Locally Propagated Activation Immediately After
Internal Defibrillation
Nipon Chattipakorn, MD; Bruce H. KenKnight, PhD; Jack M. Rogers, PhD; Robert G. Walker, BA;
Gregory P. Walcott, MD; Dennis L. Rollins, MS;
William M. Smith, PhD; Raymond E. Ideker, MD, PhD
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Background—Electrical mapping studies indicate an interval of 40 to 100 ms between a defibrillation shock and the
earliest activation that propagates globally over the ventricles (globally propagated activation, GPA). This study
determined whether activation occurs during this interval but propagates only locally before being blocked (locally
propagated activation, LPA).
Methods and Results—In five anesthetized pigs, the heart was exposed and a 504-electrode sock with 4-mm interelectrode
spacing was pulled over the ventricles. Ten biphasic shocks of a strength near the defibrillation threshold (DFT) were
delivered via intracardiac catheter electrodes, and epicardial activation sequences were mapped before and after
attempted defibrillation. Local activation was defined as dV/dt #20.5 V/s. Postshock activation times and wave-front
interaction patterns were determined with an animated display of dV/dt at each electrode in a computer representation
of the ventricular epicardium. LPAs were observed after 40 of the 50 shocks. A total of 173 LPA regions were observed,
each of which involved 262 (mean6SD) electrodes. LPAs were observed after both successful and failed shocks but
occurred earlier (P,.0001) after failed (3568 ms) than successful (41616 ms) shocks, although the times at which the
GPA appeared were not significantly different. On reaching the LPA region, the GPA front either propagated through
it (n5135) or was blocked (n538). The time from the onset of the LPA until the GPA front propagated to reach the
LPA region was shorter (P,.01) when the GPA front was blocked (32612 ms) than when it propagated through the
LPA region (63620 ms).
Conclusions—LPAs exist after successful and failed shocks near the DFT. Thus, the time from the shock to the GPA is
not totally electrically silent. (Circulation. 1998;97:1401-1410.)
Key Words: defibrillation n mapping n waves
E
lectrical mapping studies of defibrillation shocks slightly
lower than the DFT have established the presence of an
interval of '40 to 100 ms between the shock and the earliest
GPA.1– 4 Because it was assumed that electrical activity was
absent during this postshock interval at the GPA origin, the
interval was described as the isoelectric window.4 Using
optical mapping techniques, Kwaku and Dillon5 found that an
isoelectric window does not exist but that activation immediately begins to propagate away from the borders of regions
that are excited directly by the shock field. Both findings can
be accommodated if activation occurs during the isoelectric
window, but this activation can propagate only locally within
a small region before being blocked because of postshock
tissue refractoriness. To test this hypothesis, we mapped the
activation sequences on the epicardium of both ventricles
after shocks of a strength near the DFT to determine whether
locally propagated activation (LPA) occurs.
The results demonstrate that the so-called isoelectric window is not truly electrically silent. LPAs, which occur on the
epicardium soon after the shocks, propagate locally and then
block before another activation, the GPA, appears and propagates across the entire heart.
Methods
Animal Preparation
Five healthy pigs (30 to 35 kg) were tranquilized with acepromazine
(1.1 mg/kg) and ketamine (22 mg/kg), anesthetized with sodium
pentobarbital (30 mg/kg initially; 0.05 mg z kg21 z min21 maintenance
dose), intubated, mechanically ventilated with supplemental oxygen,
and given intravenous fluids. The ECG and arterial blood pressure
were monitored continuously. Core body temperature, arterial blood
gas values, and electrolyte levels were maintained within the normal
range throughout the experiment. The chest was opened through a
median sternotomy, and the heart was suspended in a pericardial
cradle.
Received August 4, 1997; revision received November 4, 1997; accepted November 6, 1997.
From the Cardiac Rhythm Management Laboratory, Division of Cardiovascular Diseases, Department of Medicine (R.G.W., G.P.W., D.L.R., W.M.S.,
R.E.I.), Department of Physiology and Biophysics (N.C.), and Department of Biomedical Engineering (J.M.R.), University of Alabama at Birmingham;
and Department of Therapy Research, CPI, St Paul, Minn, a Division of Guidant Corp (B.H.K.).
Correspondence to Raymond E. Ideker, MD, PhD, Cardiac Rhythm Management Laboratory, 1670 University Blvd, Volker Hall, B140, Birmingham,
AL 35294-0019.
E-mail [email protected]
© 1998 American Heart Association, Inc.
1401
1402
Locally Propagated Activation After Defibrillation
DFT
GPA
LPA
LV
LVFA
RV
SVC
VF
Selected Abbreviations and Acronyms
5 defibrillation threshold
5 globally propagated activation
5 locally propagated activation
5 left ventricular, left ventricle
5 last VF activation before shock
5 right ventricular, right ventricle
5 superior vena cava
5 ventricular fibrillation
Electrode Placement
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Catheter-mounted, platinum-coated titanium coil electrodes (CPIGuidant Corp) were used for defibrillation. A 34-mm catheter
(surface area, 390 mm2) was inserted into the right external jugular
vein and advanced to the RV apex and served as cathode for the first
phase of the biphasic shock. A 68-mm catheter (surface area,
780 mm2) was inserted into the left external jugular vein, and its
distal tip was positioned at the junction of the SVC and right atrium.
The position of the catheters was verified with fluoroscopy (Fig 1A),
and then the catheters were secured with a ligature at the venotomy
site to stabilize their positions. Bipolar pacing electrodes (Ag/AgCl,
1 mm in diameter, with 2- to 3-mm interpolar spacing) sutured to the
RV outflow tract, LV apex, and anterobasal LV free wall were used
to pace the heart so that the mapping array could be oriented. The
bipolar electrode at the RV outflow tract also was used as the
fibrillating electrode.
