Full text of DOI:10.1080/10903120190940029

PHYSICIAN INTERPRETATION AND QUANTITATIVE MEASURES OF
ELECTROCARDIOGRAPHIC VENTRICULAR FIBRILLATION WAVEFORM
Christopher B. Lightfoot, BSc, Thomas J. Sorensen, MD, Michael D. Garfinkel, MD,
Lawrence D. Sherman, MD, Clifton W. Callaway, MD, PhD, James J. Menegazzi, PhD
trical defibrillation or cardioversion is indicated, are
routinely implanted in patients. Furthermore, auto-
mated external defibrillators (AEDs) are now being
deployed for use by laypersons on unconscious sub-
jects.1 However, the qualitative discrimination of
unorganized rhythms such as ventricular fibrillation
(VF) from organized supraventricular rhythms may
ignore valuable information in the ECG waveform,
which may have implications for the prognosis of
treatment decisions.2
The morphology of the VF waveform has been asso-
ciated with different physiological states of the heart.
In animals, the amplitude of VF declines over time.3–5
Furthermore, the frequency characteristics of induced
VF follow a predictable pattern during ischemia.6–12 In
ABSTRACT
Objectives. The characteristics of the ventricular fibrillation
(VF) waveform may influence treatment decisions and the
likelihood of therapeutic success. However, assessment of
VF as being fine or coarse and the distinction between fine
VF and asystole are largely subjective. The authors sought to
determine the level of agreement among physicians for
interpretation of varying VF waveforms, and to compare
these subjective interpretations with quantitative measures.
Methods. Six-second segments of waveform from LIFEPAK
300 units were collected. Fifty segments, including 45 VF
and five ventricular tachycardia (VT) distracters, were
graphed to simulate rhythm strips. These waveforms were
quantitatively described using scaling exponent, root-mean-
squared amplitude, and centroid frequency. Thirty-two
emergency medicine residents were asked to interpret the
arrhythmias as VT, “coarse” VF, “fine” VF, or asystole. Their
responses were compared with the qantitative measures.
Interphysician agreement was assessed with the kappa sta-
tistic. Results. One thousand four hundred forty interpreta-
tions were analyzed. There was fair agreement between
physicians about the classification of arrhythmias (κ = 0.39).
Mean values associated with coarse VF, fine VF, and asys-
tole differed in all three quantitative measure categories. The
decision whether to defibrillate was highly correlated with
the distinction between VF and asystole (Pearson chi-square
= 1,170.40, df = 1, p[two-sided] < 0.001). Conclusions. With
only fair agreement on the threshold of fine VF and asystole,
defibrillation decisions are largely subjective and caregiver-
specific. These data suggest that quantitative measures of
the VF waveform could augment the current standard of
subjective classification of VF by emergency care providers.
Key words: cardiac arrest; heart arrest; ventricular fibrilla-
tion; asystole; waveform; scaling exponent; amplitude; cen-
troid frequency.
PREHOSPITAL EMERGENCY CARE 2001;5:147–154
Automated analysis of electrocardiographic (ECG)
signals is well established. Discrimination of different
rhythms is sufficiently reliable that automatic internal
defibrillators, that can detect rhythms for which elec-
Received July 12, 2000, from the Department of Emergency Medicine,
University of Pittsburgh (CBL, TJS, MDG, LDS, CWC, JJM),
Pittsburgh, Pennsylvania. Revision received December 8, 2000;
accepted for publication December 8, 2000.
Supported by the Center of Excellence Grant from the Emergency
Medicine Foundation, Dallas, Texas.
Address correspondence and reprint requests to: James J. Menegazzi,
PhD, Department of Emergency Medicine, 230 McKee Place, Suite
500, Pittsburgh, PA 15213. e-mail: <menegazz+@pitt.edu>.
FIGURE 1. Example of the variability found within the segments of
waveform. Five panels of waveform represent the visible changes
from coarse ventricular fibrillation to asystole.
