Reprinted with permission from

JOURNAL OF PACING AND CLINICAL ELECTROPHYSIOLOGY , Volume 24, No. 1, January 2001
Copyright © 2001 by Futura Publishing Company, Inc., Armonk, NY 10504-0418.

Prospective Randomized Comparison of

50%/50% Versus 65%/65% Tilt Biphasic
Waveform on Defibrillation in Humans
MICHAEL O. SWEENEY,* ANDREA NATALE,† KENT J. VOLOSIN,‡
CHARLES D. SWERDLOW,§ JAMES H. BAKER,\ and PAUL DEGROOTA
From the *Brigham and Women’s Hospital, Boston, Massachusetts, †Durham V.A. Medical Center,
Durham, North Carolina, ‡Our Lady of Lourdes Hospital, Camden, New Jersey,
§Cedars-Sinai Medical Center, Los Angeles, California, \Centennial Hospital, Nashville,
Tennessee, and AMedtronic, Incorporated, Minneapolis, Minnesota

SWEENEY, M.O., ET AL.: Prospective Randomized Comparison of 50%/50% Versus 65%/65% Tilt Bipha-
sic Waveform on Defibrillation in Humans. It is unknown if there is a single optimal biphasic waveform
for defibrillation. Biphasic waveform tilt may be an important determinant of defibrillation efficacy. The
purpose of this study was to compare acute defibrillation success with a three-electrode configuration in
humans using 50%/50% versus 65%/65% tilt truncated exponential, biphasic waveforms delivered
through a 110-mF capacitor. Acute DFTs for biphasic waveforms with 50%/50% versus 65%/65% tilt were
measured in random order in 60 patients using a binary search method. The electrode configuration con-
sisted of a RV coil as the cathode, and a SVC coil plus a pectoral active can emulator (CAN) as the anode.
The waveforms were derived from an external voltage source with 110-mF capacitance, and the leading
edge voltage of phase 2 was equal to the trailing edge voltage of phase 1. Stored energy DFT (9.2 6 5.7
[50%/50%] vs 10.8 6 6.4 [65%/65%] J, P 5 0.007), current DFT (10.9 6 4.0 [50%/50%] vs 12.0 6 4.4
[65%/65%] A, P 5 0.002) and voltage DFT (391 6 118 [50%/50%] vs 424 6 128 [65%/65%] V, P 5 0.004)
were significantly lower for the 50%/50% tilt waveform versus the 65%/65% tilt waveform using this
three-electrode configuration and a 110-mF capacitor. For an RV (2)/SVC plus CAN (1) electrode config-
uration and a 110-mF capacitor, a 50%/50% tilt biphasic waveform results in a 15% reduction in energy
DFT, 9% reduction in current DFT, and 8% reduction in voltage DFT versus a 65%/65% tilt biphasic
waveform. (PACE 2001; 24:60–65)
defibrillation, waveforms, implantable cardioverter defibrillators

Introduction forms. 1 ,2 Despite this observation, there is no

Minimization of the ventricular defibrilla-
tion energy requirement is a principal goal of re-
search in implantable cardioverter defibrillator
(ICD) therapy. Many interacting factors influence
the success of ventricular defibrillation in hu-
mans using ICDs. These can be broadly grouped
as the ICD system, implantation technique, and
patient-specific variables. Of these, the ICD sys-
tem can be controlled most precisely. Recently,
intense interest has focused on the features of the
defibrillating waveform. It is well established
that biphasic waveforms reduce defibrillation
energy requirements versus monophasic wave-
Supported by a research grant from Medtronic, Inc., Min-
neapolis, Minnesota.