A 504-electrode array constructed from an elastic sock was
used to record unipolar epicardial electrograms. There were 14
rows of electrodes with 6, 14, 22, 26, 32, 34, 38, 40, 44, 46, 48,
50, 50, and 54 Ag/AgCl2 1-mm-diameter electrodes per row from
apex to base of the ventricles. Electrodes were spaced '4 mm
from center to center. For data analysis, the sock was divided
conceptually into apical (7 rows), middle (4 rows), and basal (3
rows) regions (Fig 1B and 1C). The sock was pulled until its
center contacted the LV apex and its base extended above the
atrioventricular groove (Fig 1B). Two 3-mm-diameter Ag/AgCl
disk reference electrodes were sutured to the aortic root 5 mm
apart. One served as the reference for unipolar recordings and one
as the ground electrode for the mapping system.
Data Acquisition
The 504 epicardial unipolar electrogram signals referenced to the
aortic root were simultaneously band-pass filtered between 0.5 and
500 Hz and recorded digitally with a 528-channel computer-assisted
cardiac mapping system.6,7 The procedure of recording events before,
during, and after attempted defibrillation via electrical shock was as
follows. At 10 ms before the beginning of the shock, an external
timing device signaled the mapping system microprocessor8 to
switch the attenuators on, change the amplifier coupling from AC to
DC, and decrease the gain of each channel. Approximately 5 ms after
the end of the shock, the external timing device signaled the
microprocessor to switch the attenuators off. During the next 4 ms,
the attenuators finished switching, the amplifier coupling was
changed to AC, and the gain of each channel was increased to the
preshock value. The signals were recorded digitally at a rate of 2000
samples per second per channel, together with real-time information
indicating the sequence of events during the recording.9 Positioning
of the return and ground electrodes to the aortic root allowed
electrograms to be observed quickly after the shock.10
Defibrillation Protocol
Figure 1. Diagram representing defibrillation electrode configuration, mapping array, and shock waveform. A, Defibrillation
electrodes were positioned at SVC and RV apex. Electrode at
RV apex was cathode and electrode at SVC was anode for first
phase of biphasic waveform. B, A 504-electrode elastic sock
was pulled over heart. For data analysis, ventricles were conceptually divided into (I) apical, (II) middle, and (III) basal thirds.
C, Polar projection of sock was used in animated computer display of dV/dt at each electrode site (squares). D, Biphasic waveform had interphase delay of 0.2 ms. Ao indicates aorta; LAD,
left anterior descending coronary artery; PA, pulmonary artery;
RA, right atrium; and RVOT, right ventricular outflow tract.
Brief runs of normal sinus rhythm and of pacing from each of the
three bipolar pacing electrodes and from the tip of the defibrillation
catheter at the RV apex were recorded at the beginning of each
experiment to determine the orientation of the mapping array. VF
was then electrically induced by a 60-Hz alternating current delivered via the bipolar pacing electrode at the RV outflow tract. After
10 seconds of VF, defibrillation was attempted with 10-ms biphasic
truncated exponential shocks generated by a 150-mF defibrillator
(Ventritex HVS-02). Biphasic shocks were 6 ms in duration for the
first phase and 4 ms in duration for the second phase, with a 0.2-ms
interphase delay. A single-capacitor discharge was emulated by
adjustment of the leading-edge voltage of phase 2 to approximately
the same level as the trailing-edge voltage of phase 1; the overall tilt
varied with system impedance11 (Fig 1D). The delivered voltage and
current were displayed on the monitor of a waveform analyzer
(DATA 6100, Analogic Inc). Total delivered energy was calculated
by the waveform analyzer.12
To obtain an approximately equal number of successes and
failures at the same shock energy, DFTs were determined with a
modified three-reversal up/down protocol.13,14 The initial shock
strength, 400 V, was chosen because it was near the expected 50%
probability of success. The leading-edge voltage was then decreased
or increased in 80-V steps after a successful or failed defibrillation
attempt, respectively, until a first reversal in defibrillation outcome
was achieved. At each reversal point, the sign of the voltage change
of the algorithm was reversed; after the first reversal, the voltage step
was changed to 40 V, and after the second reversal, it was changed
to 20 V. The successful shock strength that was part of the pair of
shocks forming the third reversal was defined as the DFT.
VF was induced for 662 episodes (mean6SD) in each animal to
determine the DFT. Next, shocks of the DFT strength were given
after 10 seconds of VF during 10 different VF episodes in each
animal. After unsuccessful shocks, defibrillation was obtained within
Chattipakorn et al
10 seconds with a higher-energy shock delivered through the same
electrode pair. Only the 10 shocks at the DFT strength were used for
analysis. A minimum of 4 minutes was allowed to elapse between
VF episodes. At the end of the study, the anesthetized animal was
euthanized by fibrillation.
Terminology
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Earliest GPA is the first activation (dV/dt #20.5 V/s) after the
shock that gives rise to an activation front that propagates across
almost the entire epicardium. The GPA was determined from the
dV/dt animation as described previously.15 The GPA epicardial site
of origin (GPA origin) is the epicardial site at which the GPA is
first detected.
LPAs are activations (dV/dt #20.5 V/s) that occur before
activation occurs at the GPA origin. They do not spread to activate
the entire epicardium but rather propagate only locally before
blocking and becoming extinguished. An LPA region is a group of
spatially adjacent electrode sites at which LPAs are recorded. LPA
sites are the individual recording electrode sites within the LPA
region that record LPAs.
The postshock interval is the interval between the beginning of the
shock and the first postshock activation at an electrode site. The
postshock interval at the GPA origin has been called the isoelectric
window in previous studies.1– 4 The preshock interval is the interval
between the last activation before the shock and the beginning of the
shock at an electrode site.