147

148
PREHOSPITAL EMERGENCY CARE APRIL/JUNE 2001 VOLUME 5 / NUMBER 2
TABLE 1. Guidelines for Classifying Kappa
The widespread distribution of AEDs and advisory
defibrillators suggests that machine-based quantita-
tive ECG measures may soon be available in routine
clinical practice. Rhythm interpretations by para-
medics and other health care providers have shown to
be highly variable.17–19 However, the extent to which
quantitative measures should influence caregiver
judgment is unknown. Therefore, we sought to assess
the level of agreement between different physicians
for classifying VF as coarse, fine, or asystole and for
deciding whether to defibrillate. As a secondary aim,
we compared the physicians’ interpretations with
three different ECG measures—amplitude, centroid
frequency, and scaling exponent. We hypothesized
that quantitative measures derived from the VF wave-
form could augment the physician discriminations
between coarse VF, fine VF, and asystole.
κ Value
Level of Agreement
<0.00
Poor
Slight
Fair
Moderate
Substantial
Almost perfect
0.00–0.20
0.21–0.40
0.41–0.60
0.61–0.80
0.81–1.00
humans, fewer systematic data are available.
However, common clinical experience also suggests
that VF amplitude declines over time and lower dom-
inant frequencies of VF are associated with more pro-
longed ischemia.2,12,13 Deciding whether to defibrillate
VF is dependent on caregivers’ making the distinction
between VF and asystole. The probability of success-
ful outcome may be related to the physiological state
of the heart. Because the decrement of ECG amplitude
from coarse VF to asystole is continuous, distinguish-
ing fine VF from asystole is arbitrary.14,15 This discre-
tionary distinction is seen clinically and may affect
patient treatment and outcome.16 Therefore, describ-
ing the quality of the VF waveform may augment
research and clinical care by stratifying patients into
different prognostic groups and by more precisely
defining asystole.
METHODS
This study was approved by the Biomedical
Institutional Review Board of the University of
Pittsburgh. Thirty-two resident physicians from the
University of Pittsburgh Affiliated Residency in
Emergency Medicine program gave voluntary
informed consent to participate in this study. The sub-
jects included residents in their first, second, and third
FIGURE 2. Physician classification of ventricular fibrillation (VF) waveform versus scaling exponent. The percentage of respondents classifying
rhythms strips as ventricular tachycardia (VT), coarse VF, fine VF, or asystole is plotted versus the values of the scaling exponent. Five strips
were rated by 32 physicians within each bin.

Lightfoot et al. INTERPRETATION OF ECG VF WAVEFORM
149
TABLE 2. Quantitative Measures of Ventricular Fibrillation (VF) Waveform*
Scaling Exponent
Amplitude (mV)
Centroid Frequency (Hz)
Coarse VF
Fine VF
Asystole
Defibrillate
1.333 ± 0.166 (1.318, 1.348)
1.610 ± 0.170 (1.596, 1.624)
1.870 ± 0.118 (1.856, 1.883)
1.464 ± 0.226 (1.451, 1.478)
0.160 ± 0.085 (0.152, 0.168)
0.073 ± 0.026 (0.071, 0.075)
0.062 ± 0.022 (0.059, 0.064)
0.122 ± 0.083 (0.117, 0.127)
4.845 ± 0.804 (4.772, 4.917)
5.230 ± 1.191 (5.132, 5.327)
6.732 ± 2.025 (6.500, 6.965)
5.055 ± 1.066 (4.993, 5.118)
Intravenous medication and
continue cardiopulmonary resuscitation
1.845 ± 0.146 (1.828, 1.861)
0.063 ± 0.025 (0.060, 0.066)
6.582 ± 2.007 (6.359, 6.804)
*Mean ± standard deviation (95% confidence interval).
years. At the time of the study, they were all currently
certified in American Heart Association Advanced
Cardiac Life Support (ACLS).
