2128
Strength-Duration and Probability of
Success Curves for Defibrillation
With Biphasic Waveforms
Scott A. Feeser, MD, Anthony S.L. Tang, MD.
Katherine M. Kavanagh. MD, Dennis L. Rollins, MS, William M. Smith, PhD,
Patrick D. Wolf, MS. and Raymond E. Ideker, MD, PhD
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Certain biphasic waveforms require less energy to defibrillate than do monophasic pulses of
equal duration, although the mechanisms of this increased effectiveness remain unclear. This
study used strength-duration and percent success curves for defibrillation with monophasic
and biphasic truncated exponential waveforms to explore these mechanisms. In part 1,
defibrillation thresholds were determined for both high- and low-tilt waveforms. The monophasic pulses tested ranged in duration from 1.0 to 20.0 msec, and the biphasic waveforms had first
phases of either 3.5 or 7.0 msec and second phases ranging from 1.0 to 20.0 msec. In part 2,
defibrillation percent success curves were constructed for 6.0 msec/6.0 msec biphasic waveforms
with a constant phase-one amplitude and with phase-two amplitudes of approximately 21%,
62%, 94%, and 141% of phase one. This study shows that if the first phase of a biphasic
waveform is held constant and the second phase is increased in either duration or amplitude,
defibrillation efficacy first improves, then declines, and then again improves. For pulse
durations of at least 14 msec, the second-phase defibrillation threshold voltage of a high-tilt
biphasic waveform is higher than that of a monophasic pulse equal in duration to the biphasic
second phase (p<0.05), indicating that the previously proposed hypothesis of stimulation by
the second phase is not the sole mechanism of biphasic defibrillation. These facts indicate the
importance of the degree of tilt for the defibrillation efficacy of biphasic waveforms and suggest
at least two mechanisms exist for defibrillation with these waveforms, one that is more effective
for smaller second phases and another that becomes more effective as the second phase is
increased. (Circulation 1990;82:2128-2141)
C ertain biphasic waveforms require less energy to defibrillate or produce less cardiac
dysfunction than do monophasic pulses of
equal duration.1-4 The lower defibrillation threshold (DFT) for certain biphasic waveforms has important implications for the effectiveness and
lifespan of automated, implantable cardioverter/
defibrillators with power supplies that are constrained by space considerations.5-7
Research seeks to answer the following two closely
related questions. First, what is the mechanism that
From the Departments of Medicine (S.A.F., A.S.L.T., K.M.K.,
W.M.S., D.L.R., R.E.I.) and Pathology (R.E.I.), Duke University
Medical Center, and the School of Engineering (W.M.S., P.D.W.),
Duke University, Durham, N.C.
Supported in part by National Institutes of Health research
grants HL-42760, HL-28429, HL-44066, and HL-40092, and by
National Science Foundation Engineering Research Center grant
CDR862201.
Address for correspondence: Raymond E. Ideker, Box 3140,
Duke University Medical Center, Durham, NC 27710.
Received October 18, 1989; revision accepted July 3, 1990.
makes biphasic shocks so effective? Several explanations have been offered. Jones et a18 suggested that the
first phase acts as a hyperpolarizing "conditioning
prepulse" that activates sodium channels, and the
second phase acts as a depolarizing "excitation pulse."
Other possible reasons are 1) that the first phase
may shorten cardiac cell refractoriness, enabling
more cells to be excited by the second phase, 2) that
reduced cardiac impedance during the second
phase permits greater current delivery during that
phase, and 3) that the biphasic waveform is less able
to stimulate myocardium to reinitiate fibrillation
after the shock.9
Information regarding the first question may help
answer the second question. Which are the optimum
biphasic waveforms for defibrillating, based on energy
requirements,'3,4,9-13 postshock cardiac function,2'4,15
or other criteria?8 '6 For exponential decay waveforms,
several authors have shown that the second-phase
duration must be less than or equal to the first-phase
Feeser et al Success Curves for Defibrillation
duration to achieve a minimum DFI.3,49 Others have
shown the importance of the relative amplitudes of the
two phases,1-'3 the pulse separation,17 or the electrode
orientations.18'19
This study uses strength-duration curves and percent success curves for defibrillation with monophasic
and biphasic truncated exponential waveforms to explore further the mechanism of defibrillation by biphasic waveforms while also providing information on
which waveforms require minimum energy for defibrillation. The study shows that two separate mechanisms
of defibrillation must exist to explain the behavior of
biphasic waveforms with lower energy second phases
versus those with higher energy second phases.
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Methods
This study comprised five similar but distinct protocols, described subsequently as part la-d and part 2.
Part 1 protocols determined defibrillation strengthduration curves, and the part 2 protocol measured
percent success curves. The common portions of each
protocol are described first.
Surgical Preparation
Fifty-two mongrel dogs (22.3 ±3.3 kg body wt) were
anesthetized with pentobarbital (30-35 mg/kg). The
dose of pentobarbital was adjusted according to the
depth of anesthesia assessed by signs such as shivering,
eyelid reflexes, and pedal reflexes.20 Each dog was
intubated with a cuffed endotracheal tube and ventilated with 30-60% oxygen through a Harvard respirator (Harvard Apparatus, South Natick, Mass.). Succinylcholine was initially given at a dose of 1 mg/kg and
later at 0.25-0.5 mg/kg no more than once per hour to
decrease muscle contraction induced by the electric
shocks. This dose is much less than that required to
alter cardiac excitability.21 A femoral arterial line was
inserted to continuously monitor systemic blood pressure. Lactated Ringer's was continuously infused. Normal metabolic status was maintained throughout the
study by taking blood every 30-60 minutes, determining the pH, Po2, Pco2, base excess, CO2 and HCO3
contents, and calcium, potassium, and sodium concentrations, and correcting any abnormal values.
The chest was opened by median sternotomy, and
the pericardial sac opened to expose the heart. A
33-cm2 contoured mesh electrode sutured onto the
right ventricular epicardium served as the anode and
a 39-cm2 electrode sutured onto the left ventricle was
the cathode, as previously described.3
Protocol
Each dog underwent up to 90 fibrillation/defibrillation sequences performed at 5-minute intervals.
For each sequence, fibrillation was induced by using
a 35 V, 60-Hz pulse for 1-2 seconds delivered
between pacing wires inserted into the right ventricle. Defibrillation was attempted after 10 seconds by
applying to the epicardial electrodes one of the
truncated exponential waveforms described herein
(Figures 1-3). If this initial attempt failed, a rescue
2129
shock with the minimum reliable defibrillation energy was applied between these same electrodes. A
Ventritex HVS 150 gF defibrillator (Ventritex,
Sunnyvale, Calif.) delivered all shocks in parts la, lb,
and 2. A custom 750-,uF defibrillator was used during
parts lc and ld to achieve a longer decay time
constant. In part ld, the time constant was further
lengthened by wiring the defibrillator in series with a
200-fl resistor. During each attempted defibrillation,
the applied current and voltage across the heart were
sampled at 20 kHz by a DATA 6000 waveform
analyzer (Data Precision, Danvers, Mass.), and the
impedance between the electrodes and the total
delivered energy of each phase of the pulse were
computed. To ensure proper equipment functioning
and to stabilize the ionic and hormonal milieu of the
heart, each study was begun by conducting three
conditioning fibrillation/defibrillation sequences.
