- Email: [email protected]
Defibrillation with the sequential pulse technique: Reproducibility with repeated shocks
Rosenfeld
12.
13.
14.
15.
16.
May,1996
et al.
American
surviving acute myocardial infarction. Am J Cardiol 50:223, 1982. Richards DA, Cody DV, Denniss AR, Russell PA, Young AA, Uther JB: Ventricular electrical instability: A predictor of death after myocardial infarction. Am J Cardiol 51:75, 1983. Mann DE, Luck JC, Griffen JC, Herre JM, Limacher MC. Magro SA, Robertson NW, Wyndham CRC: Induction of clinical ventricular tachycardia using programmed stimulation: Value of third and fourth extrastimuli. Am J Cardiol 52:501, 1983. Buxton AE, Waxman HL, Marchlinski FE, Untereker WJ, Waspe LE, Josephson ME: Role of triple extrastimuli during electrophysiologic study of patients with documented sustained ventricular tachyarrhythmias. Circulation 69:532, 1984. Brugada P, Green M, Abdollah H, Wellens HJJ: Significance of ventricular arrhythmias initiated by programmed ventricular stimulation: The importance of the type of ventricular arrhythmia induced and the number of premature stimuli required. Circulation 69:87, 1984. Brugada P, Abdollah H, Heddle B, Wellens HJJ: Results of a
17.
18.
19.
20.
Heart Journal
ventricular stimulation protocol using a maximum of 4 premature stimuli in patients without documented or suspected ventricular arrhythmias. Am d Cardiol 52:1214, 1983. Doherty JU, Kienzle MG, Waxman HL, Buxton AE, Marchlinski FE, Josephson ME: Programmed ventricular stimulation at a second right ventricular site: An analysis of 100 patients, with special reference to sensitivity, specificity and characteristics of patients with induced ventricular tachycardia. Am J Cardiol 52:1184. 1983. Robertson JF, Cain ME, Horowitz LN, Spielman SR, Greenspan AM, Waxman HL, ,Josephson ME: Anatomic and electrophysiologic correlates of ventricular tachycardia requiring left ventricular stimulation. Am J Cardiol 46263, 1981. Denker S, Lehmann M, Mahmud R, Gilbert C, Akhtar M: Facilitation of ventricular tachycardia induction with abrupt changes in ventricular cycle length. Am .J Cardiol 53:508, 1984. McPherson CA, Rosenfeld LE, Batsford WP: Day-to-day reproducibility of responses to right ventricular programmed electrical stimulation: Implications for serial drug testing. Am J Cardiol 55689, 1985.
Defibrillation with the sequential pulse technique: Reproducibility with repeated
shocks
The development of the automatic implantable defibrillator has created the need to assess the effects of interventions on defibrillation success. However, first it is important to determine the spontaneous variability of defibrillation threshold (DFT) over time. We repeatedly determined DFT over a maximum of 2-l/2 hours in open-chested, halothane-anesthetized pigs. Ten seconds after induction of fibrillation, defibrillation was attempted by passing a sequential pulse shock through an indwelling catheter and patch electrodes. Ninety-seven fibrillation episodes (FE@ were induced in eight pigs, with a maximum of 30 shocks in an animal. DFT remained stable and fitted a flat least-squares regression equation (y = 6.35 + 0.0055x, where x is the ventricular FE number and y is the DFT, r = 0.0115, p = NS). The mean DFT over time for the eight animals was 7.6 2 1.9 J (range 4.8 to 16.3 J). The inter-animal vartability for DFT was 3.7 J and the mean intra-animal DFT variability over time was 3.6 i 2.3 J (range 0 to 6.8 J). We conclude that, using our methodology, DFT is reproducible and consistent over at least 2 hours. This model provides the basis to assess the effects of acute interventions on the ability to defibrillate. (AM HEART J 111:874, 1986.)
Max F. Rattes, M.D., Douglas L. Jones, Ph.D., Anand Sohla, Arjun D. Sharma, M.D., Eileen Jarvis, and George J. Klein, M.D. London, Ontario, Canada
From-the Ontario;
Departments and University
of Medicine Hospital.
and Physiology,
University
of Western
Supported by the Heart and Stroke Foundation of Ontario and by the Medical Research Council. Dr. C.J. Klein is a senior research Fellow of the Heart and Stroke Foundation of Ontario. Dr. D.L. Jones and Dr. A.D. Sharma are Ministry of Health Career Scientists. Received accepted
for publication Oct. 1, 1985.
Apr.
23, 1985;
revision
Reprint requests: Dr. D. L. Jones, Departments ogy, University of Western Ontario, London,
674
received
Aug.
2, 1985;
of Medicine and PhysiolOntario, Canada N6A 5Cl.
The development of the automatic implantable defibrillator has created the need to assess the effect of drugs and disease states on the ability to defibrillate. Before investigating these effects, it is important to establish the reproducibility and consistency of the minimum amount of energy necessary to defibrillate the heart (i.e., “defibrillation threshold”). We assessed the reproducibility of defibrilla-
Volume Number
111 5
Defibrillation
tion threshold over 2 ti. 2-G hours in anesthetized pigs using an internal sequential pulse system in order to evaluate its utility in assessing the effect of acute interventions on the ability to defibrillate. METHODS Animals
DEFIBRILLATOR VOLTAGE
CURRENT
’
of ventricular
fibrillation
and defibrillation.