Address for reprints: Michael O. Sweeney, M.D., Cardiovascu-
lar Division, Brigham and Women’s Hospital, 75 Francis St.,
Boston, MA 02115. Fax: (617) 732-7134; e-mail: mosweeney6
partners.org
Received March 7, 2000; revised June 14, 2000; accepted June
22, 2000.
consensus on the reason(s) for increased defibril-
lation efficacy using a biphasic waveform. Fur-
ther, if there is a single optimal biphasic wave-
form for all patients is unknown. Waveform tilt,
truncation, and pulse duration may all affect de-
fibrillation success.2–13 Tilt refers to the percent-
age decline in leading edge voltage at which the
waveform phase is truncated. Previous studies
have suggested that biphasic waveforms with a
40%–50% tilt in each phase yielded a lower de-
fibrilla tion threshold (DFT) energy versus a
waveform with 65% tilt in each phase, although
these conclusions may be dependent on the elec-
trode configuration and capacitance from which
the waveforms are derived.4,9 Currently, a three-
electrode configuration and a 90–110-mF capaci-
tance are most commonly used for human defib-
rillation using ICDs. The optimal tilt for this
combination of electrodes and capacitance is un-
known. This study prospectively compared
DFTs for 50%/50% (phase 1/phase 2) and
65%/65% tilt truncated exponential, biphasic
waveforms.

60 January 2001 PACE, Vol. 24

 EFFECT OF TILT ON BIPHASIC DEFIBRILLATION
Methods
The study population consisted of 60 patients
undergoing ICD implantation at five centers. All
patients gave written, informed consent according
to the rules and regulations of the Human Re-
search Committees at the investigative sites. Base-
line characteristics of the study population are
shown in Table I. A three-electrode pectoral
transvenous ICD system was used. In 59 patients,
a right ventricular (RV) sensing/pacing/defibrilla-
tion electrode was positioned with the tip at the
distal RV apex and a separate transvenous coil
electrode was positioned with the tip at the junc-
tion of the superior vena cava (SVC) and the right
atrium (Table II). In one patient, a single-pass dual
coil electrode with a fixed separation between the
distal and proximal coil was used (Table II). An
active can emulator (CAN) was placed in the left
mid-pectoral position. The RV coil served as the
defibrillation cathode (2) and the SVC coil plus
CAN served as the defibrillation anode (1). Defib-
rillation waveforms were derived from a
Medtronic (Minneapolis, MN, USA) Model 2934
external cardioverter defibrillator with 110-mF ca-
pacitance. Ventricular fibrillation induction and
termination testing was performed under tran-
sient light general anesthesia with intravenous
agents or general anesthesia with inhaled agents
and endotracheal intubation. Ventricular fibrilla-
tion was induced using a low energy monophasic
shock delivered between the RV coil and the SVC
coil/CAN and synchronized to the T wave during
ventricular pacing or 50-Hz burst pacing via the
Table I.
Clinical Characteristics of the Study Population
Characteristic N 5 60 Patients
Mean age, years 65 6 11
Male gender, %, (N) 90 (54/60)
Mean ejection fraction, % 32 6 15
Clinical arrhythmia, %, (N)
Aborted sudden cardiac death 32 (19/60)
Sustained VT 58 (35/60)
Nonsustained VT 10 (6/60)
Heart disease, %, (N)
Coronary artery disease 70 (42/60)
Nonischemic dilated CMP 27 (16/60)
Other 3 (2/60)
Antiarrhythmic drugs at implant,
%, (N)
Amiodarone 25 (15/60)
Class IA 3 (2/60)
Class IB 3 (2/60)
CMP 5 cardiomyopathy; VT 5 ventricular tachycardia
PACE, Vol. 24
Table II.