(A)
7.0
-
1
10.5
-
L_14.0
1
~~~17.5
(B)
(C)
(D)
FIGURE 1. Waveforms tested in parts la and lb. Tilts
ranged from 10% to 90% based on phase duration. Panel A:
Six monophasic waveforms used in both parts la and lb.
Panel B: Eight one-capacitor biphasic waveforns used in part
la, with the phase-two leading edge voltage (V21) equal to the
phase-one trailing edge voltage (V,,). Panel C: Eight onecapacitor biphasic waveforms used in part lb. Panel D: Eight
two-capacitor biphasic waveforms used in part lb, with V2,
equal to the phase-one leading edge voltage (V,,).
Circulation Vol 82, No 6, December 1990
2130
(A)
vi'
Waveform A
V21N1 1
20%
r 50%
Waveform B k
67%
K 83%
Waveform C
100%
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Waveform D
140%
FIGURE 3. Waveforms used in part 2. Tilt for each phase
averaged 55%. For waveforms B and C, the specific V21V,1
ratio used within a given dog was determined according to the
algorithm in Figure 4.
FIGURE 2. Waveforms tested in parts ]c and Id. Panel A:
Five low-tilt biphasic waveforms tested in part 1c. Tilts ranged
from 1% to 17% depending on pulse duration. Panel B:
Voltage profile of 11 monophasic and 11 biphasic low-tilt
waveforms tested in part Id. The tilts ranged from 1% to 14%.
The rounding of the leading edge of each phase resulted from
connecting the nonconstant impedance of the heart in series
with a 200-Q resistor to achieve long decay times. The current
profiles did not show this pronounced rounding.
Part 1
In parts la and lb, strength-duration curves for
high-tilt waveforms (exponential decay time constant, of 7-9 msec) were determined, and in parts
lc and ld these curves were determined for low-tilt
waveforms (7=40-180 msec). For the biphasic waveforms studied, the first-phase duration was held
constant, whereas the second-phase duration was
varied. The goal of these protocols was to compare
the defibrillation thresholds of monophasic shocks
with those of biphasic shocks, paying particular attention to the regions of the strength-duration
curves where the biphasic DFTs change from being
lower to being higher than the monophasic DFTs.
The DFT for each waveform was determined by a
modified Bourland protocol.22 Each trial consisted of
first attempting defibrillation at a voltage that was
our best estimate of the DFT for that waveform. The
phase-one leading edge voltage (V,,) was increased in
r,
increments of approximately 20% after a failed attempt or was decremented by the same percentage
after a success. When the transition from failure to
success or success to failure was found, a final shock
was tested with V11 midway between the successful
and unsuccessful voltages. The minimum successful
voltage as measured across the heart was considered
the DFT for that waveform. This value was thereby
established within ±+ 10%. Because the Ventritex
HVS defibrillator is limited to 10-V increments,
however, when V11 at the DFT is less than 100 V in
parts la, lb, and 2, the potential error of the DFT
determination is greater than 10%. To avoid damaging the heart, V11 above 560 V was not given. In the
seven of 666 DFT determinations that did not defibrillate with this maximum voltage, the DFT was
assumed to be 10% higher than the highest delivered
shock. Had the DFT determination been completed
in these cases, the resulting DFT would have been at
least as high as the assumed value. If the "true" DFT
were higher than the assumed value, it would only
have strengthened the statistical significance of the
results reported for these data points. All DRTs were
determined in random order. Eight dogs were used in
parts la, eight were used in part lb, 14 were used in
part lc, and 14 were used in part ld.
Part la. In part la, the following waveforms were
tested: 1) six monophasic truncated exponential
monophasic waveforms of durations of 3.5, 7.0, 10.5,
14.0, 17.5, and 20.0 msec and 2) eight biphasic
truncated exponential waveforms with phase-one durations of 3.5 msec and phase-two durations of 1.0,
2.0, 3.5, 7.0, 10.5, 14.0, 17.5, and 20.0 msec. These
biphasic waveforms were "one-capacitor" pulses with
Feeser et al Success Curves for Defibrillation
'I3est'
Compute 2 DFTs for 50%
waveformn;
'Worst'
of the 2 trials
(2
67%. 83%. and
waveforms. 4
tmes
defibril
dogs)
by
feutue
100%
10o
high, therefore
olts for
each.future shoks.
Doal
rarforms defibri1la c
4 o
=
imut
Attempt defibrillation with the
50%,
llations
,ks
least defibrillations
per 4 shodlks
most
averagepe4shl
VII m
4
r
_
(6dogs)
es
Test the 67% 8 1100%
waveforms 4 time.ss each. If the
67%6 waveform su,cceeds at
least 2 more time.s than the
these
(I dog). ifuse
not, use the
and
~~~~~~50% 83% wa,iveforms (1 dog).
~~~~~~~~~pulses
/
yes
2131
(S dogs)
n c
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
best & worstL
btween
forms
2:2?
Choose the best atind worst
shocks as the 2 w,;aveforms
for fiurther testing.*, In case of
a tle for best or worst. choose the
og
which ppairs either the
r
(1
dogl ~wavefonn
1
,
~~~~50%& 83% wave]forms or the
100%w raveforms for
67urhe study if the 50%
Raise VII by 10 volts for
I
83% and
for0n
berstnticllhe
pulse 4 times.
)r worst, choose
the 50% and 83%6 waveforms).
FIGURE 4. Diagram showing algorithm for selecting
waveforms to be tested in part 2. The objective of the
procedure was to choose the two waveforms most likely
to show a dip in the defibrillation percent success curve
with increasing energy. Differences in the percent success of the waveforns could best be shown by setting V1,
to a level that resulted in an average of about 50%
success for the tested waveforns. Setting V,, too high or
too low could result in all waveforms either succeeding
or failing evety time. In five dogs, waveforms B and C
were selected after the initial four sets offour shocks. In
two dogs, an additional eight shocks were delivered; and
in one dog, 16 extra shocks were given.
n1
the phase-two leading edge voltage (V2,) equal to the
phase-one trailing edge voltage (V,,) (Figure 1, panels A, B). Subsequently, biphasic waveforms are
referred to by the notation "phase-one duration/
phase-two duration," as in, for example, a 3.5/10.5
one-capacitor waveform.
Part lb. In part lb, the waveforms used were 1) six
monophasic waveforms identical to those in part la,
2) eight biphasic waveforms that differed from the
part la biphasic shocks only in that the phase-one
duration was 7.0 msec, and 3) eight "two-capacitor"
biphasic waveforms (V21=V11) with 7.0-msec first
phases and the same duration second phases as the
other biphasic shocks (Figure 1, panels A, C, D).