Data Analysis
The mapping data were transferred to a computer workstation (Sun
Microsystem, Inc), where all activations were analyzed by visualization of an animation of the first derivative (dV/dt) of the unipolar
electrograms.16,17 The dV/dt was determined over a 2-ms sliding
window spanning five data points.18 Any electrode with a dV/dt
#20.5 V/s was defined as active.19 This method of analysis has
several advantages over isochronal mapping20,21 and has been used
previously to observe wave-front propagation and the presence of
collision or reentry.15 Beginning 20 ms after the leading edge of the
shock, ie, 10 ms after the end of the shock, activations were
determined in 0.5-ms steps until the GPA front had traversed the
epicardium. The first 10 ms after the end of the shock was not
examined because of the restrictions of the switched mapping system
hardware.
Determination of LPA
For an event to be described as an LPA, it was required that the
activation at the site be dV/dt #20.5 V/s, that it occur at least 20 ms
after the leading edge of the shock but before the GPA arose, and that
it be excluded from noise by comparison of the display of the
unipolar electrogram to its derivative (dV/dt). The numbers of LPA
sites and LPA regions and the minimum dV/dt at each LPA site and
at the GPA origin were determined for all defibrillation episodes.
Determination of Preshock and Postshock Interval
and VF Cycle Length
The preshock and postshock intervals were determined at all LPA
sites and GPA origins for each defibrillation episode from the
animated computer display. Mean VF cycle length was determined at
all LPA sites and GPA origins in all episodes. At each LPA site and
GPA origin, the time of each activation, identified as the time at
which dV/dt was a minimum and was also #20.5 V/s, was
determined for 10 VF cycles before the shock. The average of the 10
activation intervals was used to define the mean VF cycle length at
that site.
Determination of the Wave-Front Interaction Time
Wave-front interaction time at an LPA region was defined as the
interval between the time of activation at the LPA site closest to the
approaching GPA wave front in each region and the time at which
the GPA wave front propagated to an electrode adjacent to that LPA
site. Wave-front interaction time was determined for all LPA regions
April 14, 1998
1403
for each defibrillation episode. In some cases, an interaction between
the LPA and the GPA front could be observed: propagation of the
GPA wave front blocked in the LPA region. Block was considered to
be present if the interval between the activation criteria being first
met at adjacent electrodes was .40 ms, implying a conduction
velocity of ,0.1 m/s, which is thought to be unphysiological.22–24
Statistical Analysis
Comparison of data between successful and unsuccessful defibrillation episodes was performed with Student’s t test for paired and
unpaired data. Spatial distributions of LPA sites and GPA origins
were analyzed by one-way ANOVA. When statistical significance
was found, individual comparisons were carried out with Fisher’s
post hoc test. Values are shown as the mean6SD. Differences were
considered to be significant for P#.05.
Results
Of the 50 defibrillation attempts in the five pigs, 23 were
successful and 27 were not. The DFT was 4246132 V (range,
280 to 640 V). Heart weight was 220630 g.
Existence of LPAs
LPAs were seen in 40 of the 50 defibrillation attempts and
were associated with successful defibrillation in 18 episodes.
A total of 173 LPA regions (65 for successful episodes and
108 for failed episodes) and 339 LPA sites (155 for successful
episodes and 184 for failed episodes) were observed in the
five animals.
Recordings from one electrode in one animal after two
different shocks are shown in Fig 2. In Fig 2A, an LPA is seen
with a GPA. The mean minimum dV/dts for the LVFA, the
LPA, and the GPA for this event were 21.1, 21.7, and 20.9
V/s, respectively. No LPA occurs in Fig 2B, with the mean
minimum dV/dt for the LVFA and the GPA being 21.2 and
21.9 V/s, respectively. Fig 3 shows propagation of LPAs.
The four regions seen in frame 2 all become blocked and
extinguished by frame 8. Examples of recordings in and
around the LPA region from Fig 3 at which the GPA front
blocked are given in Fig 4. Tracings a through e are from
LPA sites (arrows), and tracings 1 through 5 are from
electrodes surrounding but just outside the LPA region. The
minimum dV/dts for the LPAs in electrograms a through e
and for the GPA in electrograms 1 through 5 were
21.2460.47 and 23.3260.68 V/s, respectively. The GPA in
tracings 1 through 5 does not activate the LPA region at
which electrograms a through e were recorded, although it
does cause a small (possibly electronic) deflection. The
minimum dV/dt for the small deflections after LPAs in
electrograms a through e was 20.3960.08 V/s.
LPAs were significantly more common in the apical and
middle thirds of both ventricles (Table 1A and Fig 5A),
whereas GPAs were significantly more common in the apical
third, and mostly in the LV (Table 1A and Fig 5B). The
minimum dV/dt for the GPA origins was 23.262.0 V/s,
significantly less than the minimum dV/dt for the LPAs
(21.6061.64 V/s).
Interaction of the GPA Front With the
LPA Region
The GPA front blocked when it reached 38 of the 173 LPA
regions. The GPA front did not pass through this area that
1404
Locally Propagated Activation After Defibrillation
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
of LPA sites and regions was not significantly different
between successful and unsuccessful shocks. However, the
mean time until the LPAs occurred was longer for successful
than for failed shocks (Table 2A). The time from earliest
activation within an LPA region and the activation time at the
GPA origin was 26611 ms for the 65 LPA regions during
successful defibrillation episodes and was 2969 ms for the
108 LPA regions during unsuccessful episodes (P#.05). This
interval for successful and failed episodes combined was
27610 ms. Because there was no significant difference for
the postshock interval at the GPA origin between successful
and failed shocks (Table 2A), the time interval between the
LPA and the GPA origin was longer for unsuccessful than for
successful episodes. Thus, the time at which GPA wave
fronts encountered the LPA region was usually later for failed
shocks. The incidence of block of the GPA front in the LPA
region was not significantly different for successful and failed
shocks (Table 2B). In about one half of all episodes with
LPAs, the GPA front propagated through one or more LPA
regions but blocked in one or more other LPA regions.
