- Email: [email protected]
Intraoperative comparison of sequential-pulse and single-pulse defibrillation in candidates for automatic implantable defibrillators
lntraopecativeComparisonof Sequential-Pulse and Single-PulseDefibrillationin Candidates for htomatic ImplantableDefibrillators GUST H. BARDY, MD, ROBERT B. STEWART, BS, TOM D. IVEY, MD, ELLEN L. GRAHAM, RN, GUR C. ADHAR, MD, and H. LEON GREENE, MD
During defibrillation threshold determination, voltage and current waveforms were recorded and integrated to determine delivered energy. Average defibrillation threshold leading-edge voltage for the sequential pulse technique was 496 f 140 V, compared with 365 f 157 V for the single-pulse technique (p <0.005). Defibrillation threshold leadingedge current for the sequential-pulse technique was 6.0 f 2.3 A, compared with 10.6 f 5.1 A for the single-pulse method (p
Sixteen survivors of cardiac arrest underwent intraoperative comparison of the effectiveness of sequentiaf-pulse and single-pulse defibrillation. Defibrillation was tested alternately with the single-pulse or sequential-pulse technique 10 seconds into an episode of ventricular fibrillation that was induced with alternating current. The sequential-pulse defibrillation technique using truncated exponential pulses was performed with a right ventricular endocardial catheter and a left ventricular epicardial patch electrode. The first pulse was delivered between the right ventricular apical and the superior vena caval electrode on the right ventricular endocardial catheter. The second pulse was delivered between the right ventricular apical electrode and the left ventricular patch electrode 0.2 ms after termination of the first pulse. Single-pulse defibrillation was performed with a standard intracardiac defibrillation system in which a single truncated exponential pulse was delivered across 2 epicardial patch electrodes positioned over the anterolateral right ventricle and the posterolateral left ventricle.
(Am J Cardiol 1967;60:616-624)
A
lthough the clinical value of the automatic implantable defibrillator for treatment of life-threatening arrhythmias has been demonstrated,l-g problems with the device persist. Defibrillator size, electrode config-
From the Division of Cardiology, Department of Medicine and Division of Cardiovascular Surgery, Department of Surgery, University of Washington, Seattle, Washington. This work was supported in part by grants from the American Heart Association, Washington Chapter, Seattle, Washington; Grant HL3617001 from the National Institutes of Health, Bethesda, Maryland; and Medtronic, Inc., Minneapolis, Minnesota. Manuscript received February 10, 1987; revised manuscript received May 18, 1987, accepted May 24,1987. Address for reprints: Gust H. Bardy, MD, Harborview Medical Center, 325 Ninth Avenue, Seattle, Washington 98104. 818
uration, battery longevity, arrhythmia, detection, defibrillation effectiveness, antitachycardia pacing, lowenergy defibrillation alternatives, and the requirement for major surgery for implantation are aspects of the present defibrillation system that need improvement.@lOOne step toward improving some aspects of the automatic defibrillator concept would be to have an endocardial pacing lead as part of the system. Some efforts have been made to design a defibrillator that delivers its energy by way of a right ventricular endocardial lead system only, but this system has been only partially effective, making such a method too unreliable for clinical use in patients requiring defibrillation.lAll-16
Sequential-pulse defibrillation using a right ventricular catheter in combination with a left ventricular patch electrode bridges part of the gap between the
September
TABLE I
Pt
Clinical
Age (~0 & Sex
Data in 16 Cardiac
Heart Disease
1
72M
CAD, DMI
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
39M 79M 64M 71M 70M 67M 72M 76M 63M 67M 78M 59M 47M 65F 66M
HC MR, CAD CAD, AMI, CABG CAD, AMI CAD, AMI CAD CAD, GABG X 4 CAD CAD, IDC IDC HC IDC, CAD IDC CAD, IDC CAD, AMI
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
BW
Arrest Survivors
LVEF
Drugs at Time of Defib. Testing
VF
0.29
0
VF VT VF VF VF VF VF VF VT VF, AF VF VF VF VF VF
0.57 0.51 0.39 0.51 0.35 0.77 0.62 0.23 0.16 0.14 0.70 0.20 0.59 0.33 0.30
Clinical Arrhythmia
11
CABG, endocardial resection AICD MVR, CABG AICD CABG, AICD patches CABG, AICD patches CABG CABG, AICD patches CABG, AICD patches CABG, AICD patches AICD AICD CABG, AICD AICD CABG, AICD AICD, endocardial resection
Mean
66f
AF = farction: coronary infarction; ejection
atrial fibrillation; AICD = automatic implantable cardioverter-defibrillator; AMI = anterior myocardial inAS = aortic stenosis; AVR = aortic valve replacement; CABG = coronary artery bypass grafting; CAD = artery disease: CHF = congestive heart failure: Defib. = defibrillation; DMI = diaphragmatic myocardial HC = hypertrophic cardiomyopathy; IDC = idiopathic dilated cardiomyopathy; LVEF = left ventricular fraction; MVR = mitral valve replacement; VF = ventricular fibrillation; VT = ventricular tachycardia.