Ventricular fibrillation was induced by passinga 60 Hz train of 5 mA pulsesof 500 to 1000msecduration to the epicardial right ventricular patch electrode. Ventricular
875
shocks
-I -
‘k
DEFIBRILLATOR VOLTAGE
with sequential
I -\’
and surgery. Eight pigs weighing 15 to 18 kg
were fasted overnight and received a preanesthetic mixture of ketamine hydrochloride (Rogar/STB Inc., 400 mg intramuscularly) and diazepam (Roche, 10 mg intramuscularly) for transportation to the laboratory. Each pig was intubated and ventilated at 7.5 respirations per minute with room air and halothane (Hoechst, 0.5% to 1.5%) blended with 100% oxygen to maintain anesthesiafor the duration of the experiment. Body temperature was monitored esophageallyand wasmaintained with a circulating water blanket and temperature controller. A countershockcatheter (Medtronic, 6880)wasinserted into the left jugular vein. This catheter hasbeen described previously’ and consistsof a pair of electrodesat the distal tip of the catheter and a second pair of ring electrodes located 100to 125cm proximally, each with a surfacearea of 1.25cm2.The catheter wasadvanced until the distal tip rested in the right ventricular apex and the proximal electrode pair was positioned immediately above the superior vena cava-right atrial junction region. An elliptical mesh plaque electrode of surface area 2.5 cm* (Medtronic, TX-7) wassutured on the epicardium of the lateral basalleft ventricle. The countershockcatheter and plaque electrode were connected to two custom-designeddefibrillators to deliver a sequentialpulse shock.Monitor outputs from the defibrillators were connected to a dual-beam storageoscilloscope(Tektronix, 5113) to monitor current, voltage, and the duration of the defibrillation shocks.The displayed pulses were photographed (Fig. 1) and the leading edgepeak voltage and current were measuredand the delivered energy was calculated. A quadripolar patch electrode was sutured over the right ventricular outflow tract. Two polesof the quadripolar patch electrode were used for recording ventricular electrograms, while the other two poles were used for electrical stimulation of the ventricle and induction of ventricular fibrillation. Electrical monitoring. Limb leadsI and II, two unipolar electrograms and a bipolar right ventricular endocardial electrogram from the distal catheter electrodes, and two unipolar electrograms and a bipolar right ventricular epicardial electrogram from the quadripolar patch electrode were monitored on an Electronics for Medicine VR16 recorder and were recorded on magnetic tape (Ampex PR 2230) throughout the experiment. The limb leadsand unipolar electrogramswere recorded at frequency settings of 1 to 250 Hz and the bipolar electrograms were recorded at frequency settings of 30 to 500 Hz. Induction
reproducibility
L
200
v
Imr
-.I Ims
2A
2 ’ -T-\
CURRENT
lsl
2nd -IiSEPARATION
1. Oscilloscopetracings obtained during a successful sequential pulse shock. Solid and open arrows indicate the leading edge peak voltage and current for defibrillators 1 and 2, respectively. The top two tracings and the bottom two tracings are from the monitor outputs of defibrillators 1 and 2, respectively. Fig.
fibrillation wasverified by observing multiple ECG channels and intracardiac electrogramsz3 Ten secondsafter the onset of ventricular fibrillation, a sequential pulse defibrillation shock consistingof two trapezoidal pulsesof equal energy content, approximately 3 msec in duration, 30% tilt, and separatedby 0.2 msecwas delivered to the heart. The first pulse wasdelivered between the proximal pair of catheter electrodes (anode) and the distal pair of catheter electrodesin the right ventricular apex (cathode). The second pulse was delivered between the plaque electrode on the left ventricular epicardium (anode) and distal catheter electrodes (cathode).* Determining defibrillation threshold. In order to determine the reproducibility of the minimum amount of energy necessaryto defibrillate a given animal (“defibrillation threshold”), multiple episodesof ventricular fibrillation and defibrillation attempts were repeated for a maximum of 2-‘/z hours with the meannumber of fibrillation episodes per animal of 12.1 f 0.35 FE (x f S.E., range 11 to 14 FE). A minimum of 10 minutes separated each ventricular fibrillation episode. The stored voltage for the first shock was set to deliver approximately 4.2 J. All subsequentinitial stored voltages were basedon the results of the immediately preceding ventricular fibrillation episode.If the initial shock of the immediately preceding ventricular fibrillation episodewas successful,then the stored voltage for the next ventricular fibrillation episodewas set at one level lower than that which previously achieved defibrillation. This processwas repeated until the initial shockof a ventricular fibrillation episodewas unsuccessful. If the initial shock of a ventricular fibrillation episode was unsuccessful,the stored voltage wasincreasedto the next setting (Table I) and defibrillation was attempted after an additional 5 seconds.A singlerescueshock of 25 J
my, 1986
676
Rattes
et al.