Part ic. In part lc, low-tilt truncated exponential
biphasic waveforms were studied. These pulses had
3.5-msec first phases and 1.0-, 2.0-, 3.5-, 6.0-, and
8.5-msec second phases. With rm'40 msec, the tilts
varied between 1% and 17% depending on pulse
duration (Figure 2, panel A).
Part Id. The part id waveforms were 1) 11 very-lowtilt truncated exponential monophasic pulses with
durations of 1.0, 2.0, 3.0, 3.5, 4.0, 5.0, 6.0, 7.0, 8.5, 10.5,
and 14.0 msec and 2) 11 very-low-tilt biphasic stimuli
with 3.5-msec first phases and second phases of 1.0,
2.0, 3.0, 3.5, 4.0, 5.0, 6.0, 7.0, 8.5, 10.5, and 14.0 msec.
Average tilt ranged from 1% to 14% with r=180 msec
(Figure 2, panel B). The long time constant was
achieved by connecting a 200 Q resistor in series with
the nonconstant impedance of the heart, which varies
with both shock voltage and duration. Thus, the actual
voltage pulse seen by the heart was not a pure
exponential decay waveform, particularly during the
first several milliseconds of each phase. The 200 fl
resistor acted as a voltage divider such that the
defibrillator could deliver a maximum of about 200 V
to the heart, which was not always enough to defibrillate. In these cases, most often involving the 1.0- and
2.0-msec monophasic pulses, the series resistance was
reduced to 100 Ql or 50 Ql. The tilts remained below
14% even with these changes.
Part 2
Kavanagh et al13 have shown that a 6.0/6.0 onecapacitor pulse defibrillates with a lower V11 than
does a comparable two-capacitor form. This result
suggests that if the first-phase amplitude is held
constant for this 6.0/6.0 waveform, the probability of
success curve will decrease as V21 is raised from
approximately 50% of V1 to 100% of V1y. This
possibility was tested in part 2.
First, to establish the voltage range at which both
defibrillation successes and failures could be expected, the DFT was determined for a 6.0/6.0 biphasic truncated exponential waveform with V2 about
50% of Vy1. The DFT was determined twice by using
the protocol previously described, and the average of
the two V11 values was computed.
Next, to search for the two waveforms that would
most dramatically demonstrate a decrease in defibrillation efficacy with increasing phase-two amplitude,
the defibrillation success or failure of four sets of four
shocks was tested. The four waveforms in each set
were 6.0/6.0 biphasic truncated exponentials with Vy
as just established and V21 levels that were approximately 50%, 67%, 83%, and 100% of V,, (Figure 3).
2132
Circulation Vol 82, No 6, December 1990
--
500
Monophasic
-
Phase 1=3.5ms, 1 -cap
Phase 1 =7ms, 1 -cap
Phase 1 =7ms, 2-cap
-e--*
300/3P
0
0
X
-
100
10
0
20
30
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
15 -
10
00
o3
/
0
/
5>
C
-
/
to/1
Sf
/
_
,^
FIGURE 5. Scatterplots of leading edge voltage
(panel A), leading edge current (panel B), and
total delivered energy (panel C) versus totalpulse
duration for parts la and lb waveforms at
jthreshold. The monophasic thresholds from parts
la and lb are combined. On at least one occasion for each of the one-capacitor biphasic waveforms with second phases of 17.5 or 20.0 msec
@ Z (B) (m) defibrillation was not achieved with the
maximum shock of V11=560 V Thus, defibrillation thresholds for these waveforms may be low
6 :!testimates. For biphasic waveforms with a 7.0msec first phase, the leading edge voltage and
current are lower for the two-capacitor waveforn
than for the one-capacitor waveforn at all total
durations except 140 and 17.5 msec (§p<0.05,
tp<0.02). The total energy is lower for the
30
two-capacitor waveform only when the total duration is at least 24.5 msec, whereas the onecapacitor waveform performs better in terms of
energy at 10.5, 140, and 17.5 msec.
X
0S
t
t
0
0
10
20
30
20
0~
~
~
_10
0
2.s
12.5
0,
0
~~
0
~
~
~~~~
A
Total pulse duration (ms)
Feeser et al Success Curves for Defibrillation
300
-
-0--
-U--.
0
0)
CD
2133
Phase 1=3.5ms, 1-cap
Phase 1=7ms, 1-cap
Phase 1=7ms, 2-cap
Monophasic
200
00
0
(A)
0
a-4.'
FIGURE 6. Scatterplots of phase-two voltage
(panel A) and phase-two energy (panel B) at
threshold versus phase-two duration for parts la
and lb waveforms. Also included is the threshold
voltage or energyfor parts la and lb monophasic
waveforms combined. The DFT was not reached
by 560 Vin at least one dogfor certain waveforms
(X). For longer durations, V21 and J2 for the part
la biphasic waveforms were greater than the
corresponding values for monophasic pulses
equal in duration to the biphasic second phase
100 -
0
CL
0
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
0
20
10
(*p<O.O5, °p<O.Ol).
0
6
>.h
0)
L-
3
4
00
c0a
0
0
C')
CD
(B)
00
2
0
10
20
Phase two duration (ms)
The shocks within each set were given in random
order. Based on the results of these 16 shocks and the
algorithm outlined in Figure 4, two waveforms (subsequently called waveforms B and C) were selected
for further study. This complex method for choosing
waveforms was devised because pilot studies revealed
the pronounced impact on defibrillation success of
just a 10-V change in V11 or a 10% variation in V2,/V11.
The process resulted in using the 67% and 100%
waveforms in five dogs, the 50% and 83% waveforms
in two dogs, and the 67% and 83% waveforms in one
dog. Having selected waveforms B and C, these two
waveforms and two more 6.0/6.0 biphasic pulses with
the same V,, and with V2 20% and 140% of V,,
(waveforms A and D, respectively) were tested for
defibrillation success 12 times each. Again, each set
of four shocks was randomly ordered. Only these
final 12 trials of each waveform were used in determining the defibrillation percent success.
After completing one of the protocols just described, the dog was euthanized by electrically induced fibrillation. The electrodes were removed, and
the heart was excised and weighed.
Statistical Analysis
In part 1, all comparisons between defibrillation
threshold values of leading edge voltage, leading
edge current, and total energy were made with paired
t tests. In part 2, paired t tests were applied to
compare the percent success of waveforms A versus
B, B versus C, and C versus D. Statistically significant
comparisons discussed subsequently imply p<0.05
unless stated otherwise.
2134
Circulation Vol 82, No 6, December 1990
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
_.,| ~
TABLE 1. Leading Edge Voltage, Leading Edge Current, and Total Delivered Energy for Parts la and lb Waveforms at the
Defibrillation Threshold
Phase-two duration for biphasic waveforms and total
duration for monophasic waveforms (msec)
1.0
2.0
3.5
7.0
10.5
Leading edge voltage for
...
Part la monophasic
137.5±35.4
..
153.8+27.1
149.4±27.3
92.6±16.6
Phase 1=3.5 msec, 1-cap
101.5±18.9*
111.1+16.3
228.7±69.3t
258.8±83.6t
...
...