described in detail in previous publications and
included the scaling exponent,23 a mathematical meas-
ure of the amount of two-dimensional space the wave-
form fills; root-mean-square (RMS) amplitude, the
average vertical deflection of the waveform; and cen-
troid frequency,2 a value derived from a fast Fourier
transform, which is related to the peak-to-peak
appearance. Epochs were selected such that they
included the entire range of scaling exponents from
1.1 to 2.0. Scaling exponent values between 1.0 and 1.1
had not been encountered in our human VF record-
ings. Six-second epochs of waveform were graphed to
simulate rhythm strips with a scale of 25 mm/second
and 10 mm/mV. Figure 1 is representative of the visu-
al variability of waveforms in our study.
We conducted a prospective study in which subjects
independently interpreted ECG strips in random
order. These subjects were presented 45 segments of
VF and five ventricular tachycardia (VT) distracters.
Electrocardiographic data were obtained from AED
recordings from a police first-responder program, as
has been previously described in full.20–22 The data
were recorded on analog tapes by a LIFEPAK 300 AED
(Physio-Control, Redmond, WA) and then digitized at
400 points/second with an analog/digital converter
and software (PowerLab, AD Instruments, Castle Hill,
Australia). Quantitative descriptors of the VF seg-
ments were calculated. These descriptors have been
The 50 strips were assembled in a randomized order
FIGURE 3. Physician classification of ventricular fibrillation (VF) waveform versus root-mean-squared (RMS) amplitude. The percentage of
respondents classifying rhythms strips as ventricular tachycardia (VT), coarse VF, fine VF, or asystole is plotted versus the values of the ampli-
tude. Five strips were rated by 32 physicians within each bin.

150
PREHOSPITAL EMERGENCY CARE APRIL/JUNE 2001 VOLUME 5 / NUMBER 2
FIGURE 4. Physician classification of ventricular fibrillation (VF) waveform versus centroid frequency. The percentage of respondents classi-
fying rhythms strips as ventricular tachycardia (VT), coarse VF, fine VF, or asystole is plotted versus the values of the centroid frequency. Five
strips were rated by 32 physicians within each bin.
FIGURE 5. Recommended treatment versus scaling exponent. The percentage of respondents recommending immediate defibrillation or car-
dioplmonary resuscitation (CPR)/drug administration is plotted versus the values of the scaling exponent.

Lightfoot et al. INTERPRETATION OF ECG VF WAVEFORM
151
into booklets and handed out to each physician. The
participants were separated so as to reduce the possi-
bility of influencing each other’s responses. They were
not permitted to consult with each other, and there
was no time limit for their completing the task. Each
page contained one ECG strip and the following
script: “Your patient is adult, has no pulse, and CPR is
ongoing. The patient is intubated and has good IV
access. The rhythm you see is the same in all leads,
and your equipment is functioning properly. This is:
VT, coarse VF, fine VF, or asystole? Your next action
is: defibrillate, or give IV medication and continue
CPR?”
Interrater reliability of our samples was described
with kappa statistics. We found variability in the liter-
ature as to the interpretation of the level of agreement
when using kappa; our analysis was based on the cri-
teria found in Table 1.24 Mean values of the scaling
exponent, amplitude, and centroid frequency for seg-
ments classified as coarse VF, fine VF, and asystole
were determined. In order to determine whether the
quantitative measures were different for the various
rhythm classifications, mean values were compared
by analysis of variance (ANOVA) with a significance
level of p < 0.05. A correlational chi-square test was
performed to examine the relation between classifying
a segment of waveform as VF or asystole and making
a treatment decision. Analyses were conducted with
SAS (version 6.12, SAS Institute Inc., Cary, NC) and
SPSS (version 6.1.1, SPSS Inc., Chicago, IL).
RESULTS
With 32 raters interpreting 45 segments of VF wave-
form, there were a total of 1,440 classifications. Of
these, 79 were categorized as VT, 473 were categorized
as coarse VF, 577 were categorized as fine VF, and 294
were categorized as asystole. Several answers were
missed in the questionnaires (1.2%); thus, interrater
agreement statistics were based on completed book-
lets from 29 raters. All other observations utilized
answers from each of the 32 physicians.