Comparison of Preshock Intervals at the LPA
Sites and the GPA Origins
The preshock interval at LPA sites and GPA origins was
normalized by dividing by the mean VF cycle length at that
site. Preshock intervals at LPA sites were longer (P,.01)
than at GPA origins (Tables 2A and 1B). Minimum dV/dt for
activation during VF immediately preceding the shock at the
GPA origins (22.161.1 V/s) was less than at the LPA sites
(21.761.2 V/s) (P,.01).
Discussion
Figure 2. Unipolar epicardial electrograms from two episodes of
unsuccessful defibrillation. A, Recording at one LPA site (top)
shows VF (left), defibrillation shock (multiple vertical lines), and
signals after shock (right). Derivative of electrogram (bottom) is
also shown. Arrow labeled S indicates beginning of shock. LPA
was detected 50 ms after shock (arrow labeled LPA). GPA origin
was first detected 67 ms after leading edge of shock at a different electrode. GPA wave front reached this LPA site 80 ms after
shock (arrow labeled GPA), 30 ms after LPA at this site. LVFA at
this recording was detected 70 ms before shock (arrow labeled
LVFA). B, Recording (top) and its derivative (bottom) at same
electrode as in A but during a different defibrillation episode in
same animal. In this episode, no LPA was detected. Only GPA
(arrow) was detected when GPA front reached this site.
was activated earlier by the LPA but rather moved around it
and then propagated over the remainder of the epicardium
(Fig 3). In the other episodes, the GPA wave front propagated
(sometimes slowly) through the LPA region (Fig 6). In all
cases, the LPA activation had blocked and disappeared before
the GPA activation front arrived. The wave-front interaction
time for those episodes in which the GPA front was blocked
was 32612 ms, which was significantly shorter (P,.0001)
than the interval in those episodes in which the GPA wave
front passed through the LPA region (63620 ms).
Relationship Between LPA and GPA for
Successful and Unsuccessful Defibrillation
On average, LPAs occurred significantly earlier than GPAs
for both successful and failed shocks (Table 2A). The number
This study shows that after shocks near the DFT, small
activation fronts (LPAs) quickly appeared that propagated for
only a short distance before blocking and becoming extinguished. These LPAs appeared 27610 ms before the epicardial appearance of the first activation front that propagated
across almost the entire epicardium after the shock, ie, the
GPA front. Thus, the time interval between the shock and the
appearance of the GPA front, which previously has been
called the isoelectric window,3,4 was not electrically silent.
LPAs were observed after most (80%) defibrillation shocks
of near DFT strength, irrespective of their outcome. Both the
incidence and distribution of LPAs differed from those of
GPAs. LPAs were registered at only '2% of electrode sites
during a shock episode. LPA sites were more generally
distributed, appearing frequently in the apical and middle
thirds of both ventricles, whereas GPA activation fronts arose
primarily from the apical third of the left ventricle. Approximately 20% of GPA fronts blocked when they reached an
LPA region. When GPA block occurred, the time from the
onset of the LPA until the GPA front reached the LPA region
was significantly shorter (32612 ms) than when GPA block
did not occur (63620 ms).
Presence of LPAs
Several lines of evidence differentiate LPAs from shock
artifacts in the recordings. LPAs were observed only occasionally and were not observed repeatedly at the same
Chattipakorn et al
April 14, 1998
1405
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Figure 3. Selected frames of animation of activation after unsuccessful shock (380 V) showing block of GPA front in LPA region. Panels are polar views of ventricles with orientation of electrodes as shown in Fig 1C. Forty successive frames at 2-ms intervals are shown
beginning 48 ms after leading edge of shock. Each black dot represents local activation at 1 of 504 electrodes, signifying dV/dt #20.5
V/s for this frame. Grey dots are electrodes that are inactive for this frame, ie, dV/dt .20.5 V/s. There are 4 LPA regions (within circles
in frame 2) and 20 LPA sites in this episode. GPA was detected 66 ms after leading edge of shock (arrow in frame 10) near LV apex.
GPA wave front blocked without propagating through one LPA region (circle with arrow in frame 2). White dots from frames 8 through
40 indicate LPA region in which this block occurred. Wave-front interaction time in this case is 30 ms (from frame 1 to frame 16).
electrode (Fig 2). After different shocks, LPAs appeared in
many different electrodes dispersed over most of the ventricular epicardium and frequently occurred in clusters. The
strongest evidence that the LPAs are not artifactual is that
they altered refractoriness of the tissue in the vicinity of the
LPA. This is indicated by blocking of the GPA activation
front when it encountered an LPA region (Fig 3) in some
cases and by slowing of the propagation velocity in others
(Fig 6). When GPA block occurred, the time after the LPA at
which the GPA activation front reached the LPA region was
32612 ms, which is consistent with this region still being in
its absolute refractory period. When the GPA front slowly
propagated through the LPA region, however, this time
period was 63620 ms, which is consistent with the region
being in its relatively refractory period.25,26
Origin of LPAs
GPA fronts after shocks near the DFT do not represent unaltered
continuation of activation fronts present just before the shock.2
The same appears to be true for LPAs in that the sum of the
preshock and postshock intervals for LPAs, 114661 ms (Table
1B), was significantly longer than the mean activation rate at the
same electrodes during the previous 10 activations during VF
just before the shock (83612 ms).