500 450 380 320 260 190
American
Journal
I. Relationship between the preset stored voltage levels and the total energy delivered from the defibrillatom Table
A a 0
l
-we
0
i 4
B 450i 380 320 260 190
Heart
0 0 0
0 0
-0
0
4
I
2 3 VENTRICULAR FIBRILLATION EPISODE Fig. 2. Schematic example of a sequenceof ventricular fibrillation episodesdemonstrating the methodology used in this study to determine defibrillation threshold. Solid circles representsuccessfulshocks.Open circles represent unsuccessfulshocks. Arrows indicate the stored voltage for defibrillation threshold. A, Since the shock was successful in the first ventricular fibrillation episode, the stored voltages for the initial shock in the secondventricular fibrillation episodewere set at one level lower. This process was continued until in the third ventricular fibrillation episode the initial shock was unsuccessful. Defibrillation threshold was therefore assignedto delivered energy calculated from the secondventricular fibrillation episode (arrow). B, Since the initial shock was unsuccessfulin the first ventricular fibrillation episode, the stored voltage for the initial shock in the second ventricular fibrillation episodewasset at one level higher. The initial shock was successfulin the third ventricular fibrillation episodeand defibrillation threshold wastherefore assignedto its calculated delivered energy. C, Since the initial shock failed but the second was successful, defibrillation threshold was therefore assigned to the delivered energy calculated from the successfulshock of the first ventricular fibrillation episode(arrow).
Stared
Stored
w
Energy * (J)
190
1.6
voltage
260 320 380 450 500 550 600
2.4 3.3 4.2 6.5 7.9 9.5 11.4
W)
Energy * (J)
630 700 780 900 1000 1180 1410 1570
13.6 16.5 19.0 23.7 30.7 41.3 58.7 70.9
voltage
*Energy was calculated from the oscilloscope the difibrillators into a 110 Q test load.
tracings of the discharge of
wasdelivered with paddle electrodesapplied to the heart if defibrillation wasnot achieved after sevenattempts. No fibrillation episodewas allowed to exceed 50 seconds.fn the next ventricular fibrillation episode,the stored voltage was set at one level higher than that of the initial shock that failed to defibrillate in the immediately preceding ventricular fibrillation episode.The initial stored voltage in the next ventricular fibrillation episoderemainedunaltered only if the initial shock was unsuccessfulbut the secondshock wassuccessful.Defibrillation threshold was determined when the delivered energy of the initial or secondshockwassuccessfulin defibrillating the heart and the energy delivered at the immediately lower stored voltage setting wasunsuccessfulin the same,preceding, or following ventricular fibrillation episode.Fig. 2 gives an example of the methodology usedto determine defibrillation threshold. Statistical analysis. The comparisonbetweenmeasurements over time was made using two-way analysis of variance, treatment by subjects design.5A least-squares regressionequation6and the 95% confidence limits were also calculated for energy, leading edgepeak voltage, and current, with time as the independent variable. The Bartlett’s test of homogeneity of variance wasusedto test the adequacy of the linear fit. A probability value of less than 0.05 wasconsidered significant. RESULTS
Ninety-seven episodes of ventricular fibrillation were induced in the eight pigs, with a maximum of 30 shocks in any animal. The mean values of the parameters for successful defibrillation in these 97 episodes were 7.8 + 0.34 J, 422 + 8.8 V, and 4.7 it 0.098 A for total energy, leading edge peak voltage, and current, respectively. The frequency polygon of delivered energy for all successful defibrillation shocks is shown in Fig. 3. The mean defibrillation threshold energy, leading
Volume Number
111 5
Defibrillation
-
reproducibility
with sequential
shocks
877
IO
-
I-
L
* _
--------
--_____
.2 r
0
E
6-
W
0
zl
_---
4-
0
0 _/-
- --______--
o _____-
O
-------0
-----------
0 0
<--
/---g’
/e-c
0
0 0
-_______
___---
---s_
--a_ .
21”“““““’ 8 ENERGY
10 12 (joules)
14
16
18
Fig. 3. Frequency polygon of 97 delivered successful shocks in the eight animals. The distribution of successful energies is essentially unimodal and the limits of the most frequent successful shocks are 4.4 and 8.5 J.
edge peak voltage, and current remained stable over the 2-hour time span for the eight-animals (Fig. 4). Defibrillation threshold fitted a least-squares regression equation (y = 6.35 + 0.0055x, where x is the ventricular fibrillation episode number and y is defibrillation threshold, r = 0.0115, p = NS). The Bartlett’s test indicated that the linear regression fits the data well (x2 = 5.839, p = NS). The linear relationship with time for each parameter with a 95% confidence band for each regression line is illustrated in Fig. 4. The mean defibrillation threshold over time for the eight animals was 7.6 + 1.3 J (range 4.8 to 16.3 J). The standard deviation of 3.7 shows a low interanimal variability of defibrillation threshold. The difference between the lowest and highest defibrillation threshold energy for each animal was used as an index of intra-animal variability. The mean difference for the eight pigs was 3.6 + 0.82 J, with a range of 0 to 6.8 J. DISCUSSION
Defibrillation threshold as defined in this study is stable and reproducible over time when using direct cardiac sequential pulse shocks in halothane anesthetized pigs. The use of the sequential pulse system enabled defibrillation to be repeated using low energies, requiring a rescue shock in only 1 of 97 episodes of ventricular fibrillation. In previous studies, Babbs et al.‘eg found that defibrillation threshold did not change over time when using transthoracic shocks. Reproducibility of defibrillation threshold over time is important to permit the assessment of the
%, kt:‘r w; (3, 8% pe
400300
-----____ -
0
0
’
’ 2
0-
---em_ ~ _______
o __---
0
---- ______-9-------------------------------------
----0 0
_--- ,-o __----0 O ”
l& 4’ -I
200
0
’
’ 4
’
’ 6
’
’ 0
’
’ IO
’
’ 12
’
VENTRICULAR FIBRILLATION EPISODE Fig. 4. Mean threshold energy, leading edge peak voltage, and current for defibrillation over time. The linear regression equation was calculated using threshold values from two animals chosen at random at each time interval. Each open circle represents the mean of these two values. All parameters remained stable over the 2 to 2-S hours of the experiment.