There was only fair agreement between physicians
in classifying VF. The kappa statistic for overall inter-
rater reliability showed fair reproducibility (κ = 0.39).
The agreement on asystole was moderate (κ = 0.56);
whereas, on coarse and fine VF it was fair (κ = 0.39 and
0.32). Agreement was fair (κ = 0.40) for all VF when
coarse and fine VF were combined, post-hoc, into a
single classification. Reducing classifications to only
three categories (VT, VF, and asystole) led to a moder-
ate overall agreement (κ = 0.44). Physician agreement
to defibrillate had an overall κ value of 0.49, indicating
moderate reproducibility for treatment decisions.
The 45 segments of waveform ranged from 1.11 to
1.98 in scaling exponent, 0.04 to 0.40 mV in amplitude,
FIGURE 6. Recommended treatment versus root-mean-squared (RMS) amplitude. The percentage of respondents recommending immediate
defibrillation or cardiopulmonary resuscitation (CPR)/drug administration is plotted versus the values of the RMS amplitude.

152
PREHOSPITAL EMERGENCY CARE APRIL/JUNE 2001 VOLUME 5 / NUMBER 2
FIGURE 7. Recommended treatment versus centroid frequency. The percentage of respondents recommending immediate defibrillation or car-
diopulmonary resuscitation (CPR)/drug administration is plotted versus the values of the centroid frequency.
and 3.54 to 10.44 Hz in centroid frequency. Mean val-
ues were 1.55 ± 0.27 for the scaling exponent, 0.11 ±
0.01 mV for amplitude, and 5.40 ± 1.50 Hz for centroid
frequency. Using each of the 1,423 separate responses,
mean values of coarse VF, fine VF, and asystole dif-
fered in the scaling exponent (p[F = 1,069.88, df = 2,
1,344] < 0.001), amplitude (p[F = 433.12, df = 2, 1,344]
< 0.001), and centroid frequency (p[F = 197.23, df = 2,
1,344] < 0.001) (Table 2). Decisions to defibrillate were
highly correlated with the distinction between VF and
asystole (Pearson chi-square = 1,170.40, df = 1, p[two-
sided] < 0.001).
greater precision in classifying VF waveform, and
may assist clinicians in classification for treatment
decisions or research.
There is no clear cutoff between fine VF and asys-
tole. In animals with induced VF, amplitude declines
and scaling exponent increases continuously with
time. In the present study, the definition of asystole as
an ECG amplitude (<100 µV)15 would classify many
cases that physicians identified as fine VF, as asystole
(Fig. 3). Automated external defibrillators use multi-
tep algorithms to classify the electrical activity of the
heart. In one study, the threshold amplitude for rec-
ognizing asystole varied widely between devices.25 In
contrast, a high value of the scaling exponent was
closely associated with classification of the rhythm as
asystole (Fig. 2). Using amplitude measures alone as
the criterion for treatment may not be as effective as
the scaling exponent when accounting for the physi-
cian agreement about cases of asystole.
The relationship between interpretation and quanti-
tative measures are presented in Figures 2–4. The rela-
tionship between quantitative measures and treat-
ment decisions are presented in Figures 5–7.
DISCUSSION
This study determined that interrater agreement
among physicians for classification of VF as coarse
and fine was only fair. For extremes of waveform
morphology (Fig. 2, scaling exponent >1.9), interrater
agreement by physicians about waveform was better.