presently available defibrillation system and the ideal one. Using a catheter electrode offers the potential for antitachycardia pacing, bradycardia pacing and, theoretically, low-energy cardioversion. In addition, it can be used for noninvasive programmed stimulation, given appropriate generator design. It also has the potential for combining a hemodynamic arrhythmia detector, right ventricular endocardial impedance, with more conventional rate detection and QRS morphologic detection algorithms.17J” Sequential-pulse defibrillation has been effective in animal studies and in patients with no underlying structural heart disease.1g-2zHowever, it has not been shown to be effective in survivors of cardiac arrest. Furthermore, the transvenous sequential-pulse technique has not been shown to terminate VF with the same reliability as the single-pulse method used in the clinically available automatic implantable defibrillator. The purpose of this investigation was to compare the effectiveness of sequential-pulse defibrillation intraoperatively to the standard single-pulse system in clinically relevant patients and to record electrical measures of defibrillation success for each method.
Methods Sixteen patients gave informed consent for intraoperative determination of defibrillation thresholds during cardiac surgery through a median sternotomy. Fourteen patients survived out-of-hospital VF and 2 had a cardiac arrest from ventricular tachycardia. Thirteen patients underwent implantation of the standard cardioversion/defibrillation system (Cardiac Pacemakers Inc.) in combination with coronary artery bypass grafting; 3 patients underwent coronary artery
0.42 f
Amiodarone 0 0 0 0 0 0 0 0 0 0 0 0 0 Amiodarone
Surgical Procedure
0.20
bypass grafting, endocardial resection or mitral valve replacement as primary therapy. Clinical data for patients who underwent defibrillation threshold testing are listed in Table I. Single-pulse defibrillation with the standard automatic implantable defibrillator used 2 identical rectangular patch electrodes (CPI model L67) placed over the anterior right ventricle and posterolateral left ventricle (Fig. 1, top, and 2, left]. The area of each of these electrodes is 27.9 cm2. The anterior right ventricular patch electrode served as the anode. The sequential-pulse defibrillation system [Medtronic Inc.] consisted of a 10.5Fr defibrillation electrode (Medtronic model 6886) inserted into the innominate vein and positioned manually into the right ventricular apex. Appropriate catheter electrode lead position also was confirmed by demonstrating a left bundle branch block, left-axis QRS morphology during pacing in conjunction with a pacing threshold less than 2.0 V, a pacing resistance of less than 200 Q and an R-wave amplitude greater than 5.0 mV. Lead position was reconfirmed after each defibrillation episode. The sequential-pulse defibrillation catheter has 2 electrode pairs with each element of the pair separated by a pliable joint to improve catheter flexibility and maneuverability (Fig. 1, bottom, and 2, right]. Each electrode has a surface area of 125 mm2. The distal right ventricular apical electrode pair served as the common cathode for both pulses. The proximal electrode pair, positioned at the superior vena cava-right atria1 junction, was separated from the right ventricular apical cathode by 150 mm and served as the anode for the first pulse. The second pulse in the sequentialpulse system was delivered to a 10 cm2 surface area
620
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
oval patch electrode (Medtronic #TX7, Fig. 1, bottom and 2, right). The patch served as the second anode and was placed along the lateral margin of the left ventricle midway between the apex and the base with the long axis of the electrode parallel to the long axis of the left ventricle. The pulse generator for the standard single-pulse defibrillation system was an external battery-powered pulse generator (Intec ECVD] in which pulse amplitude was varied by using energy settings of 1 to 40 J, with 1-J increments possible from the l- to 5-J settings and 5-J increments possible above the 5-J setting. The pulse generator for the sequential-pulse defibrillation system was a battery-powered pulse generator (Medtronic model 23763in which pulse amplitude was varied according to leading-edge voltage settings. Pulse increments or decrements for the sequential-pulse system were made in 100-V steps. All defibrillation threshold determinations were conducted either without hemodynamic support or while the patient was on partial normothermic cardiopulmonary bypass. VF was induced with alternating current applied to the anterior right ventricle. After 10 seconds of VF, a defibrillation pulse was delivered. Before repeat induction of VF, surface electrocardiographic signals and hemodynamic values were allowed to return to baseline.