164.7±42.0
155.5±37.1
Part lb monophasic
171.3±42.4
134.6+33.6§
Phase 1=7.0 msec, 1-cap
116.9±21.3t
156.6±55.7
110.2±19.5§
111.2±23.2*§
90.4±23.8
103.4±23.3
113.8+29.3t
167.8±53.0
Phase 1=7.0 msec, 2-cap
98.5±22.5t
Leading edge current for
...
...
3.37±1.01
3.76±0.87
Part la monophasic
3.56±0.64
2.68±0.50
2.25±0.41
Phase 1=3.5 msec, 1-cap
2.39±0.47t
5.56±1.71t
6.44±2.06t
4.31±1.31
3.89±1.12
Part lb monophasic
...
...
4.25±1.19
3.99± 1.63
Phase 1=7.0 msec, 1-cap
2.78±0.56§
2.82±0.63t§
3.33±0.98t§
3.00±0.77t
2.25±0.67
2.59±0.67
Phase 1=7.0 msec, 2-cap
2.84±0.77t
4.51±1.80
2.42±0.57t
Total energy for
1.43+0.78
...
1.22±0.44
...
1.83+0.65
Part la monophasic
0.56±0.18
0.71±0.20
Phase 1=3.5 msec, 1-cap
0.75+0.28*
6.30±3.69t
4.61+2.33t
...
1.52±0.76
1.83±0.93
...
Part lb monophasic
2.36±1.26
0.98±0.32
1.43±0.77
Phase 1=7.0 msec, 1-cap
0.98±0.45t
1.19+0.40t 1
2.37±1.67§
4.82±2.89*
0.99±0.43
0.92±0.47
1.29±0.57
1.99±1.01*
Phase 1=7.0 msec, 2-cap
cap, capacitor; DFT, defibrillation threshold.
*p<0.05, tp<0.02 for difference between indicated biphasic waveform and corresponding monophasic waveform of equal total duration.
I p<O.05, §p<0.02 for difference between corresponding one and two-capacitor waveforms with phase-one duration= 7.0 msec.
t oDFT not reached at 560 V in at least one dog for these waveforms.
There has been some debate about the use of DFTs
as a means of comparing waveforms, with some authors
preferring the use of percent success curves2324 and
others finding the DFT to be a practical measure when
comparing numerous waveforms.25 In this study, DFTs
were used in part 1 because of the many waveforms
being compared, and percent success curves were used
in part 2 where just four waveforms were compared.
Results
Parts Ja and lb
The mean DFT leading edge voltages (V1, and V21),
leading edge current (111), and total and phase-two
energies (Jt and J2) for all waveforms are shown in
Figures 5 and 6. The monophasic DFTs from parts la
and lb have been combined to give 16 measurements
per data point, and the biphasic DFTs each represent
eight dogs per point.
Whether examining voltage, energy, or current
(Figure 5), the mean DFT for monophasic waveforms
is greater than for one-capacitor biphasic waveforms
of equal total duration when the biphasic second
phase is equal to or shorter than the first phase, but
the monophasic DFT is lower when the biphasic
second phase is longer. For the biphasic waveforms
with a 7.0-msec first phase, V11 at threshold for
two-capacitor waveforms is significantly lower than
for one-capacitor waveforms for all second-phase
durations except 7.0 and 10.5 msec. Because the
second phase of the two-capacitor waveform is larger
than the second phase of a one-capacitor waveform
with the same V1y, these results are not also true for
shock energy; the total delivered energy is significantly lower for the two-capacitor form only when the
second phase is at least 17.5 msec. With increase in
the second-phase duration, the defibrillation effectiveness of the one-capacitor waveforms degrade
much more dramatically than either the monophasic
250
4,
-
200
0
0
150
0
10
100
-
CD
I-
50-
0
10
20
30
Total Pulse Duration (ms)
FIGURE 7. Scatterplot of trailing edge voltage versus total
waveforn duration for parts la and lb stimuli. For the
biphasic waveforms, the open datum points represent the
phase-one voltages and the solid points denote the phase-two
voltages.
Feeser et al Success Curves for Defibrillation
TABLE 1. Continued
Phase-2 duration for biphasic waveforms and total
duration for monophasic waveforms (msec)
20.0
17.5
14.0
164.6±44.8
184.8±58.6
315.1±88.8t
385.7±+107.8t*
177.8±35.1
165.2+37.0
459.4±117.6t§
195.7±56.2
306.2±125.9§
176.5±56.5t
4.01±1.12
7.96±2.14t
4.59±1.28
8.29±4.1511
4.58±1.75t
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
2.28±1.12
9.20±4.53t
2.77±1.14
9.48±7.10
5.17±3.41
157.0±25.5
414.3± 122.7t
248.8±66.3
525.7±46.9t§
209.4±70.4
4.19±1.12
3.79±0.58
10.90±3.80t
6.51±1.95
12.39±3.85t§
14.41±2.33t§
4.91+1.60
5.45+1.96
3.19±1.99
2.19±0.68
16.31 ±8.37i
5.72±2.81
4.46±1.54
10.08±3.61lt
13.97±7.75tt
2.44±1.06
19.18±8.30t§
6.52±3.44
23.61±3.77*§
7.73±4.37
two-capacitor biphasic waveforms. Table 1 displays the means, standard deviations, and significance levels for these data.
Figure 6 compares the defibrillation effectiveness
of the second phase of biphasic waveforms to that of
monophasic waveforms. Interestingly, a one-capacitor waveform with a 3.5-msec first phase and a
second phase at least 14.0 msec long requires more
voltage and energy to defibrillate than does a
monophasic pulse equal in duration to the biphasic
second phase. For example, for the 3.5/20.0 onecapacitor waveform, the presence of a first phase
makes defibrillation by the second phase more difficult than had the first phase been absent. Also, for
the biphasic waveform with a 7.0-msec first phase
and a long second phase, V21 and J2 approach the
corresponding monophasic values.
Figure 7 shows the phase-one and phase-two trailing edge voltages (V1, and V2t) at threshold for the
part la and lb waveforms. The shapes of the V1,
curves are similar to the leading edge voltage curves
in Figures 5 (panel A) and 6 (panel A). The V2t
curves appear to be determined at longer durations
primarily by the rapid decay time constant. That is,
the phase-two trailing edge voltage does not determine the DFT in any simple manner. The trailing
edge voltages for the subsequent low-tilt waveforms
are not shown because, as expected, the trailing edge
curves are nearly identical to the leading edge curves
for those waveforms.
or
Part ]c
Figure 8 shows the leading edge voltage and
cur-
rent and total energy at threshold versus phase-two
duration for the five low-tilt waveforms of part lc
2135
illustrated in Figure 2 (panel A). A hump in each
curve is evident when the second-phase duration is
6.0 msec. The 3.5/6.0 waveform has a significantly
higher leading edge voltage and current than the
neighboring 3.5/3.5 or 3.5/8.5 forms, and a higher
total energy than the 3.5/3.5 form. Thus, the
strength-duration curves in terms of voltage and
current increase and then, surprisingly, decrease as
the second-phase duration is increased.