However, for VF waveforms with intermediate values
of amplitude, centroid frequency, and scaling expo-
nent, large variability was observed for rhythm inter-
pretation. Thus, quantitative measures provide
The optimal treatment of early VF may differ from
the optimal treatment for prolonged VF.4,13,26–30 While
immediate defibrillation has been suggested for all
VF,31 reperfusion and reoxygenation of the heart prior
to defibrillation may improve resuscitation after pro-
longed ischemia.13,26,27,32–38 Failed defibrillation
attempts should be avoided since they are known to
be deleterious to the heart and other surrounding tis-
sues.5,15,26,36,38–47 Therefore, the ability to distinguish

Lightfoot et al. INTERPRETATION OF ECG VF WAVEFORM
153
ventricular fibrillation of varied aetiology. Eur Heart J.
1990;11:173-81.
8. Brown CG, Dzwonczyk R, Werman HA, Hamlin RL. Estimating
the duration of ventricular fibrillation. Ann Emerg Med.
1989;18:1181-5.
9. Brown CG, Griffith RF, Ligten PV, et al. Median frequency—a
new parameter for predicting defibrillation success rate. Ann
Emerg Med. 1991;20:787-9.
10. Brown CG, Dzwonczyk R, Martin DR. Physiologic measure-
ment of the ventricular fibrillation ECG signal: estimating the
duration of ventricular fibrillation. Ann Emerg Med. 1993;22:70-
4.
11. Martin G, Cosin J, Such M, et al. Relation between power spec-
trum time course during ventricular fibrillation and electro-
mechanical dissociation: effects of coronary perfusion and
nifedipine. Eur Heart J. 1986;7:560-69.
12. Martin DR, Brown CG, Dzwonczyk R. Frequency analysis of the
human and swine electrocardiogram during ventricular fibrilla-
tion. Resuscitation. 1991;22:85-91.
13. Weaver WD, Cobb LA, Dennis D, Ray R, Hallstrom AP, Copass
MK. Amplitude of ventricular fibrillation waveform and out-
come after cardiac arrest. Ann Intern Med. 1985;102:53-5.
14. Callaham M, Braun O, Valentine W, Clark DM, Zegans C.
Prehospital cardiac arrest treated by urban first-responders:
profile of patient response and prediction of outcome by ven-
tricular fibrillation waveform. Ann Emerg Med. 1993;22:1664-
77.
VF that has a high probability of responding to defib-
rillation from VF that has a low likelihood of success
may have implications for immediate therapy.
Interestingly, physicians chose to defibrillate imme-
diately most cases identified as fine VF. The quantita-
tive measures for these cases are associated with a low
likelihood of successful defibrillation within this same
data set.48 Thus, the use of quantitative measures
could augment treatment decisions.
Although our waveform segments had an even dis-
tribution over the full range of scaling exponent val-
ues, a limitation to our analysis was many segments
had a low RMS amplitude. This could lead to an
increase in ambiguous interpretations and thus
decrease our kappa values. Our study investigated
one small sample within the health care field. It is
therefore not possible to generalize our findings to all
caregivers. However, with the fact that these emer-
gency residents were trained in ACLS by the same
community training center and were part of the same
residency training program, homogeneity should bias
the study for increased agreement. We speculate that
future studies with more heterogeneous samples
would find less robust agreement.
15. Gliner BE, White RD. Electrocardiographic evaluation of defib-
rillation shocks delivered to out-of-hospital sudden cardiac
arrest patients. Resuscitation. 1999;41:133-44.
16. Amaya SC, Langsam A. Ultrasound detection of ventricular fib-
rillation disguised as asystole. Ann Emerg Med. 1999;33:344-6.
17. Pirrallo RG, Swor RA, Maio RF. Inter-rater agreement of para-
medic rhythm labeling. Ann Emerg Med. 1993;22:1684-7.
18. Westdorp EJ, Gratton MC, Watson WA. Emergency department
interpretation of electrocardiograms. Ann Emerg Med. 1992;21:
541-4.
19. Pozen M, D’Agostino RB, Stykowski PA, et al. Effectiveness of a
pre-hospital medical control system: an analysis of the interac-
tion between emergency room physician and paramedic.
Circulation. 1981;63:442-7.
