and anchored near the pericardial
reflection. These 64 electrodes were used to record
unipolar electrograms. Each cardiac surface electrode was referenced to a Wilson central terminal.
All 64 electrodes were sampled once every msec,
with a multiplexed data recording system.19 Recordings were taken with amplifiers set at a 0.03-500-Hz
bandwidth. A 12-bit A to D converter that provided
a resolution of approximately 50 gV was used.
Continuous records were sampled for 3 seconds. To
minimize beat-to-beat variations in the electrograms, all recordings were obtained with the respirator held in full expiration. At least 58 electrograms from every experiment were technically
adequate for interpretation. One to six electrograms
were excluded from analysis because of noise or
mechanical artifacts. The animals were kept on a
heated table, and right ventricular subepicardial
temperature was monitored with a thermocouple.
The heart was kept moist with saline.
Ventricular recovery properties were altered by
localized myocardial warming. A light source from
a tungsten-halogen lamp was directed through a
condenser lens assembly and directed onto the
anterior right ventricle through a circular aperture.
The aperture diameter was variable from 50 to 2,400
mm2. This technique has been used previously to
evaluate the relation of localized alterations of
ventricular recovery properties to changes in cardiac surface electrograms. 16, 1720 In two of the eight
experiments, the aperture size was fixed at 800
mm2, and measurements of electrograms and VFT
were repeated alternately during control periods
and periods of local warming. In these two experiments, only one intensity of warming was used. In
the other six experiments, localized alterations of
recovery properties were produced by directing the
light through aperture sizes of 400, 800, and 1,600
mm2. Areas of alteration of recovery properties will
be referred to as "lesions." The order in which the
effects of lesion size were evaluated was varied in
different experiments to minimize any bias that
ANTERIOR
POSTERIOR
RX
l~
FIGURE 1. Diagram of the experimental preparation.
Dots distributed over the heart indicate the 64 unipolar
electrogram recording sites. Arrows that are labeled electrode 1 and electrode 2 indicate locations of bipolar electrodes (not shown) used to measure ventricularfibrillation
threshold. Concentric circles on right ventricle indicate the
400-, 800-, and 1,600-mm2 regions warmed. AO, aorta; RA,
right atrium; LA, left atrium; POST, posterior; RV, right
ventricle; ANT, anterior; LAD, left anterior descending
coronary artery; LV, left ventricle.
might be introduced because of time. Electrograms
were recorded, and VFT was measured during five
intensities of warming of the three lesion sizes. The
heart was warmed for 5 minutes before electrograms were recorded and VFT measured. The animal was allowed to recover for 5 minutes between
each intervention. Control measurements were
obtained before each series of measurements. A
diagram of the experimental preparation is shown in
Figure 1. The anterior wall of the right ventricle was
warmed by a light spot centered at VFT test electrode 1, which was in close proximity to the unipolar recording electrode at the center of the warmed
area. The approximate areas warmed by light spots
with 400-, 800-, and 1,600-mm2 apertures are indicated in the figure. VFT test electrode 2 was in
close proximity to a unipolar recording electrode at
the boundary of the 1,600-mm2 warmed area. The
epicardial temperatures at the center of warmed
areas were measured by a thermocouple, and with
different intensities of light, these ranged from 360
to 420 C.
VFT was measured with gated trains of pulses
delivered to bipolar VFT test electrode 1 or 2. The
VFT test electrodes had interelectrode distances of
2-3 mm. The pulses were 4 msec in duration and
were delivered at 10-msec intervals (100 Hz) after
every 10th driving stimulus. The train was set to
start 80 msec after the driving stimuli and had a
duration of 200 msec. With these settings, the train
started just after the QRS complex and continued
until approximately the end of the T wave in a
vertical-lead electrocardiogram. The strength of the
pulse train was increased in 1-mA increments, and
trains of stimuli at each strength were delivered
Kubota et al
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
twice before the intensity was increased. The least
amount of current required to induce ventricular
fibrillation was taken as the VFT. After fibrillation
was induced, defibrillation was accomplished by
direct-current shock (10-15 W/sec) delivered through
catheter electrodes inserted into the cavities of the
left atrium and left ventricle through the left atrial
appendage and apex of the left ventricle. With this
electrode arrangement, defibrillation was accomplished rapidly, and QRST area distributions during
control periods were stable after as many as 30
episodes of ventricular fibrillation and defibrillation.