Part Id
Part ld expands the information provided in part
lc about low-tilt waveforms by including the
strength-duration curves for monophasic waveforms and by determining the DFTs for a wider
range of pulse durations (Figure 2, panel B). For
the four total pulse durations at which both
monophasic and biphasic waveforms were tested
(7.0, 8.5, 10.5, and 14.0 msec), the voltage, current,
and energy thresholds for biphasic waveforms were
significantly lower (p<0.02) than for monophasic
pulses, except at 8.5 msec (Figure 9). The peak seen
at 8.5 msec in the biphasic curves of Figure 9 (top
panel) is greater than the neighboring portions of
the curve for durations of 7.5 msec and 10.5 msec
(p<0.03). The apparent hump at 7.0 msec is not
significantly different from neighboring values for
current thresholds. For voltage thresholds, it is
greater than the DFT at 7.5 msec (p=0.04) but not
greater than the DFT at 6.5 msec. Given this weak
and inconsistent statistical support for the presence
of a second hump at 7.0 msec separate from the
peak at 8.5 msec, this second hump will not be
considered further. The presence of at least one
hump reaffirms the findings of part lc and again
shows that the strength-duration curve for biphasic
waveforms does not follow a simple hyperbolic
curve.
Figure 10 for part ld, like figure 6 for parts la and
lb, compares the DFT values of the biphasic second
phase with that of comparable monophasic waveforms. The phase-two leading edge voltage at threshold is lower than the threshold voltage of an equiduration monophasic pulse for all durations tested
(p<0.005). The current is lower for the biphasic
second phase at all durations (p<0.04) except 5.0
and 8.5 msec, where there is merely a trend (p= 0.06
and 0.19, respectively). These results contrast with
those shown in Figure 6 (panel A) for high-tilt
waveforms, where for longer durations, V21 of biphasic waveforms is as great or greater than the leading
edge voltage for monophasic pulses.
Part 2
Figure 11 plots the defibrillation percent success
for waveforms A-D shown in Figure 3. Phase-two
defibrillation success does not follow a sigmoidal
dose-response curve for these waveforms. Thus, increasing the total energy of a pulse does not always
assure a more probable defibrillation. As V21 is raised
from 0.61 V11 to 0.94V1W and then to 1.41V1l, the
2136
Circulation Vol 82, No 6, December 1990
240
200
t
160
-
0
(A)
120
P-
80
FIGURE 8. Scatterplots of threshold voltage (panel
A), and current and delivered energy (panel B) versus
phase-two duration for part ]c waveforns. Panel A
shows V11 for all 14 dogs, with the averages represented
by the bold solid circles. Panel B plots the average
leading edge current (n) and total delivered energy (a).
40
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
In both plots, adjacent points that are significantly
0
different from each other (p<0.05) are denoted by an
asterisk above the line connecting those points.
*
0
10
c
(B)
C
0
0
2
0.
E~
L*zzJfit
0
2
4
6
Phase 2 Duration (ms)
percent success first decreases (p<0.005) and then
increases (p<0.02). The wide range of percent success for a given waveform results from sensitivity of
success to V11. For example, in dog 3 no waveform
was more than 20% successful. Had V1y been increased 10 V and V21 adjusted proportionately, however, all four waveforms may have succeeded a
majority of the time. Also, because the average
difference in percent success between waveforms B
and C was only about 20%, the effectiveness of using
just 16 preliminary shocks to choose the optimal V21
for these waveforms may be questioned. In other
words, the protocol for choosing waveforms B and C,
even though complex due to the sensitivity of success
to slight changes in V11 and V21, probably did not
consistently select the optimal waveforms. The decline in efficacy between waveforms B and C was
significant despite this limitation.
Discussion
Mechanisms of Defibrillation With
Biphasic Waveforms
The most striking findings of this study are that for
the second phase of certain biphasic waveforms, the
defibrillation strength-duration curve is not a simple
hyperbola or parabola and the defibrillation percent
success curve is not sigmoidal. Thus, the response to
the second phase is not the same as the response to
monophasic defibrillation waveforms.23'26,27 Instead,
as the second-phase charge is increased, either by
lengthening the phase (Figures 8-10) or increasing its
amplitude (Figure 11), defibrillation efficacy first improves, then declines, and then again improves. Another major finding is that under certain conditions,
the presence of the first phase of a biphasic waveform
makes defibrillation less effective than had only the
second phase been delivered. For a high-tilt one-
Feeser et al Success Curves for Defibrillation
2137
Ti71
200 -
a
-0-*150
CD
-a--fr-
-
Monophasic Voltage
Biphasic Voltage
Monophasic Current
Biphasic Current
.6
E
.5-5
CD
0
.4
100
3
m
.
so5
2
2?
0
.
0
c
m
C)
-
E
.
CO
9
1
.
.
.
0
5
.1
.
1
1
10
2
m
E
_
x
E
~~~~~~~~~~~1.
.
15
20
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
25-
FIGURE 9. Scatterplots of maximum voltage and maximum current (top panel), and total delivered energy
(bottom panel) versus total pulse duration for part Id
waveforms at threshold. Biphasic curves have been extended by dotted lines to include the defibrillation threshold for the "biphasic" waveform with a zero-amplitude
secondphase (3.5/0.0). Hump seen in the biphasic curves
at 8.5 msec (top panel) is greater than the neighboring
portions of the curve for durations of 7.5 msec and 10.5
msec (p <O. 03).
2.0 -
w01
-
a)
1.5 -
c
0
1.00
W
L-
*-
0.5-
0.0-
-a-*.~~~
*
.
9
5
5
*
.
.
*
5
**
.
Monophasic
Biphasic
.
10
.
15
|.
20
Total Pulse Duration (ms)
capacitor biphasic waveform with a 3.5-msec first
phase and a second phase at least 14 msec long, the
phase-two voltage and energy at threshold are higher
than the threshold values for a monophasic pulse
equal in duration to that second phase (Figure 6).
Interestingly, this relation does not hold for low-tilt
waveforms (Figure 10).
These results suggest current theories about the
mechanism of biphasic waveform effectiveness must
be altered or expanded. Jones and coworkers8 have
suggested that the first phase acts as a "conditioning
prepulse" and the second phase as an "excitation
pulse." Because the resting membrane potential has
been reported to be about -60 mV during fibrillation,28 Jones et a18 posit that the first phase may serve
to hasten recovery of sodium channels so the second
phase can more easily open channels to excite the
tissue. A similar theory, based on the observation of
Cranefield and Hoffman29 that a hyperpolarizing
pulse given during phase two of the action potential
can shorten the refractory period, postulates that the
first phase of a biphasic waveform may act to shorten
refractoriness so the second phase can excite more
cells.9
These ideas are strengthened by the observation that
in rabbit hearts, the ratio of the monophasic waveform
DFT to the biphasic waveform DFT increases when the
heart fibrillates for 30 seconds before shock delivery
versus fibrillating for 5 seconds.30 If the myocytes
become progressively more depolarized during the 30
seconds of fibrillation, the ability to hasten recovery of
sodium channels and shorten refractoriness would
make the biphasic waveform especially advantageous
for longer fibrillation times.