Although it cannot be definitely ruled out that LPAs arise
from foci triggered by the shock, it is likely that LPAs are part
of an activation front that arises and propagates away from
the border of a region that is excited directly by the shock, as
has been demonstrated for GPA fronts by optical mapping.5
The possibility that early sites of postshock activation could
be caused by graded responses that slowly propagate for a
distance and then trigger activation from a less refractory
site27 may be pertinent to the origin of the GPA, because the
preshock interval suggests that tissue at the GPA origins was
partially refractory at the time of the shock. Slow propagation
of the graded response could have been missed in the
extracellular recordings and could account for the postshock
interval at GPA or LPA sites.
LPAs were observed 38613 ms after the shock, which is
not as early as would be expected if the LPAs arose from an
1406
Locally Propagated Activation After Defibrillation
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Figure 4. Unipolar electrograms from LPA region and surrounding electrodes. Diagram (top left) is from frame 3 in Fig 3. Diagram on
right is enlargement of LPA region within circle. Black squares are LPA sites. Gray squares are electrodes surrounding LPA region.
Electrograms a through e on left are from electrodes recording an LPA, also labeled a through e in top right diagram. Arrows indicate
LPA. Electrograms 1 through 5 on right are from electrodes surrounding LPA region that did not record an LPA, also labeled 1 through
5 in top right diagram. Arrows in electrograms 1 through 5 indicate GPA downslopes.
activation front that propagated away from a directly excited
region (Table 1B). The conduction velocity may be slowed
because the activation fronts are propagating through relatively refractory tissue in which the refractory period has
TABLE 1.
been extended by the shock.28 –31 Indeed, some conduction
slowing is evident near the boundary of the directly excited
region in the figures shown in the study by Kwaku and
Dillon,5 in which most shocks were much weaker than the
Spatial Distribution and Preshock and Postshock Intervals of LPA Sites and GPA Origins
LPA Site, n (%)
GPA Origin, n (%)
Apical third
159 (47%)†
24 (60%)†‡
Middle third
145 (43%)†
12 (30%)
Basal third
35 (10%)
4 (10%)
A. Spatial distribution of LPA sites and GPA origins*
Right ventricle
206 (61%)
4 (10%)
Left ventricle
133 (39%)
36 (90%)§
B. Preshock and postshock intervals at LPA sites and GPA origins for all shock episodes
Preshock intervals, ms
76662
52620\
Postshock intervals, ms
38613
61610\
114661¶
113624¶
Preshock 1 postshock intervals, ms
VF cycle length, ms
83612
81612
Preshock intervals, % VF cycle length
92673
65625\
*The sites are classified into apical, middle, and basal thirds (see Fig 1B). The total number of those sites is also divided into the left and right
ventricles. Values in parentheses show the percentage of the total number of LPA sites or GPA origins.
†P,.05 vs basal third, ‡P,.05 vs middle third (one-way ANOVA, Fisher’s post hoc analysis), §P,.05 vs right ventricle (Student’s t test); \P,.001
vs LPA, ¶P,.001 vs VF cycle length.
Chattipakorn et al
April 14, 1998
1407
Figure 5. Polar projection maps of LPA
sites and GPA origins for all shock episodes in one animal. A, Distribution of
LPA sites (stars) for successful defibrillation episodes. Circles represent LPA
sites for failed episodes. B, Distribution
of GPA origins (stars) for successful defibrillation episodes. Circles represent
GPA origins for failed episodes. Abbreviations as in Fig 1.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
DFT. Such slowing might have been exacerbated in our
study, in which shocks were closer to the DFT, resulting in
more tissue being excited directly by the shock and the
generation of the boundaries of the directly excited region in
more highly refractory tissue. An alternative explanation is
that the activation fronts giving rise to LPAs originated from
the endocardium or midwall and the postshock interval
represents the time required for these activation fronts to
reach the epicardium.
The site at which direct excitation occurs and LPAs arise
should be determined by the changes in the transmembrane
potential caused by the shock, which can be influenced by the
Figure 6. Example in which GPA front propagates slowly through LPA region. Interval between consecutive maps is 4 ms. First panel
is 26 ms after leading edge of shock (280 V). The only LPA region in this unsuccessful episode was detected 30 ms after leading edge
of shock. GPA was detected 62 ms after shock (arrow) at LV apex. GPA activation front propagated through LPA region. GPA wave
front propagated slowly in LPA region relative to surrounding area, causing wave front to arc around edges of LPA region and converge on other side. Wave-front interaction time is 88 ms (from frame 2 to frame 24).
1408
Locally Propagated Activation After Defibrillation
TABLE 2. Characteristics of LPA Sites and GPA Origins in Relationship to Success or Failure
of Defibrillation
Successful Shocks
Unsuccessful Shocks
A. No. of LPA sites and regions and postshock and preshock intervals at LPA sites and GPA origins*
No. of LPA sites for each shock
966
No. of LPA regions for each shock
462
563
Postshock interval at LPA sites, ms
41616
3568†
Postshock interval at GPA origins, ms
865
64610‡
6269‡
Preshock interval at LPA site, ms§
84672% (72663)
99674% (81661)
Preshock interval at GPA origin, ms§
60623%\ (49621)
69627%\ (55620)
B. No. (%) of defibrillation episodes in which the GPA front did or did not block in an LPA region¶
Block in all LPA regions
3 (17)
Block in no LPA region
6 (33)
1 (4)
9 (41)
Block in some LPA regions but not others
9 (50)
12 (55)
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
*Data are for the 40 shock episodes with LPA sites.