The simultaneously recorded electrograms of single complexes were processed with an interactive
computer program. The complexes were digitized;
gain and baseline were adjusted and stored on
digital magnetic tape as previously described.20
QRST deflection areas were calculated by integrating each electrogram over the QRST interval. The
root mean square (RMS) voltage of all 64 leads was
plotted against time, and onset of QRS and end of
T-wave deflections were manually selected from the
RMS curve. The RMS curve is calculated from
deflections in all electrograms and has a more
abrupt onset and termination than the QRST in
individual electrograms. The RMS curve was used
to select onset and termination of the QRST interval
to decrease the likelihood of missing low-amplitude
deflections at the beginning and end of individual
electrograms. Even if there were small errors in
selecting the interval for integration, they would
have negligible effects on the results because
excluded areas would be extremely small in comparison with the changes in QRST area induced by
warming. The right atrium was paced simultaneously with the ventricle, and the P waves were a
small constant value included in QRST areas
recorded during both control and test conditions.
Data Analysis
Data from the six experiments in which multiple
temperatures and three aperture sizes were used
were analyzed with respect to changes in QRST
areas (QRST during warming - control QRST). In
each experiment, change in QRST area in the
electrogram at the center of the lesion (AQRST1)
and the RMS value of change in QRST area in all 64
electrograms (RMSAQRST) were determined. A
total of eight correlation coefficients between VFT
and QRST changes was calculated for each experiment by means of the least-squares method of
linear regression analysis:
1. VFT versus AQRST1 for all intensities of warming of all lesion sizes.
2. VFT versus AQRST1 for all intensities of warming of 400-mm2 lesions.
3. VFT versus AQRST1 for all intensities of warming of 800-mm2 lesions.
4. VFT versus AQRST1 for all intensities of warming of 1,600-mm2 lesions.
O
QRST Area and VF Threshold in Dogs
E
173
0: Control
*: Warming
50
50
0
o
40
5-
°
30
.
20
c.
10
L-1
CWCWCWCWCWCWCW
FIGURE 2. Graph of repeated ventricular fibrillation
thresholds measured at electrode 1 (see Figure 1). Control (C) measurements were alternated with measurements made during warming (W). Warming decreased
fibrillation threshold. There was little variation in repeated
measurements during each state.
5. VFT versus RMSAQRST for all intensities of
warming of all lesion sizes.
6. VFT versus RMSAQRST for all intensities of
warming of 400-mm2 lesions.
7. VFT versus RMSAQRST for all intensities of
warming of 800-mm2 lesions.
8. VFT versus RMSAQRST for all intensities of
warming of 1,600-mm2 lesions.
Results
were
studied
Two dogs
to document that myocardial warming lowers VFT and to demonstrate the
stability of VFT and QRST area after multiple
episodes of warming and fibrillation and defibrillation. In these experiments, VFT measurements
during control periods and during myocardial warming of an 800-mm2 area were alternated. VFT was
measured at two sites. Site 1 was at the center of the
800-mm2 lesion, and site 2 was approximately 5 mm
outside of it (see Figure 1). The intensity of light
was adjusted to raise myocardial temperature to
40-420 C.