FIGURE 1. Top, epicardial patch electrodes used In single-pulse defibrillation technique (CPI model L67). Bottom, catheter-patch electrodes used in sequential-pulse defibrillation technique (Medtronic models 6660 and 7X7).
Defibrillation measurements were made in a random sequence with either the single-pulse or the sequential-pulse system tested first. The standard singlepulse method used a truncated exponential waveform where pulse width automatically varied to maintain a constant pulse tilt of 60%. The sequential-pulse method used two 5-ms truncated exponential-pulses with an interpulse separation interval of 0.2 ms. Defibrillation using the sequential-pulse method was first attempted with a 500-V setting for each pulse. If the 500-V setting was successful, pulse amplitude was then decreased by a 100-V decrement and VF was reinduced. This process was repeated until VF could not be terminated with the initial discharge. If the initially used 500-V pulse was unsuccessful, pulse amplitude was increased sequentially until sinus rhythm was restored. After restoration of sinus rhythm, the starting voltage was increased by 100 to 600 V and VF was reinduced. The starting voltage was increased by 100-V increments with each subsequent VF episode until the first pulse in an episode terminated VF. The defibrillation threshold was the lowest voltage that successfully terminated VF with 1 discharge delivered 10 seconds after VF initiation. Determination of the defibrillation threshold for the single-pulse method began by using a 10-J pulse. Pulse amplitude increments or decrements were made by varying energy settings in 5-J steps for pulses between 5 and 40 J and in 1-J steps for pulses between 1 and 5 J. If the initial 10-J setting was successful, pulse amplitude was decreased by a 5-J decrement to 5 J and VF was reinduced. This process was repeated, with subsequent pulse amplitude decreases being in 1-J steps until VF could not be terminated with the initial discharge.
FIGURE 2. retf, standard single-pulse defibrillation technique using 2 large defibrillation patch electrodes placed over the posterolateral left ventricle and the anterolateral right ventricle. Righf, electrodes used for sequential-pulse defibrillatlon technique. An endocardial lead is shown with a right ventricular cathodal electrode and a superior venal caval anodal electrode for delivery of the first pulse. A left ventricular patch serves as the anode for the second pulse. See text for further discussion.