†P,.0001 vs successful shocks, ‡P,.0001 vs postshock interval at LPA sites, §mean6SD (values in parentheses are the actual
[not normalized] preshock interval), \P,.01 vs at LPA sites; ¶values in parentheses show the percentage of the total number of
successful or unsuccessful defibrillation episodes with LPAs.
shock potential gradient field, interruptions in the myofiber
syncytium,32 and the orientation and curvature of the myofibers.33–35 Because LPAs occurred almost anywhere over both
ventricles, they appeared in portions of the heart exposed to
both high- and medium-strength shock potential gradient
fields as well as in regions exposed to the weaker shock
potential gradient field (Fig 5A). For the electrode configuration used in this study, the shock potential field is largest in
the right ventricle adjacent to the shocking electrode and falls
off with distance, being weakest in the lateral apical left
ventricular free wall.36 In contrast, most GPA activation
fronts arose in the apical third and mostly from the left
ventricle, the region in which the shock field was weakest
(Fig 5B). Thus, although LPAs could be observed even in
regions of high potential gradients, these regions were not
excited directly by the shock, even though the cells were in a
late stage of their refractory period and hence excitable.
Recent findings that interruption of the myofiber syncytium giving rise to secondary sources32 and the orientation
and curvature of myofibers with respect to the shock field
affect the transmembrane potential33–35 raise the possibility
that, even in regions of high potential gradient, the transmembrane potential may not be changed markedly during the
shock. Even though they were excitable, such regions would
not be excited directly by the shock. This may explain why
shocks did not excite LPA regions directly, even though in
many cases, these regions should have been excitable, because their mean preshock intervals were 92% of the activation rate at these sites during VF (Table 1B), suggesting that
tissue at these sites was not absolutely refractory.
Hyperpolarized and depolarized regions are interspersed
throughout the heart during a defibrillation shock.32,37,38 Roth
and Wikswo34 and Wikswo et al35 demonstrated that “make”
excitation occurs in depolarized but not hyperpolarized regions in myocardium that is recovered, whereas hyperpolarization and not depolarization leads to “break” excitation in
myocardium that is relatively refractory. Hence, depending
on the stage of refractoriness, in regions in which the
transmembrane potential is markedly altered by the shock,
direct excitation should occur in some tissue areas but not
others, depending on whether depolarization or hyperpolarization is present.
Comparison of GPA and LPA Wave Fronts
As opposed to GPA wave fronts, the pathways of LPA fronts
are small, typically traversing an area covered by only a few
electrodes, and terminate soon after they appear on the
epicardium. LPA fronts are probably extinguished by conduction block when they encounter adjacent tissue that is
refractory because of lack of recovery from its last activation
before the shock, prolongation of its refractory period by the
shock, or direct excitation by the shock.
In contrast, GPAs appear later after the shock, arise
primarily in the region exposed to the lowest shock potential
gradient, and propagate to activate all of the epicardium
except for some LPA regions in which they may be blocked.
Because they occur later after the shock than the LPAs, the
myocardium has had more time to recover and is less
refractory, which may explain why GPA fronts do not also
block and become extinguished. In addition, the preshock
interval is shorter, so the tissue at the GPA origin is more
refractory at the time of the shock and the mean minimum
dV/dt is more negative, ie, the downslope is faster. Although
the postshock interval was longer at the GPA origin and the
tissue had more time to recover from the shock, the sum of
the preshock and postshock intervals was not different. Thus,
the time from the last activation before the shock to the first
activation after the shock was not significantly different for
sites of GPA origin and sites of LPAs. The dV/dt for the VF
activation just before the shock was also more negative at the
sites of GPA origin.
LPAs and Defibrillation
The association of the outcome of defibrillation with the
length of the so-called isoelectric window is controversial.
Although some studies indicate that this interval is longer for
Chattipakorn et al
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
successful than for failed shocks3,4 and suggest that it may be
used to predict the outcome of defibrillation, other studies do
not indicate an association.2,15 In the present study, we found
no difference in the so-called isoelectric interval for successful and failed shocks. This discrepancy may reflect methodological differences. Most studies that showed differences in
the isoelectric window between successful and failed shocks
used a monophasic waveform in dogs together with shocks of
different strengths such that the mean shock strength for the
failed shocks was lower than for successful shocks.3,4 Zhou et
al,2 using both monophasic and biphasic shocks in dogs but
comparing successful and failed defibrillation shocks of the
same strength, found no differences in the so-called isoelectric window. The study by Usui et al,15 which also found no
difference in the so-called isoelectric interval for successful
and failed shocks, was similar to the present study in that they
used a biphasic waveform in pigs, an RV-SVC defibrillation
electrode configuration, and successful and failed shocks of
the same strength. A greater shock strength may be the
common cause of two effects that may be independent of each
other: an increase in the success rate of defibrillation and an
increase in the duration of the so-called isoelectric window. It
is possible that LPAs were recorded in previous studies1– 4 but
that they were not noted because LPAs are fleeting, occupy
small regions, appear close to the shock artifacts, and are
quickly overshadowed by the large GPA wave front that
passes across most of the epicardium.
Dillon has suggested that shocks defibrillate by causing a
uniform degree of refractoriness throughout the ventricles.39
However, the results of our study suggest that a uniform
degree of refractoriness is not absolutely necessary for
successful defibrillation. The LPAs, occurring 38613 ms
after the shock, should have caused the LPA regions to be
more refractory than the remainder of the myocardium after
the LPAs occurred. Yet defibrillation was still successful
even in the presence of LPAs. In fact, the postshock interval
at the LPA sites was significantly longer for successful than
unsuccessful shock episodes, suggesting that the difference in
the time of recovery between LPA regions and the remainder
of the ventricles was even greater for successful than for
unsuccessful shock episodes.
Although the postshock interval at the GPA origins was not
different between successful and failed shocks, LPAs occurred
earlier after failed shocks than after successful shocks (Table
2A). However, LPAs alone did not reinitiate VF, because LPAs
existed immediately after shocks that ultimately succeeded as
well as after shocks that failed to defibrillate. It is not known
whether the interactions between LPA and GPA fronts play a
causative role in the outcome of defibrillation or whether the
small but significant differences in mean times of LPAs after
successful and unsuccessful shocks is a concomitant effect of
some other causative mechanism for defibrillation. The answer
to this question will require additional studies in which recording
electrodes are spaced more closely than in this study to allow
determination of whether activation fronts arise in the LPA
region after the GPA front and lead to the reinitiation of VF.