The results of one of these experiments are
shown in Figures 2-4. Fibrillation threshold measured at site 1 averaged 44.1 + 3.9 mA (mean + SD)
during control periods and decreased to 4.7 ± 1.6
mA during myocardial warming (p<0.01). VFT
measured at site 2 averaged 45-+±4.2 mA during
control periods and was not significantly different
during warming, 46.73+3.2 mA. The results of the
other experiment were comparable. The findings
confirm previous reports that warming decreases
VETl0 and that VFT is dependent on the site used to
make the measurement.10.21 Figure 2 clearly illustrates that the differences between VFT during
control periods and VFT during warming greatly
exceeded the differences in measurements during a
given state. As illustrated in Figure 3, the QRST
area recorded near VFT test site 1 changed more
during warming than the QRST area recorded near
Circulation Vol 78, No 1, July 1988
174
4000
Site
Control
-
3000
2000
-
1000
-
T /7 ~ T [7
i:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/
_
c~
c
T
W
C
T
1
1
W C WC W CW
Site 2
300
* Control
O Warming
:
cn
a
iI
0bUU ,
400
14000 1
Co
/ ~~~~~~~~~~~
0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Warming
2000 -
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
1 300
-
O-
cl,
-Qnn
-I - I
-4U
C W
raO_CW
W
o_ D_
C
I
W C
W
,*
I
T
C W
T TT
C
W
FIGURE 3. Graphs of QRST areas recorded near ventricularfibrillation threshold test sites 1 and 2 during repeated
control and warming periods. Site 1 was within and site 2
was approximately 5 mm outside of the warmed area.
Data are from same experiment illustrated in Figure 2.
VFT test site 2. The figure also shows that QRST
areas at site 1 were less variable during repeated
recordings in a given state than QRST areas during
control recordings compared with QRST areas during warming. Warming had little effect on activation
sequence, see Figure 4.
In the other six experiments, observations were
made during control periods and during five intensities that warmed the three lesion sizes. Mean
control values of VFT ranged from 15.7 mA to 41.3
mA in the six experiments. However, in individual
experiments, the variability of the control measurements was relatively small, and the maximum difference between measurements ranged from 1 mA
to 9 mA. The differences in repeated control measurements were small with respect to the difference
between VFT during control and warming periods.
The relation of changes in QRST area at site 1
(AQRST1) to VFT of one dog is shown in Figure 5.
The correlation coefficients between VFT and
AQRST1 for lesions of five intensities in 400-, 800-,
and 1,600-mm2 areas were - 0.95, - 0.99, and
-0.96, respectively. The correlation coefficient
between VFT and AQRSTI calculated without regard
to lesion size was -0.94. The results of all six
experiments were comparable and are summarized
in Table 1. Correlation coefficients between VFT
and AQRST1 for lesions of all sizes ranged from
-0.80 to -0.95, mean -0.88. The averages of
correlation coefficients for all six experiments,
FIGURE 4. Diagram of activation sequence maps of the
anterior cardiac surface. Maps were constructed from
data recorded during a control and warming period.
Arrow indicates the right ventricular pacing site. Data
are from same experiment illustrated in Figures 2 and 3.
All isochrone lines are at 5-msec intervals.
between VFT and AQRST, for 400-, 800-, and
1,600-mm2 lesions, were - 0.95, - 0.94, and - 0.96,
respectively. In a previous study,20 we showed that
QRST area at the center of a lesion is highly
correlated with lesion severity and is independent
of lesion size. The results of the present study show
there is also a high correlation between VFT and
AQRSTI, that is, lesion severity, which is independent of lesion size. As shown in Figure 3, there was
TABLE 1. Correlation Coefficient Between Ventricular Fibrillation
Threshold and AQRST
Dog
1
2
3
4
5
6
Mean
All
-0.95
-0.86
-0.80
-0.85
-0.94
- 0.88
- 0.88
Lesion size
400 mm2
800 mm2
-0.97
-0.92
-0.97
- 0.97
-0.90
-0.91
-0.99
-0.95
-0.95
-0.99
-0.94
-0.88
- 0.95
-0.94
1,600 mm2
-0.98
-0.95
-0.93
-0.94
-0.96
-0.99
-0.96
Kubota et al
QRST Area and VF Threshold in Dogs
175
30S
E
*
400 mm2
*
800 mm 2
10
0
(U
*
*
0
20[
1600 mm2
r=-0.94
*
400 mm2
*
800
*
1600 mm2
A
0
mm
2
c
0
c
0
0
An
'U
0
U
0
10a
-
C
(U
.U
a
11
0
r_
£
-#
L7 FY1
200
600
1000
0
\
C
,S~~~~~
1400
0
AORST, (mV.ms)
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
FIGURE 5. Graph of relation between ventricular fibrillation threshold and change in QRST area at electrode 1
(AQRST1) duringfive intensities ofwarming of400-, 800-,
and 1,600-mm2 areas (dog 5). Correlation was independent of lesion size.