September
TABLE II
Defibrillation
Threshold Voltage
Sequential Puke
Pt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Mean f SD p Value
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
Data
(V)
Current (A) Single Plhe
500 480 490 680 278 365 280 480 477 570 752 380 745 470 470 520 496 f 140
1, 1987
Sequential Pulse
575 590 660 360 185 215 275 220 455 430 420 265 465 145 360 220 365 z!z 157
Single Pulse
7.8 6.7 5.2 5.8 1.6 3.0 2.8 7.5 5.7 6.2 8.5 4.6 11.2 5.6 5.8 7.2 6.0 f- 2.3
<0.005
Impedance
21.0 14.0 13.1 11.2 3.6 5.8 10.4 6.0 10.2 11.3 13.0 5.6 18.2 4.5 15.6 6.3 10.6 f 5.1
current; impedance
If the initial 10-J setting was unsuccessful, a rescue shock was delivered and, upon repeat fibrillation, the starting amplitude was increased by 5 to 15 J. This process of increasing the initial energy setting by 5-J increments was repeated until the first pulse in an episode terminated VF. The defibrillation threshold was the lowest energy that successfully terminated VF with 1 discharge delivered 10 seconds into VF. For both methods of defibrillation, the voltage and current waveforms of the defibrillator pulse were recorded on a Tektronics 2354 digital oscilloscope. Delivered energy was determined by integrating the voltage and current waveforms. Impedance was determined by the equation:
Energy(J)
Single Pulse
70 89 98 97 111 84 75 78 101 111 102 97 84 101 94 72 92 f 13
<0.0005
Voltage is leading-edge voltage: current is leading-edge NS = not significant: SD = standard deviation.
Z = (- t/C)(l/ln
Sequential Pulse
(a)
36 52 59 60 59 42 36 43 59 49 57 56 43 53 36 35 48 f 10
Sequential Pulse 13.1 12.4 11.0 18.8 2.0 4.1 2.5 12.7 11.6 76.2 28.5 7.2 31.1 11.1 11.2 14.9 13.0 + 8.1
<0.005 is mean impedance
calculated
Single Pulse 24.3 23.8 29.5 13.3 2.2 3.0 5.9 3.3 15.2 13.1 17.3 4.7 18.9 1.9 11.2 3.9 12.0 * 8.9 NS
from leading- and trailing-edge
voltages
pulse technique was 496 f 140 V. Mean leading edge voltage for the sequential-pulse system was considerably higher than that for the single-pulse technique (p <0.005] (Fig. 31. The leading-edge current for the single-pulse technique was 10.6 f 5.1 A. That for the sequential-pulse technique was 6.0 f 2.3 A. The mean leading-edge current defibrillation threshold for the sequentialpulse technique was significantly higher than that for the single-pulse technique (p
VT/VL)
where t = pulse width, C = capacitance, Vr = trailingedge voltage and Vt, = leading-edge voltage. Capacitance for the sequential-pulse system is 50 yF and for the single-pulse system, 133 yF. Single-pulse and sequential-pulse leading-edge voltage, leading-edge current, impedance and energy were compared using paired t tests. The 2 values for voltage, current, impedance and energy for the sequential-pulse technique were averaged before comparing with single-pulse values,
DFT Voltage eoo-
Voltage (Volts)
4oo
Results Defibrillation data are summarized in Table II and Figures 3 through 6. The number of VF episodes induced for determining the defibrillation threshold was similar for each method. With the single-pulse system 3.9 f 0.8 VF episodes were induced. With the sequential-pulse system, 4.0 f 0.9 VF episodes were induced. Leading-edge voltage defibrillation threshold for the single-pulse technique was 365 f 157 V. Leading-edge voltage defibrillation threshold for the sequential-
0,
Sequential Pulse
Single Pulse
FIGURE 3. Comparison of defibrillation threshold edge voltage for sequential pulse catheter-patch pulse patch-patch defibrillation methods.
(DFT) leadingand for single-
622
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
technique differed significantly from that for the single-pulse technique (p X0.005) (Fig. 5). The minimum energy needed for defibrillation using the single-pulse technique was 12.0 f 8.9 J and that for defibrillation using the sequential-pulse technique was 13.0 f 8.1 J (both pulse energies combined). There was no statistical difference in energy requirements for either defibrillation method (Fig. 6).
Discussion The standard automatic implantable cardioverterdefibrillator has been available for 7 years without significant alterations in its pulsing method. Although numerous canine studies have been performed in an attempt to improve on the presently available defibrillation system, few human studies have validated the
DFT Current 25
. 20-
. .
.rf + 8 :l
15-
Current (Amps)
;
.
:
lo-
5-
.
0’
I
Sequential Pulse
p40.005
I
Single Pulse
FIGURE 4. Defibrillation threshold (DFT) leading-edge current for sequential pulse catheter-patch and for single-pulse patch-patch defibrillation methods.
,DFT Impedance
..