Limitations of the Study
The limitations of the study concern evaluation of the LPAs
and are primarily inherent in electrical mapping. Because
April 14, 1998
1409
transmural recording was not performed and the epicardial
electrodes were '4 mm apart, actual postshock windows for
the LPAs and GPAs were probably shorter than reported in
this study. The potential gradient field was not estimated. We
could not observe activations for 10 ms after the end of the
shock, nor could we tell which regions were excited directly
by the shock or directly visualize the action potentials in the
LPA regions or in the surrounding tissue where the LPA is
blocked. Conduction velocity was not measured because,
unlike for an electrode plaque, the interelectrode spacings
within the sock were not constant. Because these limitations
can be overcome by optical mapping, such a study is needed
to confirm our initial observation of LPAs and to learn more
about how and why they occur.
Acknowledgments
This study was supported in part by National Institutes of Health
research grants HL-42760 and HL-33637. The authors wish to thank
Frederick G. Evans, Catherine M. Sreenan, and Fiona Hunter for
their statistical analyses and helpful editorial comments.
References
1. Chen P-S, Wolf PD, Melnick SD, Danieley ND, Smith WM, Ideker RE.
Comparison of activation during ventricular fibrillation and following
unsuccessful defibrillation shocks in open chest dogs. Circ Res. 1990;66:
1544 –1560.
2. Zhou X, Daubert JP, Wolf PD, Smith WM, Ideker RE. Epicardial
mapping of ventricular defibrillation with monophasic and biphasic
shocks in dogs. Circ Res. 1993;72:145–160.
3. Shibata N, Chen P-S, Dixon EG, Wolf PD, Danieley ND, Smith WM,
Ideker RE. Epicardial activation following unsuccessful defibrillation
shocks in dogs. Am J Physiol. 1988;255:H902–H909.
4. Chen P-S, Shibata N, Dixon EG, Wolf PD, Danieley ND, Sweeney MB,
Smith WM, Ideker RE. Activation during ventricular defibrillation in
open-chest dogs: evidence of complete cessation and regeneration of
ventricular fibrillation after unsuccessful shocks. J Clin Invest. 1986;77:
810 – 823.
5. Kwaku KF, Dillon SM. Shock-induced depolarization of refractory myocardium prevents wave-front propagation in defibrillation. Circ Res.
1996;79:957–973.
6. Wolf PD, Rollins DL, Blitchington TF, Ideker RE, Smith WM. Design for
a 512 channel cardiac mapping system. In: Mikulecky DC, Clarke AM,
eds. Biomedical Engineering: Opening New Doors. Proceedings of the
Fall 1990 Annual Meeting of the Biomedical Engineering Society. New
York, NY: New York University Press; 1990:5–13.
7. Smith WM, Wolf PD, Simpson E, Danieley ND, Ideker RE. Mapping
ventricular fibrillation and defibrillation. In: Shenasa M, Borggrefe M,
Breithardt G, eds. Cardiac Mapping. Mount Kisco, NY: Futura Publishing Co; 1993:251–260.
8. Wolf PD, Wharton JM, Wilkinson CD, Smith WM, Ideker RE. A method
of measuring cardiac defibrillation potentials. In: Proceedings of the 39th
Annual Conference on Engineering in Medicine and Biology. Washington, DC: Alliance for Engineering in Medicine and Biology; 1986:4.
9. Wolf PD, Ideker RE, Smith WM. A data stream formatter for a cardiac
mapping system. In: Kondraske GV, Robinson CJ, eds. Proceedings of
the 8th Annual International Conference of the IEEE Engineering in
Medicine and Biology Society. Fort Worth, TX: Institute of Electrical and
Electronics Engineers, Inc; 1986:491– 493.
10. Witkowski FX, Penkoske PA, Plonsey R. Mechanism of cardiac defibrillation in open-chest dogs with unipolar DC-coupled simultaneous activation and shock potential recordings. Circulation. 1990;82:244 –260.
11. KenKnight B, Eyüboglu M, Ideker RE. Impedance to defibrillation countershock: does an optimal impedance exist? Pacing Clin Electrophysiol.
1995:2068 –2087.
12. Feeser SA, Tang ASL, Kavanagh KM, Rollins DL, Smith WM, Wolf PD,
Ideker RE. Strength-duration and probability of success curves for defibrillation with biphasic waveforms. Circulation. 1990;82:2128 –2141.
13. Topham SL, Cha Y-M, Peters BB, Chen P-S. Effects of lidocaine on
relation between defibrillation threshold and upper limit of vulnerability
in open-chest dogs. Circulation. 1992;85:1146 –1151.
1410
Locally Propagated Activation After Defibrillation
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
14. Huang J, KenKnight BH, Walcott GP, Walker RG, Smith WM, Ideker
RE. Effect of electrode polarity on internal defibrillation with monophasic
and biphasic waveforms using an endocardial lead system. J Cardiovasc
Electrophysiol. 1997;8:161–171.
15. Usui M, Callihan RL, Walker RG, Walcott GP, Rollins DL, Wolf PD,
Smith WM, Ideker RE. Epicardial sock mapping following monophasic
and biphasic shocks of equal voltage with an endocardial lead system.
J Cardiovasc Electrophysiol. 1996;7:322–334.
16. Laxer C, Ideker RE, Smith WM, Wolf PD, Simpson EV. A graphical
display system for animating mapped cardiac potentials. In: Proceedings
of the 3rd IEEE Symposium on Computer-Based Medical Systems. Chapel
Hill, NC: IEEE Computer Society; 1990:197–204.