20. Davis EA, Mosesso VN. Performance of police first responders
in utilizing automated external defibrillation on victims of sud-
den cardiac arrest. Prehosp Emerg Care. 1998;2:101-7.
21. Davis EA, McCrorry J, Mosesso VN Jr. Institution of a police
automated external defibrillation program: concepts and prac-
tice. Prehosp Emerg Care. 1999;3:60-5.
22. Mosesso VN Jr, Davis EA, Auble TE, Paris PM, Yealy DM. Use
of automated external defibrillators by police officers for treat-
ment of out-of-hospital cardiac arrest. Ann Emerg Med.
1998;32:200-7.
CONCLUSIONS
There was only fair interrater agreement between
physicians about the classification of VF waveform.
Treatment decisions are normally made on the dis-
tinction of VF from asystole. The subjectivity of wave-
form classification could be circumvented by the use
of quantitative measures of waveform morphology.
Quantitative ECG may improve precision in the selec-
tion of immediate therapy.
The authors thank Dr. Thomas E. Auble and Dr. Margaret Hsieh for
their assistance with the statistical analyses.
References
1. Clinton W. Memorandum on automated external defibrillators
in federal buildings. Weekly Compilation of Presidential
Documents. 2000;36:1177-8.
23. Callaway CW, Sherman LD, Scheatzle MD, Menegazzi JJ.
Scaling structure of electrocardiographic waveform during pro-
longed ventricular fibrillation in swine. Pacing Clin
Electrophysiol. 2000;23:180-91.
2. Brown CG, Dzwonczyk R. Signal analysis of the human electro-
cardiogram during ventricular fibrillation: frequency and
amplitude parameters as predictors of successful countershock.
Ann Emerg Med. 1996;27:184-8.
24. Landis JR, Koch GG. The measurement of observer agreement
for categorical data. Biometrics. 1977;33:159-74.
3. Jones DL, Klein GJ. Ventricular fibrillation: the importance of
being coarse? J Electrocardiol. 1984;17:393-9.
25. Clifford AC. Comparative assessment of shockable ECG
rhythm detection algorithms in automated external defibrilla-
tors. Resuscitation. 1996;32:217-25.
4. Reich H, Angelos M, Safar P, et al. Cardiac resuscitability with
cardiopulmonary bypass after increasing ventricular fibrillation
ties in dogs. Ann Emerg Med. 1990;19:887-90.
26. Niemann JT, Cairns CB, Sharma J, Lewis RJ. Treatment of pro-
longed ventricular fibrillation: immediate countershock versus
high-dose epinephrine and CPR preceding countershock.
Circulation. 1992;885:281-7.
27. Cobb LA, Fahrenbruch CE, Walsh LTR, et al. Influence of car-
diopulmonary resuscitation prior to defibrillation in patients
with out-of-hospital ventricular fibrillation. JAMA. 1999;281:
1182-8.
5. Noc M, Weil MH, Tang W, et al. Electrocardiographic predic-
tion of the success of cardiac resuscitation. Crit Care Med.
1999;27:708-14.
6. Dzwonczyk R, Brown CB, Werman HA. The median frequency
of ECG during ventricular fibrillation: its use in an algorithm for
estimating the duration of cardiac arrest. IEEE Trans Biomed
Eng. 1990;37:640-6.
7. Carlisle EJ, Allen JD, Kernohan WG, et al. Fourier analysis of

154
PREHOSPITAL EMERGENCY CARE APRIL/JUNE 2001 VOLUME 5 / NUMBER 2
28. Menegazzi JJ, Seaberg DC, Yealy DM, Davis EA, MacLeod BA.
Combination pharmacotherapy with delayed countershock vs.
standard Advanced Cardiac Life Support after prolonged ven-
tricular fibrillation. Prehosp Emerg Care. 2000;4:31-7.
29. Menegazzi JJ, Davis EA, Yealy DM, et al. An experimental algo-
rithm versus standard Advanced Cardiac Life Support in a
swine model of out-of-hospital cardiac arrest. Ann Emerg Med.