effects of pharmacologic agents and disease states on the ability to defibrillate. One can experimentally obtain a control value of the amount of energy necessary to defibrillate an animal before a drug is administered or interventions such as ischemia are induced. Observed alterations in defibrillation threshold could then be attributed to an intervention. The low intra-animal and interanimal variability of our sample facilitates demonstration of the effect of an intervention on defibrillation threshold. Successful energies for defibrillation in the eight animals appeared to follow a normal distribution, and it is reasonable to assume that our sample is representative of the porcine population. The mean and the 95% confidence intervals for successful defibrillation in this group of pigs was 7.1 + 3.1 J.
Rattes et al.
American
The upper confidence limit indicates that in order to assure 95% successful defibrillation with the use of sequential pulse shocks, it is necessary to deliver energy 2.4 S.E. above the population mean. These data provide a guideline to assess the effect of interventions on defibrillation threshold. For example, in an unpaired sample of 15 animals in each group and allowing an alpha and beta error of 5% and 15%, respectively, a difference of 4.0 J is necessary to demonstrate a statistical difference due to an intervention. In summary, this study demonstrates the reproducibility and consistency of defibrillation threshold over at least 2 hours with the use of an internal sequential defibrillation system in pigs. By inference, one could predict a level of energy which defibrillates 95% of a population from a sample mean and the standard error of successful energy. This model also provides the basis to assess the effect of acute interventions on the ability to defibrillate. The
authors
wish
Protective fibrillation
to thank
Dr. A.P. Dormer
for his assistance
with statistical manuscript.
analysis
and Mrs.
D. Vigna
May, 1986 Heart Journal
for preparation
of the
REFERENCES
1. Yee R, Jones DL, Jarvis E, Donner AP, Klein GJ: Changes in pacing threshold and R wave amplitude after transvenous catheter countershock. J Am Co11 Cardiol 4:543, 1984. 2. Jones DL, Klein GJ, Gulamhusein S, Jarvis E: The repetitive ventricular response: Relationship to ventricular fibrillation threshold in dogs. PACE 6:1258, 1983. 3. Jones DL, Klein GJ: Ventricular fibrillation: The importance of being coarse? J Electrocardiol 17:393, 1934. 4. Jones DL, Klein GJ, Kellog MJ: Improved defibrillation threshold with sequential pulse energy delivered to two different lead orientations in pigs. Am J Cardiol 55:821. 1985. 5. Sokal RR, Rohlf FJ: Biometry, 1981. San Francisco, 1981, Freeman and Company. 6. Colton T: Statistics in medicine. Boston, 1974. Little, Brown & Company. 7. Babbs CF, Whistler SJ, Geddes LA: Reproducibility of defibrillation threshold data in experimental animals (abstrl. Med Instrum 12:52, 1978. 8. Babbs CF, Whistler SJ, Yim GKW: Temporal stability and precision of ventricular defibrillation threshold data. Am J Physiol 4:553, 1978. 9. Babbs CF: Drug induced changes in ventricular defibrillation threshold. Thesis, 1977 Purdue University, West Lafayette, Ind.
effect of tiapamil against ventricular during coronary artery occlusion
Calcium channel antagonists differ in their effects on myocardial excitable properties. This study examines whether tiapamil (100 Ag/kg/min intravenously) is capable of reducing the susceptibility to ventricular fibrillation (VF) during acute occlusion and reperfusion of the left anterior descending coronary artery. During occlusion, tiapamil elevated the VF threshold to 17.5 ? 8.2 mA compared to a control of 8.6 + 5.9 mA (2~ < 0.01). However, no significant effect was noted upon reperfusion of the vessel. Adrenergic stimulation with norepinephrine, 0.5 Ag/kg/min, lowered the VF threshold by 32% (2~ < 0.02), and by 6% (N-S.) when tiapamil was infused concurrently. Thus, tiapamil protects the heart against VF during coronary occlusion, but not during reperfusion. This appears to be mediated in part by an antiadrenergic action of the drug. (AM HEART J 111:878, 1986.)
Ernst A. Raeder, M.D., Richard Boston,
L. Verrier, Ph.D., and Bernard Lown, M.D.
Mass.
From the Cardiovascular School of Public Health Women’s Hospital.
Laboratories, Department of Nutrition, Harvard and the Department of Medicine, Brigham and
This study was supported in part by Grant No. HL-33567 from the National Heart, Lung, and Blood Institute of the National Institutes of Health, United States Public Health Service, Bethesda, Md.; by the Rappaport International Program in Cardiology; and by a grant from the F. Hoffmann-LaRoche & Co., Base& Switzerland. Received
for publication
Reprint requests: of Public Health,
878
Sept.
Bernard Lawn, 665 Huntington
17, 1985;
accepted
Oct. 21, 1985.
M.D., Dept. of Nutrition, Harvard Ave., Boston, MA 02115.
School
The calcium channel blocking agents are an important category of drugs that have changed medical management of angina pectoris. Though they share a common property of affecting smooth vascular tone, they differ among one another in important respects, as they exert negative inotropic effects on the myocardium and differing effects on atrial and nodal tissue. When arrhythmias of the heart are due to ischemia, calcium antagonists have been found to
12.