ICD Electrodes Used in the Study
ICD Electrodes N 5 60 Patients
Right ventricular endocardial, %, (N)
Medtronic Model 6932 52 (31/60)
Medtronic Model 6936 28 (17/60)
Ventritex Model RVO2 17 (10/60)
CPI Model 0125 2 (1/60)
Intermedics Model 497-19 2 (1/60)
Superior vena cava, %, (N)
Medtronic Model 6933 13 (8/60)
Medtronic Model 6937 67 (40/60)
Ventritex Model SVO2 17 (10/60)
CPI Model 0125 2 (1/60)
Intermedics Model 497-22 2 (1/60)
Pectoral active can emulator, %, (N)
Medtronic Model 7221 27 (16/60)
Medtronic Model 7223 73 (44/60)
Ventritex, Sunnyuale, CA, USA; CPI, Cardiac Pacemakers, Inc.,
St. Paul, MN, USA. ICD 5 implantable cardioverter defibrillator.
RV endocardial electrode. Ventricular fibrillation
was defined as rapid, irregular, disorganized sur-
face electrocardiographic activity accompanied by
a mean cycle length , 200 ms on simultaneous in-
tracardiac electrograms. Ventricular fibrillation
was required to persist for 10 seconds prior to de-
livery of a defibrillating pulse. DFT metrics (en-
ergy, voltage, current, and resistance) were deter-
mined for truncated exponential, biphasic
waveforms with 50%/50% versus 65%/65% tilts.
The waveform voltage and duration were mea-
sured directly by the test instrument. Resistance
was calculated using leading and trailing edge
voltage and equations of exponential decay, and
current was calculated using Ohm’s Law. For both
waveforms, phase1/phase 2 tilts were equal and
phase 2 leading edge voltage equaled phase 1 trail-
ing edge voltage. In each patient the stored energy
DFT, defined as the minimum stored energy that
successfully terminated ventricular fibrillation on
a single occasion, was prospectively determined
for each waveform in random order using a binary
search method (Fig. 1). For each episode of ven-
tricular fibrillation, only the first pulse was used
to determine the DFT. If the first pulse was unsuc-
cessful, testing continued with the next prescribed
energy level after restoration of sinus rhythm and
a 5-minute rest period.
Statistical Methods
To detect a difference of 15% on a mean
stored energy DFT of approximately 10 J, (a 5
0.05, 1-b 5 0.80) and assuming a standard devia-
January 2001 61
 SWEENEY, ET AL.
Figure 1. Binary search tree diagram used to determine
the defibrillation threshold in the clinical study.
tion of differences of 4 J, a sample size of 60 was
required. Differences between DFT metrics within
each patient for the two different biphasic wave-
form tilts were compared using the paired Stu-
dent’s t-test and McNemar’s test.
Results
DFT metrics for the 50%/50% and 65%/65%
tilt biphasic waveforms are shown in Table III.
Stored energy (Joules), voltage (volts), and current
(amperes) thresholds were significantly lower for
the 50%/50% tilt waveform. The stored energy
threshold was 15% lower for the 50%/50% versus
65%/65% tilt waveform (P 5 0.007). The current
DFT was 9% lower for the 50%/50% versus
65%/65% tilt waveform (P 5 0.002). The voltage
DFT was 8% lower for the 50%/50% versus
65%/65% tilt waveform (P 5 0.004). The average
phase 1/phase 2 pulse durations for the 50%/50%
tilt waveform were 3.23 6 0.58/3.35 6 0.54 ms
versus 4.82 6 0.87/4.99 6 0.85 ms for the
65%/65% tilt. Since tilt was specified for each
waveform phase, the second phase of each wave-
form was slightly longer than the first phase due to
Table III.
Defibrillation Threshold Metrics for 50%/50% versus 65%/65% Tilt Biphasic Waveforms
DFT Metric 50%/50% Tilt
Stored energy, Joules 9.2 6 5.7
Delivered energy, Joules 8.6 6 5.4
Delivered/stored energy, % 93.7 6 0.01
Voltage, Volts 391 6 118
Current, Amperes 10.9 6 4.0
Resistance, ohms 37.4 6 7.7
DFT 5 defibrillation threshold.
62 January 2001
the increased resistance seen with lower voltages
in the second phase.