little difference in QRST area recorded at the periphery of the warmed area during control and warming
periods. Therefore, VFT measured at site 1 was not
correlated with QRST area recorded at the periphery of the warmed area.
Unlike the relation of VFT to AQRSTI, which
was relatively independent of lesion size, the relation of VFT to RMSAQRST was dependent on
lesion size. As shown in Figure 6, there was a linear
relation between VFT and RMSAQRST for each
lesion size. Also shown in the figure, for a given
RMSAQRST, VFT was lower for smaller lesions
than for larger lesions. These data are from the
same experiment illustrated in Figure 5. Data from
all six experiments are summarized in Table 2.
Correlation coefficients between VFT and
RMSAQRST for lesions of a given size were between
- 0.75 and - 0.99 and averaged - 0.97, - 0.93, and
- 0.93 for 400-, 800-, and 1,600-mm2 lesions, respectively. The correlation coefficients calculated without regard to lesion size ranged from -0.35 to
-0.84 and averaged -0.59. RMSAQRST, which
was calculated from all 64 electrograms, reflects
TABLE 2. Correlation Coefficient Between Ventricular Fibrillation
Threshold and RMSAQRST
Lesion size
800 mm
Total
400 mm2
Dog
1,600 mm2
1
-0.77
-0.97
-0.94
-0.95
2
-0.38
-0.97
-0.98
-0.96
3
-0.35
-0.97
-0.94
-0.78
4
-0.56
-0.97
-0.93
-0.75
5
-0.66
-0.95
-0.97
-0.96
-0.99
-0.84
-0.99
-0.96
6
-0.93
-0.59
-0.97
-0.93
Mean
RMS, root mean square.
100
200
300
400
RMSAQRST (mV.ms)
FIGURE 6. Graph of relation between ventricularfibrillation threshold and RMSAQRST during five intensities
of warming of 400-, 800-, and 1,600-mm2 areas (dog 5).
Correlation coefficients were -0.95, -0.96, and -0.97
for 400-, 800-, and 1,600-mm2 lesions, respectively. Correlation coefficient calculated without regard to lesion
size was considerably lower, -0.66. Regression line for
400-mm2 lesions was shifted to the left of regression line
for 1,600-mm2 lesions.
lesion size as well as lesion severity. In all six
experiments, the regression line for VFT versus
RMSAQRST for 400-mm2 lesions was shifted to the
left of the regression line for VFT versus RMS
AQRST for 1,600-mm2 lesions. We have previously
shown that gradients of QRST area between warmed
and surrounding areas increase as the intensity of
myocardial warming increases but that the locations
where QRST area changes occur are relatively
unaffected by the intensity of myocardial warming.20
The results therefore suggest that lesion severity
plays a major role in arrhythmia vulnerability.