1993;22:235-9.
prolonged arrest and experimental cardiopulmonary resuscita-
tion. Ann Emerg Med. 1985;14:521-8.
39. Warner ED, Dahl C, Ewy GA. Myocardial injury from transtho-
racic defibrillator countershock. Arch Pathol. 1975;99:55-9.
40. Vogel U, Wanner T, Bultmann B. Extensive pectoral muscle
necrosis after defibrillation: nonthermal skeletal muscle damage
caused by electroporation. Int Care Med. 1998;24:743-5.
41. Kerber RE, Martins JB, Gascho JA, Marcus ML, Grayzel J. Effect
of direct-current countershocks on regional myocardial contrac-
tility and perfusion. Experimental studies. Circulation.
1981;63:323-32.
30. Warner LL, Hoffman JR, Baraff LJ. Prognostic significance of
field response in out-of-hospital ventricular fibrillation. Chest.
1985;87:22-8.
31. Emergency Cardiac Care Committee and Subcommittees AHA.
Part 6: Advanced cardiovascular life support. Section 2:
Defibrillation. Resuscitation. 2000;46:109-13.
42. Knox MA, Hughes HCJ, Tyers GF, Seidl D, Demers LM. The
induction of myocardial damage by open-chest low-energy
countershock. Med Instrum. 1980;14:63-6.
32. Paradis NA, Marin GB, Rivers EP, et al. Coronary perfusion
pressure and return of spontaneous circulation in human car-
diopulmonary resuscitation. JAMA. 1990;263:1106-13.
33. Niemann JT, Haynes KS, Garner D, et al. Postcountershock
pulseless rhythms: response to CPR, artificial cardiac pacing,
and adrenergic agonists. Ann Emerg Med. 1986;15:112-20.
34. von Planta I, Weil MH, von Planta M, et al. Cardiopulmonary
resuscitation in the rat. J Appl Physiol. 1988;65:2641-7.
35. Tang W, Weil MH, Noc M, Sun S, Gazmuri RJ, Bisera J.
Augmented efficacy of external CPR by intermittent occlusion
of the ascending aorta. Circulation. 1993;88:1916-21.
43. Tacker WA, Van Vleet JF, Geddes LA. Electrocardiographic and
serum enzymatic alterations asociated with cardiac alterations
induced in dogs by single transthoracic damped sinusoidal
defibrillator shocks of various strengths. Am Heart J.
1979;98:185-93.
44. Ewy GA, Taren D, Bangert J, et al. Comparison of myocardial
damage from defibrillator discharges at various dosages. Med
Instrum. 1980;142:9-12.
45. Gaba DM, Talner NS. Myocardial damage following transtho-
racic direct current countershock in newborn piglets. Pediatr
Cardiol. 1982;2:281-8.
36. Xie J, Weil MH, Sun S, et al. High-energy defibrillation increas-
es the severity of postresuscitation myocardial dysfunction.
Circulation. 1997;96:683-8.
37. Strohmenger HU, Lindner KH, Brown CG. Analysis of the ven-
tricular fibrillation ECG signal amplitude and frequency
parameters as predictors of countershock success in humans.
Chest. 1997;111:584-9.
46. Karch SB, Billingham ME. Morphologic effects of defibrillation.
A preliminary report. Crit Care Med. 1984;12:920-1.
47. Tokano T, Bach D, Chang J, et al. Effect of ventricular shock
strength on cardiac hemodynamics. J Cardiovasc Electro-
physiol. 1998;9:791-7.
48. Callaway CW, Sherman LD, Holt E, Deietrich TJ, Mosesso VN.
Ventricular fibrillation waveform predicts defibrillation success
by automatic external defibrillators [abstract]. Acad Emerg
Med. 2000;7:438.
38. Niemann JT, Criley JM, Rosborough JP, Niskanen RA, Alferness
C. Predictive indices of successful cardiac resuscitation after