13.
14.
15.
16.
May,1996
et al.
American
surviving acute myocardial infarction. Am J Cardiol 50:223, 1982. Richards DA, Cody DV, Denniss AR, Russell PA, Young AA, Uther JB: Ventricular electrical instability: A predictor of death after myocardial infarction. Am J Cardiol 51:75, 1983. Mann DE, Luck JC, Griffen JC, Herre JM, Limacher MC. Magro SA, Robertson NW, Wyndham CRC: Induction of clinical ventricular tachycardia using programmed stimulation: Value of third and fourth extrastimuli. Am J Cardiol 52:501, 1983. Buxton AE, Waxman HL, Marchlinski FE, Untereker WJ, Waspe LE, Josephson ME: Role of triple extrastimuli during electrophysiologic study of patients with documented sustained ventricular tachyarrhythmias. Circulation 69:532, 1984. Brugada P, Green M, Abdollah H, Wellens HJJ: Significance of ventricular arrhythmias initiated by programmed ventricular stimulation: The importance of the type of ventricular arrhythmia induced and the number of premature stimuli required. Circulation 69:87, 1984. Brugada P, Abdollah H, Heddle B, Wellens HJJ: Results of a
17.
18.
19.
20.
Heart Journal
ventricular stimulation protocol using a maximum of 4 premature stimuli in patients without documented or suspected ventricular arrhythmias. Am d Cardiol 52:1214, 1983. Doherty JU, Kienzle MG, Waxman HL, Buxton AE, Marchlinski FE, Josephson ME: Programmed ventricular stimulation at a second right ventricular site: An analysis of 100 patients, with special reference to sensitivity, specificity and characteristics of patients with induced ventricular tachycardia. Am J Cardiol 52:1184. 1983. Robertson JF, Cain ME, Horowitz LN, Spielman SR, Greenspan AM, Waxman HL, ,Josephson ME: Anatomic and electrophysiologic correlates of ventricular tachycardia requiring left ventricular stimulation. Am J Cardiol 46263, 1981. Denker S, Lehmann M, Mahmud R, Gilbert C, Akhtar M: Facilitation of ventricular tachycardia induction with abrupt changes in ventricular cycle length. Am .J Cardiol 53:508, 1984. McPherson CA, Rosenfeld LE, Batsford WP: Day-to-day reproducibility of responses to right ventricular programmed electrical stimulation: Implications for serial drug testing. Am J Cardiol 55689, 1985.
Defibrillation with the sequential pulse technique: Reproducibility with repeated
shocks
The development of the automatic implantable defibrillator has created the need to assess the effects of interventions on defibrillation success. However, first it is important to determine the spontaneous variability of defibrillation threshold (DFT) over time. We repeatedly determined DFT over a maximum of 2-l/2 hours in open-chested, halothane-anesthetized pigs. Ten seconds after induction of fibrillation, defibrillation was attempted by passing a sequential pulse shock through an indwelling catheter and patch electrodes. Ninety-seven fibrillation episodes (FE@ were induced in eight pigs, with a maximum of 30 shocks in an animal. DFT remained stable and fitted a flat least-squares regression equation (y = 6.35 + 0.0055x, where x is the ventricular FE number and y is the DFT, r = 0.0115, p = NS). The mean DFT over time for the eight animals was 7.6 2 1.9 J (range 4.8 to 16.3 J). The inter-animal vartability for DFT was 3.7 J and the mean intra-animal DFT variability over time was 3.6 i 2.3 J (range 0 to 6.8 J). We conclude that, using our methodology, DFT is reproducible and consistent over at least 2 hours. This model provides the basis to assess the effects of acute interventions on the ability to defibrillate. (AM HEART J 111:874, 1986.)
Max F. Rattes, M.D., Douglas L. Jones, Ph.D., Anand Sohla, Arjun D. Sharma, M.D., Eileen Jarvis, and George J. Klein, M.D. London, Ontario, Canada
From-the Ontario;
Departments and University
of Medicine Hospital.
and Physiology,
University
of Western
Supported by the Heart and Stroke Foundation of Ontario and by the Medical Research Council. Dr. C.J. Klein is a senior research Fellow of the Heart and Stroke Foundation of Ontario. Dr. D.L. Jones and Dr. A.D. Sharma are Ministry of Health Career Scientists. Received accepted
for publication Oct. 1, 1985.
Apr.
23, 1985;
revision
Reprint requests: Dr. D. L. Jones, Departments ogy, University of Western Ontario, London,
674
received
Aug.
2, 1985;
of Medicine and PhysiolOntario, Canada N6A 5Cl.
The development of the automatic implantable defibrillator has created the need to assess the effect of drugs and disease states on the ability to defibrillate. Before investigating these effects, it is important to establish the reproducibility and consistency of the minimum amount of energy necessary to defibrillate the heart (i.e., “defibrillation threshold”). We assessed the reproducibility of defibrilla-
Volume Number
111 5
Defibrillation
tion threshold over 2 ti. 2-G hours in anesthetized pigs using an internal sequential pulse system in order to evaluate its utility in assessing the effect of acute interventions on the ability to defibrillate. METHODS Animals
DEFIBRILLATOR VOLTAGE
CURRENT
’
of ventricular
fibrillation
and defibrillation.