Discussion
A relation between QRST area distribution and
VFT is suggested by previous studies5-8 that documented that agents known to increase arrhythmia
vulnerability increased disparity of ventricular repolarization and by the ventricular gradient hypothesis proposed by Wilson et al. 1 Their hypothesis
stated that the QRST area was a measure of inhomogeneity of ventricular repolarization. Later studies on experimental animals provided direct evidence that the QRST deflection areas of cardiac
surface electrograms do in fact reflect disparity of
repolarization. 16,17 Methods of analyzing QRST area
distributions and relating these distributions to
arrhythmia vulnerability have been evaluated in
both an animal study4 and in clinical studies.1-3 In
the animal study,4 interventions known to enhance
arrhythmia vulnerability such as digitalis intoxica-
176
Circulation Vol 78, No 1, July 1988
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
tion and hypothermia were used, but vulnerability
to arrhythmias was not directly measured. In the
clinical studies,'-3 body surface QRST area distributions of patients at risk of ventricular tachyarrhythmias were analyzed. As in the animal study, there
was no direct measure of arrhythmia vulnerability.
The results of both the animal and clinical studies
suggested that analysis of QRST area distributions
provides prognostic information regarding arrhythmia vulnerability. The present study, however, provides direct evidence that VFT is correlated with
QRST distributions of cardiac surface electrograms. VFT is a commonly used method of quantifying arrhythmia vulnerability in animal experiments. Although the exact physiological basis for
the measurement has not been defined, VFT
decreases in conditions associated with spontaneous fibrillation such as coronary occlusion. 10 A
relation between VFT and occurrence of spontaneous arrhythmias in the clinical setting seems reasonable but remains speculative.
In these experiments, the lesions were localized
areas of abnormally short repolarization properties
induced by warming the anterior surface of the right
ventricle. Localized warming has previously been
reported to lower fibrillation threshold10 and enhance
reentrant activity.22 Kuo et al,9 however, were
unable to induce ventricular tachyarrhythmias by
programmed stimulation in dogs during regional
coronary artery perfusion with warm blood unless
the animals were systemically hypothermic. The
reason for the differences between their findings
and our findings is uncertain. Our method of warming the myocardium produces relatively sharp borders between warmed areas with short repolarization properties and surrounding areas with normal
repolarization properties.20 On the other hand, it is
likely that regional perfusion of warm blood results
in more gradual transitions in repolarization properties transmurally and at the boundaries of the
perfusion field than our model and that the gradient
of properties may be inadequate to increase ventricular vulnerability. In addition, the methods for
evaluating vulnerability were different in our study
and in the study by Kuo et al.
Increased disparity of ventricular repolarization
properties was the likely mechanism for the decrease
in VFT during myocardial warming. Warming shortens ventricular repolarization properties,9,10,20 and
localized shortening of repolarization properties
would be expected to increase T-wave amplitude in
regionally sensitive electrocardiographic leads and
alter QRST area distributions. In our experiments,
VFT had a high inverse correlation with the alterations in QRST areas because of lesions with short
repolarization properties. Although we did not assess
the correlation between VFT and QRST alterations
due to localized lesions with prolonged repolarization properties, it seems likely there would also be
a high correlation between VFT and QRST alterations in this state. However, the correlation would
be a positive one rather than a negative one because
localized prolongation of repolarization properties
due to cooling would decrease both T-wave amplitude and VFT.
Local disparity of repolarization properties, however, is not the only factor that plays a role in
arrhythmia vulnerability, and other factors affecting
arrhythmogenesis may not be detectable in QRST
area distributions. It seems unlikely that arrhythmia
vulnerability due to enhanced automaticity would
be detected in maps of QRST area distribution.
Slow conduction, which also plays a role in arrhythmia vulnerability, may have small effects on QRST
waveform. Recent studies from our laboratory23
have shown that conduction delays electrotonically
modulate repolarization properties. Whether or not
the modulation of repolarization properties due to
conduction delays is sufficient to alter QRST area
distributions remains to be investigated.