The mean ratio of delivered to stored energy
at the DFT was 5% lower for the 50%/50% tilt
waveform versus the 65%/65% tilt waveform (P ,
0.001). The delivered energy threshold for the
50%/50% tilt waveform was also lower because a
50%/50% tilt waveform delivers a smaller per-
centage of the stored energy versus a 65%/65% tilt
waveform due to shorter pulse durations. Pathway
resistance at the stored energy DFT was slightly
higher for the 50%/50% tilt versus the 65%/65%
waveform (P 5 0.005), however, the difference
was small (difference between pairs 0.62 6 1.64).
Resistance at 9 J (the starting energy in each test
sequence for each patient) was not different be-
tween the 50%/50% and the 65%/65% tilt wave-
form (36.9 6 7.6 vs 36.6 6 7.4 V, P 5 0.16).
Thirty-two (53%) patients had a lower stored
energy DFT with the 50%/50% tilt waveform, 12
(20%) patients had a lower DFT with the
65%/65% tilt waveform, and 16 (27%) patients
had no measurable difference in DFT with either
waveform (P , 0.001). Nine (15%) of 60 patients
had a stored energy DFT $ 24 J. Three patients
were in the 50%/50% tilt group and six patients
were in the 65%/65% tilt group (5% vs 10%, re-
spectively, P 5 0.78). Two of nine patients with
stored energy DFT $24 J with either waveform
had a DFT equal to 18 J with the opposite wave-
form, and in both cases the failure at 24 J occurred
with the waveform randomized to second in the
testing sequence. This raised the possibility that
the higher stored energy DFTs with either wave-
form were due to the order of testing. To deter-
mine if the order of testing affected stored energy
DFT, the data were grouped and analyzed accord-
ing to whether it was collected first or second in
the randomization sequence. In 27 (45%) of 60 pa-
tients, the 50%/50% tilt waveform was tested first
and in 33 (55%) of 60 patients the 65%/65% tilt
waveform was tested first. The mean stored energy
DFTs for the waveforms that were tested first ver-
sus waveforms tested second were 9.8 6 5.7 J vs
65%65% Tilt D (%) P Value
10.8 6 6.4 15 0.007
10.6 6 6.3 19 0.001
98.3 6 0.01 5 , 0.001
424 6 128 8 0.004
12.0 6 4.4 9 0.002
36.7 6 7.5 2 0.005
PACE, Vol. 24
 EFFECT OF TILT ON BIPHASIC DEFIBRILLATION
10.1 6 6.5 J (P 5 0.62). This indicates that the or-
der of testing did not bias the stored energy DFT
for either waveform.
Discussion
This prospective, randomized comparison in
60 patients using a RV (2)/SVC plus CAN (1)
electrode configuration and a 110-mF capacitance
voltage source demonstrates that a 50%/50% tilt
biphasic waveform results in significantly lower
stored energy, voltage, and current DFTs versus a
65%/65% tilt biphasic waveform. Capacitance,
lead configuration, and testing protocol were held
constant within and between patients to minimize
the potentially confounding influence of these
variables on defibrillation efficacy. Overall, fewer
patients had a high stored energy DFT ($ 24 J)
with the 50%/50% tilt versus the 65%/65% tilt
waveform (5% vs 10%, respectively) although this
was not statistically significant.
The delivered energy for the 50%/50% tilt
waveform was less than the 65%/65% tilt wave-
form partly because of shorter average pulse dura-
tions. A 50%/50% tilt waveform delivers a lower
delivered percentage of stored energy than a
65%/65% tilt waveform (93.7% vs 98.3%, respec-
tively). Results also demonstrate that stored en-
ergy, current, and voltage DFTs were also reduced
by the 50%/50% tilt waveform.