This study suggests more is involved in lowering
the DFT of biphasic waveforms than just reactivating
Circulation Vol 82, No 6, December 1990
2138
200
0
0.
lU
r
0
150
0
E
0
C
0
C)
0
CD
C"
04
0
0
100
0
CO
0
0.
E
Q
50
E
a._E
x
X
CD
0
0
10
FIGURE 10. Scatterplot of maximum phase-two voltage
and current at threshold versus phase-two duration for part
Id biphasic waveforms. Also shown are the voltage and
current defibrillation thresholds (DFTs) for part Id
monophasic waveforms plotted versus total pulse duration.
Biphasic curves have been extended by dotted lines to
include the DFT for the "biphasic" waveform with a
zero-amplitude second phase (3.5/0. 0). Biphasic waveforms
defibrillate with a lower phase-two voltage than the voltage
of the equiduration monophasic pulse for all waveforms
tested (p<0.0005). Current is lower for the biphasic second
phase at all durations (p<0.04) except 5.0 and 8.5 msec (*).
15
Duration (ms)
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
sodium channels. If the first phase simply speeds
recovery of sodium channels so the second phase can
more easily excite tissue, increasing the second phase
amplitude for high-tilt waveforms (part 2) or duration for low-tilt waveforms (part lc and part ld)
should continually improve defibrillation efficacy, as
is true for monophasic waveforms.23,27 A decline in
efficacy should occur only when the second phase
begins to damage tissue. Once that occurs, further
increasing phase two should not improve its efficacy.
Contrary to this expectation, we found that inceasing
the second phase of some biphasic waveforms causes
a decrease in defibrillation efficacy followed by an
increase in efficacy as the second phase is increased
still further. Because the decline in defibrillation
efficacy occurs with either increases in the duration
or amplitude of the second phase, the decline is not
likely just due to time limitations of the sodium
channel recovery.
If the second phase is indeed the excitatory phase, it
must be able to depolarize previously hyperpolarized
membranes. For each of the waveform types tested,
(0
0
U
Q
:1
100
80
U,)
60
c
0
40-
am
a-
adding a 1.0-msec second phase to a monophasic
stimulus reduces the DFT. For example, a 3.5/1.0
biphasic stimulus is more effective than a 3.5-msec
monophasic pulse. It is not clear if a 1.O-msec pulse can
depolarize previously hyperpolarized membranes. Estimates of myocyte membrane time constants range from
0.2 to 4.9 msec.31 If the membrane behaves linearly, the
second phase of a 3.5/1.0 square biphasic stimulus can
depolarize a membrane hyperpolarized by the first
phase only if the membrane time constant is shorter
than 1.6 msec (see "Appendix"). For longer time
constants, there is not enough energy in the second
phase to overcome the hyperpolarizing effects of the
first phase. Thus, depending on the membrane properties during fibrillation, depolarization may or may not
be achieved by a 1.0-msec second phase. Even if
depolarization is achieved, it is small in magnitude and
lasts less than a millisecond.
Some models of a discrete cardiac strand yield
different results, suggesting a 1.0-msec second phase
can depolarize the myocyte membrane.32-34 These
models indicate that the transmembrane potential
-Zui~Zi
20-
0
Waveform A
---O----
Dogs 1 - 8 FIGURE 11. Scatterplot of defibrilla-
Average
tion percent success versus phase-two
leading edge voltage for part 2 waveforms. When increasing V2, from 21% to
141% of V the percent success first in-
creases (p <0.01), then decreases
(p<0.005), and then increases again
B
17±3
51±13
(.21±.03) (.62±.08)
C
D
77±17
116±23
(.94±.08) (1.41±.04)
V21±s.d. (V21/V1 1)
(p<0.02).
Feeser et al Success Curves for Defibrillation
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
induced by an applied field consists of an aperiodic
term and a periodic term. The aperiodic term changes
from depolarizing close to the cathode to hyperpolarizing close to the anode and arises with a time constant
of about 20 msec. The periodic term varies from
hyperpolarization to depolarization across each cell
and achieves its maximum value within 1 msec after
applying the stimulus. The aperiodic term has a significant effect only near the defibrillation electrodes,
whereas the periodic term is responsible for the effects
of the defibrillation shock throughout most of the
myocardium. Thus, the fast response of the periodic
term could explain how very short second phases can
excite tissue and reduce DFTs. The gradation from
depolarization to hyperpolarization induced within
each cell has also been predicted by Chernysh and
coworkers35 based on a different mathematical model.
The results discussed thus far suggest at least two
mechanisms exist for defibrillation with biphasic
waveforms, one that is more effective for smaller
second phases and another that becomes more effective as the second phase is increased. The concept
that each phase of a stimulus induces a gradation
from hyperpolarization to depolarization along each
myocyte provides a starting point for hypothesizing
one possible physical interpretation for the dual
mechanisms. If the first phase of a biphasic waveform
opens sodium channels on one half of each myocyte
and the second phase opens channels on the other
half without closing all of those on the first half, the
biphasic pulse might be more likely to defibrillate
than if only the first phase were given. If the total
charge of the second phase were increased by either
lengthening that phase (part lc and part ld) or
increasing its amplitude (part 2), however, the second phase might completely close channels on the
half of the membrane initially depolarized by phase
one. This would undo the positive interaction between the phases, and could explain the poor performance of the 3.5/6.0 waveform in part lc, the 3.5/5.0
and 3.5/6.0 waveforms in part ld, and waveform C in
part 2. The ability to halt myocyte excitation even
after sodium channels have opened and the threshold
has been reached has been demonstrated by Weidmann.36 He showed that delivering a hyperpolarizing
pulse of sufficient amplitude to a cell that has been
depolarized above threshold can halt excitation during phase 0 of the action potential. This mechanism
could offer a reason for the decline in efficacy as the
second-phase amplitude or duration is increased. As
the second phase is further increased, it should gain
sufficient energy to excite cells by itself, opening
channels on half of the membrane just as a monophasic pulse would. This could explain the increase in
efficacy with waveform D in part 2 and with those
waveforms in part lc and ld having second phases at
least 7.0 msec long.
The idea that the first phase hastens recovery of
sodium channels can be incorporated to help explain
the part ld data. If sodium channel activation occurs
within the portion of each cell hyperpolarized by the
2139
first phase, the second phase of a biphasic waveform
should defibrillate by using less energy than a comparable monophasic pulse equal in duration to that
second phase. This fact is demonstrated in part ld
(Figure 10).