Ventricular fibrillation was induced by passinga 60 Hz train of 5 mA pulsesof 500 to 1000msecduration to the epicardial right ventricular patch electrode. Ventricular
875
shocks
-I -
‘k
DEFIBRILLATOR VOLTAGE
with sequential
I -\’
and surgery. Eight pigs weighing 15 to 18 kg
were fasted overnight and received a preanesthetic mixture of ketamine hydrochloride (Rogar/STB Inc., 400 mg intramuscularly) and diazepam (Roche, 10 mg intramuscularly) for transportation to the laboratory. Each pig was intubated and ventilated at 7.5 respirations per minute with room air and halothane (Hoechst, 0.5% to 1.5%) blended with 100% oxygen to maintain anesthesiafor the duration of the experiment. Body temperature was monitored esophageallyand wasmaintained with a circulating water blanket and temperature controller. A countershockcatheter (Medtronic, 6880)wasinserted into the left jugular vein. This catheter hasbeen described previously’ and consistsof a pair of electrodesat the distal tip of the catheter and a second pair of ring electrodes located 100to 125cm proximally, each with a surfacearea of 1.25cm2.The catheter wasadvanced until the distal tip rested in the right ventricular apex and the proximal electrode pair was positioned immediately above the superior vena cava-right atrial junction region. An elliptical mesh plaque electrode of surface area 2.5 cm* (Medtronic, TX-7) wassutured on the epicardium of the lateral basalleft ventricle. The countershockcatheter and plaque electrode were connected to two custom-designeddefibrillators to deliver a sequentialpulse shock.Monitor outputs from the defibrillators were connected to a dual-beam storageoscilloscope(Tektronix, 5113) to monitor current, voltage, and the duration of the defibrillation shocks.The displayed pulses were photographed (Fig. 1) and the leading edgepeak voltage and current were measuredand the delivered energy was calculated. A quadripolar patch electrode was sutured over the right ventricular outflow tract. Two polesof the quadripolar patch electrode were used for recording ventricular electrograms, while the other two poles were used for electrical stimulation of the ventricle and induction of ventricular fibrillation. Electrical monitoring. Limb leadsI and II, two unipolar electrograms and a bipolar right ventricular endocardial electrogram from the distal catheter electrodes, and two unipolar electrograms and a bipolar right ventricular epicardial electrogram from the quadripolar patch electrode were monitored on an Electronics for Medicine VR16 recorder and were recorded on magnetic tape (Ampex PR 2230) throughout the experiment. The limb leadsand unipolar electrogramswere recorded at frequency settings of 1 to 250 Hz and the bipolar electrograms were recorded at frequency settings of 30 to 500 Hz. Induction
reproducibility
L
200
v
Imr
-.I Ims
2A
2 ’ -T-\
CURRENT
lsl
2nd -IiSEPARATION
1. Oscilloscopetracings obtained during a successful sequential pulse shock. Solid and open arrows indicate the leading edge peak voltage and current for defibrillators 1 and 2, respectively. The top two tracings and the bottom two tracings are from the monitor outputs of defibrillators 1 and 2, respectively. Fig.
fibrillation wasverified by observing multiple ECG channels and intracardiac electrogramsz3 Ten secondsafter the onset of ventricular fibrillation, a sequential pulse defibrillation shock consistingof two trapezoidal pulsesof equal energy content, approximately 3 msec in duration, 30% tilt, and separatedby 0.2 msecwas delivered to the heart. The first pulse wasdelivered between the proximal pair of catheter electrodes (anode) and the distal pair of catheter electrodesin the right ventricular apex (cathode). The second pulse was delivered between the plaque electrode on the left ventricular epicardium (anode) and distal catheter electrodes (cathode).* Determining defibrillation threshold. In order to determine the reproducibility of the minimum amount of energy necessaryto defibrillate a given animal (“defibrillation threshold”), multiple episodesof ventricular fibrillation and defibrillation attempts were repeated for a maximum of 2-‘/z hours with the meannumber of fibrillation episodes per animal of 12.1 f 0.35 FE (x f S.E., range 11 to 14 FE). A minimum of 10 minutes separated each ventricular fibrillation episode. The stored voltage for the first shock was set to deliver approximately 4.2 J. All subsequentinitial stored voltages were basedon the results of the immediately preceding ventricular fibrillation episode.If the initial shock of the immediately preceding ventricular fibrillation episodewas successful,then the stored voltage for the next ventricular fibrillation episodewas set at one level lower than that which previously achieved defibrillation. This processwas repeated until the initial shockof a ventricular fibrillation episodewas unsuccessful. If the initial shock of a ventricular fibrillation episode was unsuccessful,the stored voltage wasincreasedto the next setting (Table I) and defibrillation was attempted after an additional 5 seconds.A singlerescueshock of 25 J
my, 1986
676
Rattes
et al.