Previous reports on the effect of biphasic
waveform tilt on defibrillation efficacy have been
conflicting. Swartz et al.4 reported a 31% reduc-
tion in the stored energy DFT using a 42%/42%
tilt versus a 65%/65% tilt biphasic waveform de-
livered from a 120-mF capacitor. That study used
two transvenous electrodes (RV and SVC or coro-
nary sinus) and a subscapular patch electrode. In
contrast, Poole et al.14 found no difference in de-
fibrillation efficacy between a 50%/50% tilt ver-
sus a 65%/65% tilt biphasic waveform delivered
from a 120-mF capacitor using a unipolar high
voltage configuration (RV/CAN). Similarly, Li et
al.15 found no difference in defibrillation efficacy
between 50%/50% versus 65%/65% tilt biphasic
waveforms using two transvenous electrodes
(RV/SVC). Natale et al.9 observed a reduction in
biphasic DFT with a three-electrode configuration
when the tilt was decreased from 65%/65% to
50%/50%, but no further reduction when tilt was
decreased to 40%/40%. The disparate results of
these studies may be related to differences in elec-
trode configurations, capacitances, DFT testing
techniques, and sample size. Prior research has
demonstrated, for example, that the increased de-
fibrillation efficacy of biphasic versus monopha-
sic waveforms is influenced by electrode configu-
ration. 16 Further, the optimal phase 1 duration
and capacitance may vary with electrode configu-
PACE, Vol. 24 January 2001
ration and pathway resistance.10,17 Perhaps most
importantly, because these conflicting studies
used a much smaller sample size (9–19 patients)
than the current study, they would have had a low
probability of detecting 10%–15% differences in
defibrillation efficacy.
The reason for the increased defibrillation ef-
ficacy of the 50%/50% fixed tilt waveform versus
the 65%/65% fixed tilt waveform in our study is
unknown. The increased efficacy of the 50%/50%
tilt waveform may be related to better optimiza-
tion of pulse duration, higher negative phase peak
voltage, or both, as both waveform parameters are
affected by altering tilt.
The Weiss-Lapicque relationship for cardiac
tissue stimulation using rectangular pulses pre-
dicts a minimum mean current or rheobase for
stimulation and a chronaxie, which is the stimu-
lation pulse duration at twice rheobase cur-
rent.18,19 This relationship produces a U-shaped
curve for energy as a function of pulse duration as
has been demonstrated with cardiac pacing. The
lowest energy requirement for electrical stimula-
tion of the heart using rectangular waveforms,
therefore, occurs when the pulse duration approx-
imates the chronaxie time.20 Animal and human
studies have shown a similar strength-duration re-
lationship for defibrillation using a monophasic
truncated exponential waveform17,21,22 and phase
1 of biphasic truncated exponential defibrilla-
tion. 11,12,17,23 A pulse duration between 2 and 4
ms has been observed to yield a defibrillation en-
ergy minima for monophasic shocks and phase 1
of biphasic shocks.11,12,17 Quantitative models of
defibrillation based on this strength-duration rela-
tionship recite a necessary balance between the
exponential-decay time constant of the capacitive
discharge, pulse duration, and chronaxie
time.5–8,17
To explain the increased efficacy of the
50%/50% tilt waveform versus the 65%/65% tilt
waveform on the basis of pulse durations, several
reasonable but unproven assumptions must be
made. First, that phase 1 of a biphasic shock acts
as does a monophasic shock,6,24 and that the prin-
cipal effect of phase 1 is to capture the my-
ocardium, whereas phase 2 “burps” the residual
charge left on the cell membrane by phase 1.6 Sec-
ond, that monophasic shocks follow a Weiss-
Lapicque relationship. If those assumptions hold
true, then the increased efficacy of the 50%/50%
tilt waveform might be due to phase 1 pulse dura-
tions being nearer the chronaxie time. Animal
models have shown a monophasic defibrillation
chronaxie time between 2 and 4 ms, and studies in
humans have suggested a chronaxie time in a sim-
ilar range.17,25,26
Numerical models of defibrillation with trun-
63
 SWEENEY, ET AL.
cated exponential waveforms indicate a linkage
between the exponential-decay time constant of
the capacitive discharge, the time-response pa-
rameter of cardiac myocytes to the shock, and
pulse duration.5–8,17,27–30 The time constant of the
capacitive discharge, or ts, is the product of path-
way resistance and capacitance. In the hyperbolic
average current model5,6 the time-response pa-
rameter of cardiac myocytes is the chronaxie time
tc (as defined by the Weiss-Lapicque relationship).