High-tilt biphasic waveforms, however, perform
quite poorly as the second phase is extended (Figures 5
and 6). Recall that a long second phase of a truncated
exponential biphasic waveform requires a greater voltage and current to defibrillate than a comparable
monophasic pulse (Figure 6), whereas a long square
second phase performs better than the comparable
monophasic stimulus (Figure 10). The poor performance of truncated exponential waveforms with long
tails, reported previously,37'38 has been suggested to
result from reinducing fibrillation38 and has been reproduced with a mathematical model based on linear
dynamic system theory.39 If the poor performance of
exponential waveforms is indeed due to reinducing
fibrillation, this implies that not only can some biphasic
waveforms defibrillate more effectively than monophasic pulses, but certain biphasic waveforms can also
reinduce fibrillation more easily. If each of these processes are forms of activating tissue, it is not contradictory to suggest biphasic waveforms both defibrillate and
refibrillate effectively. Wharton et al,40 however, found
that the peak voltage that will induce fibrillation when
delivered during the vulnerable period is lower for a
3.5/2.0 biphasic stimulus than for a 5.5 -msec monophasic shock. And, certain biphasic waveforms are less
able than monophasic pulses to excite tissue during
the relative refractory period,41'42 suggesting that
defibrillation efficacy might be enhanced for biphasic
waveforms because they are less able to reinduce
fibrillation in partially refractory tissue.43 Thus, although no unifying explanation is yet apparent for
these seemingly contradictory findings, it may be
inferred that more is involved in defibrillating myocardium than simply exciting tissue.
Most Effective Wavefomns
This study confirms that biphasic waveforms defibrillate with less voltage, current, and energy than
equiduration monophasic pulses when the biphasic
second phase is equal to or shorter than the first
phase.13'4'9'12'13 As the second phase is lengthened,
the performance of high-tilt one-capacitor waveforms
deteriorates dramatically, whereas the efficacy of
comparable two-capacitor waveforms degrades less
dramatically (Figures 5 and 6) and that of low-tilt
waveforms improves in terms of voltage and current
(Figure 9). Thus, these results illustrate the pronounced effect of the amount of tilt on defibrillation
efficacy with biphasic waveforms.
Apart from the broad guideline that smaller second phases are preferable to large ones, specific
recommendations about optimal waveforms are difficult to establish because of the remarkable sensitivity of the defibrillation threshold to phase durations.
For example, in part lb, the defibrillation threshold
voltage for two-capacitor waveforms was lower than
2140
Circulation Vol 82, No 6, December 1990
for comparable one-capacitor waveforms in all cases
except for the 7.0/7.0 and 7.0/10.5 shocks, in which
case the one-capacitor and two-capacitor waveforms
were not significantly different. The work by Kavanagh et al,13 however, further substantiated by part
2 of this study, showed that when each phase was only
1 msec shorter, that is, 6.0/6.0, the one-capacitor
waveform defibrillated more effectively than the twocapacitor form. Because canine hearts do not provide
an exact model of human hearts,44 this sensitivity of
defibrillation to phase-duration changes means biphasic waveforms that are optimal for defibrillating
canine hearts require testing before clinical implimentation in humans. With this qualification in mind,
the lowest defibrillation energy found in this study for
a high-tilt one-capacitor biphasic waveform is lower
when the first phase is 3.5 msec than when it is 7.0
msec (3.5/2.0 versus 7.0/2.0, p<0.01).
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Conclusion
This study has shown that more than one mechanism must operate to explain the strength-duration
and probability of success curves for defibrillating
with monophasic and biphasic stimuli. Possible
mechanisms have been hypothesized. Further experimentation will be required to verify these theories.
Appendix
Assume that a myocyte membrane behaves like a
simple resistance-capacitance circuit with a time constant r. The value of r at which the second phase of
a 3.5/1.0 square biphasic waveform has delivered
enough charge to reverse the polarity imposed on the
membrane by the first phase is computed herein.
Given a unity driving voltage, the transmembrane
voltage after the first phase is
v1=1-e- 3.5 msec/-,
The total driving voltage of the second phase is then
the v1 already on the membrane plus the unity voltage
applied by the waveform (i.e., v1+ 1). Thus, the
change in transmembrane voltage effected by the
second phase is
v2=(v1+1)(1-e -.Omsec/T)
Setting v1 plus v2 equal to zero and solving for r by
iteration gives r=1.6 msec. Only if r is less than
1.6 msec would the second phase reverse the transmembrane polarity.
Acknowledgments
We thank Ellen Dixon, Jenny Harrison, and
Sharon Melnick for their technical assistance.
References
1. Schuder JC, Gold JH, Stoeckle H, McDaniel WC, Cheung
KN: Transthoracic ventricular defibrillation in the 100 kg calf
with symmetrical one-cycle bidirectional rectangular wave
stimuli. IEEE Trans Biomed Eng 1983;30:415-422
2. Jones JL, Jones RE: Decreased defibrillator-induced dysfunction with biphasic rectangular waveforms. Am J Physiol 1984;
247:H792-H796
3. Dixon EG, Tang ASL, Wolf PD, Meador JT, Fine MJ, Calfee
RV, Ideker RE: Improved defibrillation thresholds with large
contoured epicardial electrodes and biphasic waveforms. Cir-
culation 1987;76:1176-1184
4. Chapman PD, Vetter JW, Souza JJ, Troup PJ, Wetherbee JN,
Hoffmann RG: Comparative efficacy of monophasic and
biphasic truncated exponential shocks for nonthoracotomy
internal defibrillation in dogs. J Am Coll Cardiol 1988;12:
739-745
5. Platia EV, Wakins L Jr, Mower MM, Griffith LSC, Reid PR,
Mirowski M: Automatic implantable defibrillators, in Platia
EV (ed): Management of Cardiac Arrhythmias: The Nonpharmacologic Approach. Philadelphia, JB Lippincott Co, 1987, pp
272-303
6. Gabry MD, Brodman R, Johnston D, Frame R, Kim SG,
Waspe LE, Fisher JD, Furman S: Automatic implantable
cardioverter-defibrillator: Patient survival, battery longevity
and shock delivery analysis. J Am Coll Cardiol 1987;9:
1349-1356
7. Marchlinski FE, Flores B, Miller JM, Gottlieb CD, Hargrove
WC III: Relation of the intraoperative defibrillation threshold
to successful postoperative defibrillation with an automatic
implantable cardioverter defibrillator. Am J Cardiol 1988;62:
393-398
8. Jones JL, Jones RE, Balasky G: Improved cardiac cell excitation with symmetrical biphasic defibrillator waveforms. Am J
Physiol 1987;253:H1418-H1424
9. Tang ASL, Yabe S, Wharton JM, Dolker M, Smith WM,