Despite these limitations, the results of the present study document an excellent inverse correlation
between VFT and change in QRST area in the
electrogram at the center of a localized lesion with
short repolarization properties. QRST area progressively increased and VFT progressively decreased
as the severity of the lesion was increased. The
correlation between VFT and change in QRST at
the center of the lesion was relatively independent
of lesion size (see Figure 5 and Table 1). For a given
change in QRST area at the center of the lesion,
fibrillation threshold was the same regardless of
whether the lesion was large or small. This is in
keeping with findings of a previous study from our
laboratory that showed a high correlation between
change in QRST area and lesion severity that was
independent of lesion size.20 Increased severity of
lesions increased the magnitude of QRST change
recorded from sites within the light spot but did not
increase the number of sites affected.
The correlation between VFT and RMSAQRST,
however, was dependent on lesion size (see Figure
6 and Table 2). For lesions of a given size and
different severities, the correlation coefficient was
between - 0.75 and - 0.99. Correlation coefficients
calculated for VFT and RMSAQRST for lesions of
all sizes and all severities were between - 0.35 and
- 0.84. For a given RMSAQRST, fibrillation threshold was lower for small lesions than for large
lesions. A small lesion would have to have greater
abnormalities in repolarization properties than a
large lesion to result in the same RMSAQRST.
Therefore, the dependence of the relation of VFT to
RMSAQRST on lesion size suggests that lesion
severity plays a major role in arrhythmia vulnerability. Abildskov and Steinhaus's findings24 from
computer simulations also support this conclusion.
They found that small severe lesions resulted in
striking decreases in fibrillation threshold. Fibrillation threshold, however, did not decrease further
when lesion size was increased and when lesion
severity was kept the same. With less severe lesions,
Kubota et al
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
lesion size had only a small effect on VFT, and a
10-fold increase in lesion size decreased VFT by
only 13%.
The results of our study provide direct evidence
that decreased VFT due to local disparity of repolarization properties is correlated with cardiac surface QRST area distributions. Additional studies
are needed to establish the correlation between
VFT and QRST area distribution alterations induced
by other abnormalities. We did not examine the
relation of VFT to body surface QRST area distributions. However, analysis of regionally sensitive
electrocardiographic leads provides reasonably accurate estimates of localized cardiac abnormalities.
Previous work from our laboratory has shown that
activation and recovery times estimated from the
minimum derivative of the QRS and maximum
derivative of the T wave of cardiac surface electrograms are highly correlated with the times estimated from distant electrocardiographic leads.25
We have also shown previously that localized repolarization abnormalities associated with acute coronary occlusion are clearly evident in body surface
as well as cardiac surface QRST distributions in the
presence of simulated left bundle branch block.26
Additional experimental and clinical studies to evaluate the relation of cardiac and body surface QRST
area distributions to arrhythmia vulnerability therefore seem warranted.
References
1. Tsunakawa H, Nishiyama G, Kanesaka S, Harumi K: Assessment of susceptivity for ventricular tachycardia following
myocardial infarction by non-dipolarity of QRST area maps
(abstract). Circulation 1987;74(suppl IV):IV-432
2. Gardner MJ, Montague TJ, Armstrong CS, Horacek BM,
Smith ER: Vulnerability to ventricular arrhythmia: Assessment by mapping of body surface potential. Circulation
1986;73:684-692
3. Ambroggi LD, Bertoni T, Locati E, Stramba-Badiale M,
Schwartz PH: Mapping of body surface potentials in patients
with the idiopathic long QT syndrome. Circulation 1986;
74:1334-1345
4. Urie PM, Burgess MJ, Lux RL, Wyatt RF, Abildskov JA:
The electrocardiographic recognition of cardiac states at
high risk of ventricular arrhythmias: An experimental study
in dogs. Circ Res 1978;42:350-358
5. Han J, Moe GK: Nonuniform recovery of excitability in
ventricular muscle. Circulation 1964;14:44-60
6. Han J, Garcia de Jalon PD, Moe GK: Adrenergic effects on
ventricular vulnerability. Circ Res 1964;14:516-524
7. Han J, Garcia de Jalon PD, Moe GK: Fibrillation threshold
of premature ventricular responses. Circ Res 1966;18:18-25
QRST Area and VF Threshold in Dogs
177
8. Merx W, Yoon MS, Han J: The role of local disparity in
conduction and recovery time on ventricular vulnerability to
fibrillation. Am Heart J 1977;94:603-610
9. Kuo CS, Munakata K, Reddy CP, Surawicz B: Characteristics and possible mechanism of ventricular arrhythmia
dependent on the dispersion of action potential durations.