In the leading edge-current exponential
model,7,8,27–30 the time-response parameter is re-
ferred to as the membrane time constant tm . tc and
tm are assumed to be related although this rela-
tionship is inexact and may vary by waveform.
DFT is predicted to be minimized when ts ap-
proximates the cardiac myocyte time-response pa-
rameter (either tc or tm ) in both models.19 Because
pathway resistance was nearly equal and output
capacitance was identical between waveforms,
the difference in efficacy in the current study can-
not have been due to a difference in ts . Using equa-
tions from these models6,8,17 and time-response
parameters derived from human strength-duration
curves using transvenous leads,17 we calculated
the optimal first pulse duration and tilt using the
values of capacitance and average resistance mea-
sured in this study. The leading edge current ex-
ponential model predicted that 50% tilt and 3.19
ms would be optimal, while the average current
hyperbolic and exponential models predicted op-
timal values of 64% and 65% for tilt and 4.73 ms
and 4.80 ms for pulse duration, respectively.
In our study, the average phase 1/phase 2
pulse durations for the 50%/50% tilt waveform
were 3.23 6 0.58/3.35 6 0.54 ms versus 4.82 6
0.87/4.99 6 0.85 ms for the 65%/65% tilt. Thus,
our data are in agreement with theoretical predic-
tions of the leading edge exponential model.
There are limitations to these models in pre-
dicting the response to various pulse tilts, how-
ever. Despite one model predicting exactly which
tilt would have the lower DFT in our study of one
discharge time constant, these models are all inac-
curate in predicting behavior across time con-
stants. Each model predicts that when the system
time constant, ts, decreases the optimal tilt should
increase. Studies presented in the literature did
not show a difference between 50% or 65% tilt us-
ing two-electrode systems. However, when using
three-electrode systems, which have a decreased
resistance (and therefore decreased ts ) compared
with two-electrode systems, the lower DFT was
associated with the lower tilt, not the higher tilt
value predicted by the models. We further in-
spected our data to determine if individual patient
resistances could explain differences observed in
our study. The mean resistance for patients show-
64 January 2001
ing lower DFT with 50% tilt, no difference, or
lower DFT with 65% tilt was 38 6 9, 38 6 6, and
35 6 5 V, respectively (P . 0.05). Thus, even with
the current study, the interpatient variability of re-
sistance does not explain the difference in tilt,
though this is perhaps confounded by the inaccu-
racy of a crude DFT estimation in one patient.
If there is a single optimal pulse duration for
defibrillation using a capacitive discharge wave-
form is unproven. This reflects the fact that the
strength-duration relationship for transvenous de-
fibrillation in humans is incompletely character-
ized. At least one study in humans appeared to
show that the transvenous monophasic DFT volt-
age and stored energy declined exponentially with
increasing pulse durations but were insensitive to
pulse durations . 6 ms.25 This observation is
more compatible with the quantitative model de-
veloped by Blair27,28 that describes the heart mus-
cle membrane as a capacitor and resistor in paral-
lel and predicts that DFTs are insensitive to
altering pulse durations over a wide range.