120 1 100
80
Impedances0
m
DFT Energy
2
H
: .
. a
*
(Ohms)
strengths and weaknesses of any of the newly proposed methods. Those human studies that have been performed have, for the most part, been undertaken in patients with structurally normal hearts.2zWe believe that our study is the first to investigate alternative defibrillation methods in humans with serious structural heart disease and with clinically relevant arrhythmias while also providing information on defibrillation threshold voltage, current, impedance and delivered energy for the standard defibrillation systemas well as for sequential-pulse defibrillation. As such, these data should offer some insight into methods for improving implantable automatic defibrillators. The successor failure of defibrillation is multifactorial and includes lead configuration,8,21-23 lead surface area,8 system impedance,24 position,8y22pulse waveform25r28and anatomic characteristics of the patient’s heart disease.27Defibrillation successor failure is further complicated by the observation that defibrillation is probably a statistical phenomenon, that is, the same defibrillation pulse may terminate VF on some occasionsbut not others.28We believe that in practice, the term “defibrillation threshold” is best considered a liberal approximation of the amount of energy necessaryto restore normal rhythm. A more specific statistical representation of defibrillation effectiveness would require a large number of inductions of VF, which may be difficult in clinical investigation, given the safety constraints in humans. Therefore, as a clinically reasonable alternative approach, we choseto define the “defibrillation threshold” as the lowest amplitude pulse observed to restore sinus rhythm delivered 10 seconds after VF induction. This definition of defibrillation threshold, although incomplete, is useful on a practical level for discriminating relative effectiveness of 2 defibrillation systems. Another problem in defining the parameters of defibrillation effectiveness derives from our lack of understanding of whether energy, voltage, current or some other factor is the key determinant of defibrillation success.Our study illustrates this point. The defibrillation threshold from the perspective of total delivered energy is similar for the 2 methods. However,
.T
t
.T
4r”
40-
20-
0
p
Pulse
Sirlgle Pulse
FIGURE 5. Defibrillation threshold (DFT) impedance for sequentialpulse catheter-patch and for single-pulse patch-patch defibrillation methods.
Pulse
FIGURE 6. Defibrillation threshold pulse and single-pulse defibrillation
Pulse
(DFT) energy methods.
for sequential-
September
there is considerable disparity in leading-edge voltage and in leading-edge current between the single- and sequential-pulse methods. For the most part these differences are dependent on lead system impedance, with the lower impedance, single-pulse patch-patch system yielding higher leading-edge currents and lower leading edge voltages than the higher-impedance, sequential-pulse system. It therefore becomes difficult to explain defibrillation success or failure merely on the basis of how many amperes or volts were applied, especially when both defibrillation methods are effective at similar energies. An additional difficulty in comparing the 2 methods is that the distribution of current with the 2 techniques is probably considerably different. Differences in current flux would render oscilloscopic current waveform values only partially useful when trying to understand the role of current in defibrillation, Considerations of differences in potential gradient with the two techniques would also be difficult without actually knowing tissue electric field strength in a 3-dimensional manner. Such an analysis is not yet possible. Despite the lack of precise tissue measurements of electric field intensity, some more practical considerations can be drawn upon to help assessthe relative effectiveness of the two techniques. For example, previous work suggests that encompassing more ventricular mass within the defibrillation pulse increases the likelihood of restoring sinus rhythrnz3 This would minimize the persistence of islands of fibrillating tissue that could serve as a refibrillation source for tissue that had been defibrillated. Anatomically, an area of particular interest in this regard may be the ventricular septum. The complex relation of septal myocardium to both ventricles and to the Purkinje system may make it more likely to retrigger VF if a small segment of septum continues to fibrillate in the short period immediately after