Circulation 1983;67:1356-1367
10. Burgess MJ, Williams D, Ershler P: Influence of test site on
ventricular fibrillation threshold. Am Heart J 1977;94:55-61
11. 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 1934;
10:46-61
12. Burger HC: A theoretical elucidation of the notion "ventricular gradient." Am Heart J 1957;53:240-246
13. Plonsey R: A contemporary view of the ventricular gradient
of Wilson. J Electrocardiol 1979;12:337-341
14. Geselowitz DB: The ventricular gradient revisited. Relation
to the area under the action potential. IEEE Trans Biomed
Eng 1983;30:76-77
15. Cohn RL, Rush S, Lepeschkin E: Theoretical analysis and
computer simulation of ECG. Ventricular gradient and recovery waveforms. IEEE Trans Biomed Eng 1982;29:413-423
16. Abildskov JA, Evans AK, Lux RL, Burgess MJ: Ventricular
recovery properties and QRST deflection area in cardiac
electrograms. Am J Physiol 1980;239:H227-H231
17. Abildskov JA, Green LS, Evans AK, Lux RL: The QRST
deflection area of electrograms during global alterations of
ventricular repolarization. J Electrocardiol 1982;15:103-108
18. Abildskov JA, Green LS, Lux RL: The present status of
body surface mapping. J Am Coll Cardiol 1983;2:394-396
19. Ershler PR, Steadman BW, Wyatt RF, Lux RL, Abildskov
JA: A 64 lead online system for clinical and experimental
mapping of ventricular activation sequences. Proceedings of
Computers in Cardiology. Park City, Utah, September 14-18,
1984, pp 417-419
20. Burgess MJ, Lux RL, Wyatt RF, Abildskov JA: The relation
of localized myocardial warming to changes in cardiac
surface electrograms in dogs. Circ Res 1978;43:899-907
21. Shumway N, Johnson J, Stesh R: The study of ventricular
fibrillation by threshold determinations. J Thoracic Surg
1957;34:643-653
22. Geddes JS, Burgess MJ, Millar K, Abildskov JA: Accelerated repolarization as a factor in reentry: Simulation of the
electrophysiology of acute myocardial infarction. Am Heart
J 1974;88:61-68
23. Burgess MJ, Steinhaus BM, Spitzer KW, Ershler PR: Nonuniform epicardial activation and repolarization properties
of the in vivo canine pulmonary conus. Circ Res 1988;
62:233-246
24. Abildskov JA, Steinhaus BM: Effect of lesion size on
vulnerability (abstract). J Am Coll Cardiol 1987;9(suppl
2): 127A
25. Burgess MJ, Baruffi S, Spaggiari S, Macchi E, Lux R,
Taccardi B: Determination of activation and recovery
sequences and local repolarization durations from distant
electrocardiographic leads. Jpn Heart J 1986;27(suppl
10):205-216
26. Hirai M, Burgess MJ, Haws CW: Effects of coronary
occlusion on cardiac and body surface and QRST isoarea
maps of dogs with abnormal activation simulating left bundle
branch block. Circulation 1988;77 (in press)
KEY WoRDs * arrhythmia vulnerability * isoarea maps
cardiac surface maps
Relation of cardiac surface QRST distributions to ventricular fibrillation threshold in
dogs.
I Kubota, R L Lux, M J Burgess and J A Abildskov
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Circulation. 1988;78:171-177
doi: 10.1161/01.CIR.78.1.171
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1988 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/78/1/171
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/