Higher peak negative phase voltage may also
have contributed to the increased defibrillation ef-
ficacy of the 50%/50% tilt waveform. Tomassoni
et al.31 demonstrated that for three commercially
available biphasic waveforms the waveform with
the lowest negative phase peak voltage and
longest pulse duration resulted in the highest de-
fibrillation energy requirement. Further, when
phase 1 trailing edge/phase 2 leading edge volt-
ages were held equal and phase 1 tilt was fixed at
65%, a shorter pulse duration in phase 2 did not
reduce defibrillation efficacy. However, reducing
phase 2 leading edge voltage to 1/2 phase 1 trailing
edge voltage, while holding phase1/phase 2 tilt
fixed at 65% and phase 1/phase 2 pulse duration
equal, resulted in significantly higher DFTs. They
concluded that negative phase peak voltage was
more important for defibrillation efficacy than
phase 2 pulse duration. In the present study,
phase 1 trailing edge and phase 2 leading edge
voltages were equal for both waveforms. However,
starting with any phase 1 leading edge voltage, the
phase 2 leading edge voltage will be relatively
higher for the 50%/50% versus 65%/65% tilt.
Study Limitations
The principal methodological limitation of
this study is that we did not determine strength-
duration curves for defibrillation. This is imprac-
tical in the context of ICD implantation in humans
as it would mandate an excessive, and potentially
deleterious, number of ventricular fibrillation in-
ductions and terminations. The wide energy steps
in our DFT search protocol may have been inade-
quate to resolve small but important differences
between waveforms. Although single shock DFTs
PACE, Vol. 24
 EFFECT OF TILT ON BIPHASIC DEFIBRILLATION
are useful for characterizing clinically relevant as-
pects of defibrillation efficacy, the probabilistic
nature of defibrillation is not well addressed by
this technique.32 This may partly explain the vari-
able effect of tilt on defibrillation efficacy between
patients in our study. Although slightly more than
one-half the patients had a lower DFT with the
50%/50% tilt waveform, one-fifth had a lower
DFT with the 65%/65% tilt waveform and one-
quarter had no measurable difference in DFT with
either waveform.
References
1. Schuder JC, McDaniel WC, Stoeckle H. Defibrillation of 100 kg
calves with asymmetrical, bidirectional, rectangular pulses. Car-
diovasc Res 1984; 18:419–426.
2. Bardy GH, Ivey TD, Johnson G, et al. A prospective randomized
evaluation of biphasic versus monophasic waveform pulses on
defibrillation efficacy in humans. J Am Coll Cardiol 1989; 14:
728–733.
3. Schuder JC, Stoeckle H, West JA, et al. Transthoracic ventricular
defibrillation in the dog with truncated and untruncated exponen-
tial stimuli. IEEE Trans Biomed Eng 1971; 18:410–415.
4. Swartz JF, Fletcher RD, Karasik PE. Optimization of biphasic
waveforms for human nonthoracotomy defibrillation. Circulation
1993; 88:2646–2654.
5. Irnich W. Optimal truncation of defibrillation pulses. PACE 1995;
18:673–688.
6. Kroll MW. A minimal model of the monophasic defibrillation
pulse. PACE 1993; 16:769–777.
7. Kroll MW. A minimal model of the single capacitor biphasic wave-
form. PACE 1994; 17:1782–1792.
8. Walcott GP, Walker RG, Cates AW, et al. Choosing the optimal
monophasic and biphasic waveforms for ventricular defibrillation.
J Cardiovasc Electrophysiol 1995; 6:737–750.
9. Natale A, Sra J, Krum D, et al. Relative efficacy of different tilts
with biphasic defibrillation in humans. PACE 1996; 19:197–206.
10. Swerdlow CD, Kass RM, Chen PS, et al. Effect of capacitor size and
pathway resistance on defibrillation threshold for implantable de-
fibrillators. Circulation 1994; 90:1840–1846.
11. Schauerte PN, Schondube FA, Grossman M, et al. Influence of
phase duration of biphasic waveforms on defibrillation energy
requirements with a 70-uF capacitance. Circulation 1998; 97:
2073–2078.
12. Schauerte PN, Ziegert K, Waldmann M, et al. Effect of biphasic
shock duration on defibrillation threshold with different electrode
configurations and phase 2 capacitances. Prediction by upper-
limit-of-vulnerabilit y determination. Circulation 1999; 99:
1516–1522.