500 450 380 320 260 190
American
Journal
I. Relationship between the preset stored voltage levels and the total energy delivered from the defibrillatom Table
A a 0
l
-we
0
i 4
B 450i 380 320 260 190
Heart
0 0 0
0 0
-0
0
4
I
2 3 VENTRICULAR FIBRILLATION EPISODE Fig. 2. Schematic example of a sequenceof ventricular fibrillation episodesdemonstrating the methodology used in this study to determine defibrillation threshold. Solid circles representsuccessfulshocks.Open circles represent unsuccessfulshocks. Arrows indicate the stored voltage for defibrillation threshold. A, Since the shock was successful in the first ventricular fibrillation episode, the stored voltages for the initial shock in the secondventricular fibrillation episodewere set at one level lower. This process was continued until in the third ventricular fibrillation episode the initial shock was unsuccessful. Defibrillation threshold was therefore assignedto delivered energy calculated from the secondventricular fibrillation episode (arrow). B, Since the initial shock was unsuccessfulin the first ventricular fibrillation episode, the stored voltage for the initial shock in the second ventricular fibrillation episodewasset at one level higher. The initial shock was successfulin the third ventricular fibrillation episodeand defibrillation threshold wastherefore assignedto its calculated delivered energy. C, Since the initial shock failed but the second was successful, defibrillation threshold was therefore assigned to the delivered energy calculated from the successfulshock of the first ventricular fibrillation episode(arrow).
Stared
Stored
w
Energy * (J)
190
1.6
voltage
260 320 380 450 500 550 600
2.4 3.3 4.2 6.5 7.9 9.5 11.4
W)
Energy * (J)
630 700 780 900 1000 1180 1410 1570
13.6 16.5 19.0 23.7 30.7 41.3 58.7 70.9
voltage
*Energy was calculated from the oscilloscope the difibrillators into a 110 Q test load.
tracings of the discharge of
wasdelivered with paddle electrodesapplied to the heart if defibrillation wasnot achieved after sevenattempts. No fibrillation episodewas allowed to exceed 50 seconds.fn the next ventricular fibrillation episode,the stored voltage was set at one level higher than that of the initial shock that failed to defibrillate in the immediately preceding ventricular fibrillation episode.The initial stored voltage in the next ventricular fibrillation episoderemainedunaltered only if the initial shock was unsuccessfulbut the secondshock wassuccessful.Defibrillation threshold was determined when the delivered energy of the initial or secondshockwassuccessfulin defibrillating the heart and the energy delivered at the immediately lower stored voltage setting wasunsuccessfulin the same,preceding, or following ventricular fibrillation episode.Fig. 2 gives an example of the methodology usedto determine defibrillation threshold. Statistical analysis. The comparisonbetweenmeasurements over time was made using two-way analysis of variance, treatment by subjects design.5A least-squares regressionequation6and the 95% confidence limits were also calculated for energy, leading edgepeak voltage, and current, with time as the independent variable. The Bartlett’s test of homogeneity of variance wasusedto test the adequacy of the linear fit. A probability value of less than 0.05 wasconsidered significant. RESULTS
Ninety-seven episodes of ventricular fibrillation were induced in the eight pigs, with a maximum of 30 shocks in any animal. The mean values of the parameters for successful defibrillation in these 97 episodes were 7.8 + 0.34 J, 422 + 8.8 V, and 4.7 it 0.098 A for total energy, leading edge peak voltage, and current, respectively. The frequency polygon of delivered energy for all successful defibrillation shocks is shown in Fig. 3. The mean defibrillation threshold energy, leading
Volume Number
111 5
Defibrillation
-
reproducibility
with sequential
shocks
877
IO
-
I-
L
* _
--------
--_____
.2 r
0
E
6-
W
0
zl
_---
4-
0
0 _/-
- --______--
o _____-
O
-------0
-----------
0 0
<--
/---g’
/e-c
0
0 0
-_______
___---
---s_
--a_ .
21”“““““’ 8 ENERGY
10 12 (joules)
14
16
18
Fig. 3. Frequency polygon of 97 delivered successful shocks in the eight animals. The distribution of successful energies is essentially unimodal and the limits of the most frequent successful shocks are 4.4 and 8.5 J.
edge peak voltage, and current remained stable over the 2-hour time span for the eight-animals (Fig. 4). Defibrillation threshold fitted a least-squares regression equation (y = 6.35 + 0.0055x, where x is the ventricular fibrillation episode number and y is defibrillation threshold, r = 0.0115, p = NS). The Bartlett’s test indicated that the linear regression fits the data well (x2 = 5.839, p = NS). The linear relationship with time for each parameter with a 95% confidence band for each regression line is illustrated in Fig. 4. The mean defibrillation threshold over time for the eight animals was 7.6 + 1.3 J (range 4.8 to 16.3 J). The standard deviation of 3.7 shows a low interanimal variability of defibrillation threshold. The difference between the lowest and highest defibrillation threshold energy for each animal was used as an index of intra-animal variability. The mean difference for the eight pigs was 3.6 + 0.82 J, with a range of 0 to 6.8 J. DISCUSSION
Defibrillation threshold as defined in this study is stable and reproducible over time when using direct cardiac sequential pulse shocks in halothane anesthetized pigs. The use of the sequential pulse system enabled defibrillation to be repeated using low energies, requiring a rescue shock in only 1 of 97 episodes of ventricular fibrillation. In previous studies, Babbs et al.‘eg found that defibrillation threshold did not change over time when using transthoracic shocks. Reproducibility of defibrillation threshold over time is important to permit the assessment of the
%, kt:‘r w; (3, 8% pe
400300
-----____ -
0
0
’
’ 2
0-
---em_ ~ _______
o __---
0
---- ______-9-------------------------------------
----0 0
_--- ,-o __----0 O ”
l& 4’ -I
200
0
’
’ 4
’
’ 6
’
’ 0
’
’ IO
’
’ 12
’
VENTRICULAR FIBRILLATION EPISODE Fig. 4. Mean threshold energy, leading edge peak voltage, and current for defibrillation over time. The linear regression equation was calculated using threshold values from two animals chosen at random at each time interval. Each open circle represents the mean of these two values. All parameters remained stable over the 2 to 2-S hours of the experiment.