17. Parson I, Mendler P, Downar E. On-line cardiac mapping: an analog
approach using video and multiplexing techniques. Am J Physiol. 1982;
242:H526 –H535.
18. Cabo C, Wharton JM, Wolf PD, Ideker RE, Smith WM. Activation in
unipolar cardiac electrograms: a frequency analysis. IEEE Trans Biomed
Eng. 1990;37:500 –508.
19. Bollacker KD, Simpson EV, Hillsley RE, Blanchard SM, Gerstle RJ,
Walcott GP, Callihan RC, King MC, Smith WM, Ideker RE. An
automated technique for identification and analysis of activation fronts in
a 2-D electrogram array. Comput Biomed Res. 1994;27:229 –244.
20. Berbari EJ, Lander P, Scherlag BJ, Lazzara R, Geselowitz DB. Ambiguities of epicardial mapping. J Electrocardiol. 1991;24(suppl):16 –20.
21. KenKnight BH, Bayly PV, Gerstle RJ, Rollins DL, Wolf PD, Smith WM,
Ideker RE. Regional capture of fibrillating ventricular myocardium:
evidence of an excitable gap. Circ Res. 1995;77:849 – 855.
22. Scher AM, Young AC. Spread of excitation during premature ventricular
systoles. Circ Res. 1955;3:535–542.
23. Sano T, Takayama N, Shimamoto T. Directional difference of conduction
velocity in the cardiac ventricular syncytium studied by microelectrodes.
Circ Res. 1959;7:262–267.
24. Myerburg RJ, Gelband H, Nilsson K, Castellanos A, Morales AR, Bassett
AL. The role of canine superficial ventricular muscle fibers in endocardial
impulse distribution. Circ Res. 1978;42:27–35.
25. Zhou X, Wolf PD, Rollins DL, Afework Y, Smith WM, Ideker RE.
Effects of monophasic and biphasic shocks on action potentials during
ventricular fibrillation in dogs. Circ Res. 1993;73:325–334.
26. Opthof T, Ramdat Misier AR, Coronel R, Vermeulen JT, Verberne HJ,
Frank RGJ, Moulijn AC, van Capelle FJL, Janse MJ. Dispersion of
refractoriness in canine ventricular myocardium: effects of sympathetic
stimulation. Circ Res. 1991;68:1204 –1215.
27. Gotoh M, Uchida T, Mandel WJ, Fishbein MC, Chen P-S, Karaguezian
HS. Cellular graded responses and ventricular vulnerability to reentry by
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
a premature stimulus in isolated canine ventricle. Circulation. 1997;95:
2141–2154.
Sweeney RJ, Gill RM, Reid PR. Characterization of refractory period
extension by transcardiac shock. Circulation. 1991;83:2057–2066.
Dillon SM. Optical recordings in the rabbit heart show that defibrillation
strength shocks prolong the duration of depolarization and the refractory
period. Circ Res. 1991;69:842– 856.
Jones JL, Jones RE, Milne KB. Refractory period prolongation by
biphasic defibrillator waveforms is associated with enhanced sodium
current in a computer model of the ventricular action potential. IEEE
Trans Biomed Eng. 1994;41:60 – 68.
Knisley SB, Smith WM, Ideker RE. Effect of field stimulation on cellular
repolarization in rabbit myocardium: implications for reentry induction.
Circ Res. 1992;70:707–715.
Gillis AM, Fast VG, Rohr S, Kléber AG. Spatial changes in transmembrane potential during extracellular electrical shocks in cultured
monolayers of neonatal rat ventricular myocytes. Circ Res. 1996;79:
676 – 690.
Tung L, Neunlist M, Sobie EA. Near-field and far-field stimulation of
cardiac muscle. In: Cerullo LJ, Heiferman KS, Liu H, Podbielska H, Wist
AO, Zamorano LJ, eds. Clinical Applications of Modern Imaging Technology II. Los Angeles, Calif: The International Society for Optical
Engineering; 1994:367–374.
Roth BJ, Wikswo JP Jr. Electrical stimulation of cardiac tissue: a
bidomain model with active membrane properties. IEEE Trans Biomed
Eng. 1994;41:232–240.
Wikswo JP Jr, Lin S-F, Abbas RA. Virtual electrodes in cardiac tissue: a
common mechanism for anodal and cathodal stimulation. Biophys J.
1995;69:2195–2210.
Tang ASL, Wolf PD, Afework Y, Smith WM, Ideker RE. Threedimensional potential gradient fields generated by intracardiac catheter
and cutaneous patch electrodes. Circulation. 1992;85:1857–1864.
Clark DM, Rogers JM, Ideker RE, Knisley SB. Intracardiac defibrillationstrength shocks produce large regions of hyperpolarization and depolarization. J Am Coll Cardiol. 1996;27:147A. Abstract.
Efimov IR, Cheng YN, Biermann M, Van Wagoner DR, Mazgalev TN.
Transmembrane voltage changes produced by real and virtual electrodes
during monophasic defibrillation shock delivered by an implantable
electrode. J Cardiovasc Electrophysiol. 1997;8:1031–1045.
Dillon SM. Synchronized repolarization after defibrillation shocks: a
possible component of the defibrillation process demonstrated by optical
recordings in rabbit heart. Circulation. 1992;85:1865–1878.
Locally Propagated Activation Immediately After Internal Defibrillation
Nipon Chattipakorn, Bruce H. KenKnight, Jack M. Rogers, Robert G. Walker, Gregory P.
Walcott, Dennis L. Rollins, William M. Smith and Raymond E. Ideker
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Circulation. 1998;97:1401-1410
doi: 10.1161/01.CIR.97.14.1401
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1998 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/97/14/1401
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/