13. Tang ASL, Yabe S, Wharton JM, et al. Ventricular defibrillation us-
ing biphasic waveforms: The importance of phasic duration. J Am
Coll Cardiol 1989; 13:207–214.
14. Poole JE, Bardy GH, Kudenchuk PJ, et al. Prospective randomized
comparison of biphasic waveform tilt using a unipolar defibrilla-
tion system. PACE 1995; 18:1369–1373.
15. Li HG, Yee R, Mehra R, et al. The effect of biphasic shock wave-
form tilt upon defibrillation efficacy. (abstract) Circulation 1993;
88:I-113.
PACE, Vol. 24 January 2001
Conclusions
For the most common three-electrode config-
uration and voltage source capacitance in clinical
use a 50%/50% tilt biphasic waveform yields, on
average, lower defibrillation stored energy re-
quirements than a 65%/65% tilt waveform. This
difference is probably due to a closer to optimal
pulse duration (of phase 1 in particular) for the ca-
pacitance and typical pathway resistance encoun-
tered in the study population.
16. Bardy GH, Troutman C, Johnson G, et al. Electrode system influ-
ence on biphasic waveform defibrillation efficacy in humans. Cir-
culation 1991; 84:665–671.
17. Swerdlow CD, Brewer JE, Kass RM, et al. Application of models of
defibrillation to human defibrillation data: Implications for opti-
mizing implantable defibrillator capacitance. Circulation 1997;
96:2813–2822.
18. Weiss G. Sur la possibilite de render comparable entre eux les ap-
pareils survant a l’excitation electrique. Arch Ital Biol 1901;
35:413–446.
19. Lapicque L. Definition experimentalle de l’excitabilite. Proc Soc
Biol 1909; 77:280–285.
20. Pearce JA, Bourland JD, Nielsen W, et al. Myocardial stimulation
with ultrashort duration current pulses. PACE 1982; 5:52–58.
21. Bourland JD, Tacker WA, Geddes LA. Strength-duration curves for
trapezoidal waveforms of various tilts for transchest defibrillation
in animals. Med Instrum 1978; 12:38–41.
22. Chapman PD, Wetherbee JN, Vetter JW, et al. Strength-duration
curves of fixed pulse width variable tilt truncated exponential
waveforms for nonthoracotomy internal defibrillation in dogs.
PACE 1988; 11:1045–1050.
23. Feeser SA, Tang ASL, Kavanagh KM, et al. Strength-duration and
probability of success curves for defibrillation with biphasic wave-
forms. Circulation 1990; 82:2128–2141.
24. Hillsley RE, Walker RG, Swanson DK, et al. Is the second phase of
a biphasic defibrillation waveform the defibrillating phase? PACE
1993; 16:1401–1411.
25. Gold MR, Shorofsky SR. Strength-duration relationship for human
transvenous defibrillation. Circulation 1997; 96:3517–3520.
26. Shorofsky SR, Foster AH, Gold MR. Effect of waveform tilt on de-
fibrillation thresholds in humans. J Cardiovasc Electrophysiol
1997; 8:496–501.
27. Blair H. On the intensity-time relations for stimulation by electric
currrents, I. J Gen Physiol 1932; 15:709–729.
28. Blair H. On the intensity-time relations for stimulation by electric
currrents, II. J Gen Physiol 1932; 15:731–755.
29. Cleland B. A conceptual basis for pacing waveforms. PACE 1996;
19:1177–1185.
30. Cleland B. A conceptual basis for defibrillation waveforms. PACE
1996; 19:1186–1195.
31. Tomassoni G, Newby K, Deshpande S, et al. Defibrillation efficacy
of commercially available biphasic impulses in humans. Circula-
tion 1997; 95:1822–1826.
32. McDaniel MC, Schuder JC. The cardiac ventricular defibrillation
threshold: Inherent limitations in its application and interpreta-
tion. Med Instrum 1987; 21:170–176.
65