Ideker RE: Ventricular defibrillation using biphasic waveforms: The importance of phasic duration. JAm Coll Cardiol
1989;13:207-214
10. Gurvich NL, Markarychev VA: Defibrillation of the heart with
biphasic electrical impulses. Kardiologiia 1967;7:109-112
11. Schuder JC, McDaniel WC, Stoeckle H: Defibrillation of
100-kg calves with asymmetrical, bidirectional, rectangular
pulses. Cardiovasc Res 1984;18:419-426
12. Schuder JC, McDaniel WC, Stoeckle H: Asymmetric biphasic
defibrillation: Big pulse first or last (abstract)? in Johnson AT,
Mates RE (eds): Proceedings of 40th Annual Conference on
Engineering in Medicine and Biology. Washington, DC, Alliance
for Engineering in Medicine and Biology, 1987, p 97
13. Kavanagh KM, Tang ASL, Rollins DL, Smith WM, Ideker
RE: Comparison of the internal defibrillation thresholds for
monophasic and double and single capacitor biphasic waveforms. JAm Coll Cardiol 1989;14:1343 -1349
14. Jones JL, Jones RE: Improved defibrillator waveform safety
factor with biphasic waveforms. Am J Physiol 1983;245:
H60-H65
15. Niebauer MJ, Babbs CF, Geddes LA, Bourland JD: Efficacy
and safety of the reciprocal pulse defibrillator current waveform. Med Biol Eng Comput 1984;22:28-31
16. Winkle RA, Mead RH, Ruder MA, Gaudiani V, Buch WS,
Pless B, Sweeney M: Improved low energy defibrillation
efficacy in man using a biphasic truncated exponential waveform (abstract). JAm Coll Cardiol 1987;9:142A
17. Jones DL, Sohla A, Bourland JD, Tacker WA, Kallok MJ,
Klein GJ: Internal ventricular defibrillation with sequential
pulse countershock in pigs: Comparison with single pulses and
effects of pulse separation. PACE 1987;10:497-502
18. Wetherbee JN, Chapman PD, Bach SM Jr, Troup PJ: Sequential shocks are comparable to single shocks employing two
current pathways for internal defibrillation dogs. PAGE 1988;
11:696-703
19. Jones DL, Klein GJ, Guiraudon GM, Sharma AD: Sequential
pulse defibrillation in humans: Orthogonal sequential pulse
defibrillation with epicardial electrodes. J Am Coll Cardiol
1988;11:590-596
20. Amlie JP, Owren T: The effect of prolonged pentobarbital
anaesthesia on cardiac electrophysiology and inotropy of the
dog heart is situ. Acta Pharmacol Toxicol 1979;44:264-271
21. Galindo AH, Davis TB: Succinylcholine and cardiac excitability. Anesthesiology 1962;23:32-40
Feeser et al Success Curves for Defibrillation
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
22. Bourland JD, Tacker WA Jr, Geddes LA: Strength-duration
curves for trapezoidal waveforms of various tilts for transchest
defibrillation in animals. Med Instrum 1978;12:38-41
23. Davy JM, Fain ES, Dorian P, Winkle RA: The relationship
between successful defibrillation and delivered energy in
open-chest dogs: Reappraisal of the "defibrillation threshold"
concept. Am Heart J 1987;113:77-84
24. McDaniel WC, Schuder JC: The cardiac ventricular defibrillation threshold: Inherent limitations in its application and
interpretation. Med Instrum 1987;21:170-176
25. Rattes MF, Jones DL, Sharma AD, Klein GJ: Defibrillation
threshold: A simple and quantitative estimate of the ability to
defibrillate. PACE 1987;10:70-77
26. Koning G, Schneider H, Hoelen AJ: Amplitude-duration
relation for direct ventricular defibrillation with rectangular
current pulses. Med Biol Eng Comput 1975;13:388-395
27. Chapman PD, Vetter JW, Souza JJ, Troup PJ, Wetherbee JN:
Comparative efficacy of monophasic and biphasic waveforms
for non-thoracotomy canine defibrillation (abstract). PACE
1988;11:497
28. Akiyama T: Intracellular recording of in situ ventricular cells
during ventricular fibrillation.AmJPhysiol 1981;140:H465-H471
29. Cranefield PF, Hoffman BF: Propagated repolarization in
heart attack. J Gen Physiol 1958;41:633-649
30. Jones JL, Swartz JF, Jones RE, Fletcher R: Increasing fibrillation duration enhances relative asymmetrical biphasic versus
monophasic defibrillation waveform efficacy. Circ Res 1990;67:
376-384
31. Trautwein W: Membrane currents in cardiac muscle fibers.
Physiol Rev 1973;53:793-835
32. Plonsey R, Barr RC: Effect of microscopic and macroscopic
discontinuities on the response of cardiac tissue to defibrillating (stimulating) currents. Med Biol Eng Comput 1986;24:
130-137
33. Krassowska W, Pilkington TC, Ideker RE: Periodic conductivity as a mechanism for cardiac stimulation and defibrillation. IEEE Trans Biomed Eng 1987;34:555-560
34. Krassowska W, Pilkington TC, Ideker RE: Two-scale asymptotic analysis for modeling activation of a periodic cardiac
strand. Mathl Comput Modelling (in press)
2141
35. Chernysh AM, Tabak VY, Bogushevich MS: Mechanisms of
electrical defibrillation of the heart. Resuscitation 1988;16:
169-178
36. Weidmann S: Effect of current flow on the membrane potential of cardiac muscle. J Physiol 1951;115:227-236
37. Chapman PD, Wetherbee JN, Vetter JW, Troup P, Souza J:
Strength-duration curves of fixed pulse width variable tilt
truncated exponential waveforms for nonthoracomy internal
defibrillation in dogs. PACE 1988;11:1045-1050
38. Schuder JC, Rahmoeller GA, Stoeckle H: Transthoracic ventricular defibrillation with triangular and trapezoidal waveforms. Circ Res 1966;19:689-694
39. Koning G: Some physical aspects of ventricular defibrillation:
A system analytic approach (thesis). Amsterdam, Free University of Amsterdam, 1972
40. Wharton JM, Richard VJ, Murry CE, Dixon EG, Reimer KA,
Meador J, Smith WM, Ideker RE: Electrophysiologic effects
in vivo of monophasic and biphasic stimuli in normal and
infarcted dogs. PACE 1990;13:1158-1172
41. Wharton JM, Frazier DW, Ideker RE: Electrophysiologic
effects in vivo of biphasic and monophasic stimuli (abstract). J
Am Coll Cardiol 1988;11:144A
42. Daubert JP, Frazier DW, Tang ASL, Hagler JA, Smith WM,
Ideker RE: Biphasic shocks excite refractory myocardium less
effectively than monophasic shocks (abstract). PACE 1988;11:
503
43. Ideker RE, Tang ASL, Frazier DW, Shibata N, Chen P-S,
Wharton JM: Ventricular defibrillation: Basic concepts, in
El-Sherif N, Samet P (eds): Cardiac Pacing and Electrophysiology. Philadelphia, WB Saunders, 1990 (in press)
44. Dreifus LS, Naito M, Michelson EL: What animal models
should be used to define antiarrhythmic efficacy? Acute dog
models, in Morganroth J, Moore EN, Dreifus LS, Michelson
EL (eds): The Evaluation of New Antiarrhythmic Drugs. The
Hague/Boston/London, Martinus Nijhoff Publishers, 1981, pp
17-32
KEY WORDS * arrhythmias * electrophysiology * ventricular
fibrillation
Strength-duration and probability of success curves for defibrillation with biphasic
waveforms.
S A Feeser, A S Tang, K M Kavanagh, D L Rollins, W M Smith, P D Wolf and R E Ideker
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Circulation. 1990;82:2128-2141
doi: 10.1161/01.CIR.82.6.2128
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1990 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/82/6/2128
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/