effects of pharmacologic agents and disease states on the ability to defibrillate. One can experimentally obtain a control value of the amount of energy necessary to defibrillate an animal before a drug is administered or interventions such as ischemia are induced. Observed alterations in defibrillation threshold could then be attributed to an intervention. The low intra-animal and interanimal variability of our sample facilitates demonstration of the effect of an intervention on defibrillation threshold. Successful energies for defibrillation in the eight animals appeared to follow a normal distribution, and it is reasonable to assume that our sample is representative of the porcine population. The mean and the 95% confidence intervals for successful defibrillation in this group of pigs was 7.1 + 3.1 J.
Rattes et al.
American
The upper confidence limit indicates that in order to assure 95% successful defibrillation with the use of sequential pulse shocks, it is necessary to deliver energy 2.4 S.E. above the population mean. These data provide a guideline to assess the effect of interventions on defibrillation threshold. For example, in an unpaired sample of 15 animals in each group and allowing an alpha and beta error of 5% and 15%, respectively, a difference of 4.0 J is necessary to demonstrate a statistical difference due to an intervention. In summary, this study demonstrates the reproducibility and consistency of defibrillation threshold over at least 2 hours with the use of an internal sequential defibrillation system in pigs. By inference, one could predict a level of energy which defibrillates 95% of a population from a sample mean and the standard error of successful energy. This model also provides the basis to assess the effect of acute interventions on the ability to defibrillate. The
authors
wish
Protective fibrillation
to thank
Dr. A.P. Dormer
for his assistance
with statistical manuscript.
analysis
and Mrs.
D. Vigna
May, 1986 Heart Journal
for preparation
of the
REFERENCES
1. Yee R, Jones DL, Jarvis E, Donner AP, Klein GJ: Changes in pacing threshold and R wave amplitude after transvenous catheter countershock. J Am Co11 Cardiol 4:543, 1984. 2. Jones DL, Klein GJ, Gulamhusein S, Jarvis E: The repetitive ventricular response: Relationship to ventricular fibrillation threshold in dogs. PACE 6:1258, 1983. 3. Jones DL, Klein GJ: Ventricular fibrillation: The importance of being coarse? J Electrocardiol 17:393, 1934. 4. Jones DL, Klein GJ, Kellog MJ: Improved defibrillation threshold with sequential pulse energy delivered to two different lead orientations in pigs. Am J Cardiol 55:821. 1985. 5. Sokal RR, Rohlf FJ: Biometry, 1981. San Francisco, 1981, Freeman and Company. 6. Colton T: Statistics in medicine. Boston, 1974. Little, Brown & Company. 7. Babbs CF, Whistler SJ, Geddes LA: Reproducibility of defibrillation threshold data in experimental animals (abstrl. Med Instrum 12:52, 1978. 8. Babbs CF, Whistler SJ, Yim GKW: Temporal stability and precision of ventricular defibrillation threshold data. Am J Physiol 4:553, 1978. 9. Babbs CF: Drug induced changes in ventricular defibrillation threshold. Thesis, 1977 Purdue University, West Lafayette, Ind.
effect of tiapamil against ventricular during coronary artery occlusion
Calcium channel antagonists differ in their effects on myocardial excitable properties. This study examines whether tiapamil (100 Ag/kg/min intravenously) is capable of reducing the susceptibility to ventricular fibrillation (VF) during acute occlusion and reperfusion of the left anterior descending coronary artery. During occlusion, tiapamil elevated the VF threshold to 17.5 ? 8.2 mA compared to a control of 8.6 + 5.9 mA (2~ < 0.01). However, no significant effect was noted upon reperfusion of the vessel. Adrenergic stimulation with norepinephrine, 0.5 Ag/kg/min, lowered the VF threshold by 32% (2~ < 0.02), and by 6% (N-S.) when tiapamil was infused concurrently. Thus, tiapamil protects the heart against VF during coronary occlusion, but not during reperfusion. This appears to be mediated in part by an antiadrenergic action of the drug. (AM HEART J 111:878, 1986.)
Ernst A. Raeder, M.D., Richard Boston,
L. Verrier, Ph.D., and Bernard Lown, M.D.
Mass.
From the Cardiovascular School of Public Health Women’s Hospital.
Laboratories, Department of Nutrition, Harvard and the Department of Medicine, Brigham and
This study was supported in part by Grant No. HL-33567 from the National Heart, Lung, and Blood Institute of the National Institutes of Health, United States Public Health Service, Bethesda, Md.; by the Rappaport International Program in Cardiology; and by a grant from the F. Hoffmann-LaRoche & Co., Base& Switzerland. Received
for publication
Reprint requests: of Public Health,
878
Sept.
Bernard Lawn, 665 Huntington
17, 1985;
accepted
Oct. 21, 1985.
M.D., Dept. of Nutrition, Harvard Ave., Boston, MA 02115.
School
The calcium channel blocking agents are an important category of drugs that have changed medical management of angina pectoris. Though they share a common property of affecting smooth vascular tone, they differ among one another in important respects, as they exert negative inotropic effects on the myocardium and differing effects on atrial and nodal tissue. When arrhythmias of the heart are due to ischemia, calcium antagonists have been found to