the defibrillation pulse. Such anatomic considerations may account for the sequential-pulse technique’s relative effectiveness despite lower currents. Alternately, one might explain effectiveness for the sequential-pulse method on the basis of orthogonal currents. Despite the lower current with the sequential-pulse technique than with the single-pulse method, the orthogonality of the 2 currents may promote defibrillation more readily than a similar, more unidirectional current. Patients with different forms of heart disease may have different responses when subjected to electrically dissimilar pulses. For example, patients with diaphragmatic infarcts may respond more favorably to a particular pulse than patients with anterior infarcts even though comparable current densities occur throughout the myocardium. Marked disparity in total energy needed for defibrillation in some of our patients suggests that anatomic factors may indeed alter responses to different pulses. By implication, one may infer that the optimal defibrillation method might vary between patients. From the practical vantage of patient care, the catheter-patch sequential-pulse defibrillation method has several advantages over the standard single-pulse
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
623
method. Endocardial bradycardia and antitachycardia programmed pacing would be available. This might obviate the need for 2 devices in patients with associated bradycardia. Such an option may also be used to interrupt monomorphic ventricular tachycardia in patients who manifest a relatively stable ventricular arrhythmia before development of VF, perhaps avoiding the need for defibrillation. In addition, this system may be used to perform noninvasive programmed stimulation for drug evaluation or for testing device effectiveness thereby eliminating the need for recurrent catheterizations. Furthermore, an endocardial right ventricular lead system would provide the opportunity to use a hemodynamic sensor such as impedance,17J8Such an addition would improve arrhythmia detection algorithms, Finally, the right ventricular catheter-epicardial patch configuration would facilitate further progress toward the goal of a totally transvenous defibrillator.
References 1. Mirowski M, Mower MM, Gott VL, Brawley RK, Denniston R. Transvenous automatic defibrillator-preliminary clinical tests of its defibrillating subsvstem. Trank Am Sot Artif Intern~Organs 1972;18:520-525. i Mirowski M. Mower MM. Gott VL. Bmwlev RK. Feasibihtv and effectiveness of low-energy catheterhefibrillation in man. Circulation 1973;h9-85. 3. Mirowski M. Treatment of malignant ventricular tachyarrhythmias with the automatic implantable defibrillator. Int J CardioJ X983:2:409-413. 4. Mirowski M. Reid PR. Mower MM. Watkins L, Gott VL, Schauble IF, Langer A, Heilman MS, Kolenik SA. Fischell RE, Weisfeldt ML. Termination of malignant ventricular arrhythmias with an implanted automatic defibrillator in human beings. N EnnJ I Med 1989:303:322-324. 5. Reid PR, Miroiski M, Mower MM, Platia EV, Griffith LSC, Watkins L, Bach SM, Imran M, Thomas A. Clinical evaluation of the internal automatic cardioverter/defibrilJator. Am J CardioJ 1983;51:1608-1613, 6. Winkle RA, Bach SM, Echt DS, Mead RH, Schmidt P. Practical aspects of automatic cardioverter/defibriJJator implantation. Am Heart r 1984;108: 1335-1346.
7. Echt DS, Armstrong K, Schmidt P, Oyer PE, Stinson EB, Winkle RA. Clinical exuerience. comolications, and survival in 70 patients with the automatic impjantable cardidverter/defibriJJator. Circulation 1985;71:289-296. 8. Troup PJ, Chapman PD, Olinger GN, Kleinman LH. The implanted defibrillator: relation of defibrillating lead configuration and clinical variables to defibrillation threshold. JACC 1%35;6:1315-i321. 9. Marchlinski FE, Flores BT, Buxton AE, Hat-grove WC, Addonizio VP, Stephenson LW, Harken AH, Doherty JU, Grogan EW, Josephson ME. The automatic implantable cardioverter-defibriJJator: efficacy, complications, and device failures. Ann Jntern Med 3986;104:481-488. 10. Bardy GH, Ivey TD, Stewart RB, Graham E, Greene HL. Failure of the automatic implantable defibrillator to detect ventricular fibrillation. Am J CardioJ 1986;58:1107-1108. 11. Tackman WM. Zincs DP. Low-enernv svnchronous cardioversion of ventricular tachycardia king a catheter electrode in a canine model of subacute myocardial infarction. Circulation 1982;66:187-194. 12. Yee R, Zincs DP, Gulamhusein S, Kallok MJ, Klein GJ. Low energy countershock &ing an intravascular catheter in an acute cardiac care setting. Am J CardioJ 1982;50:1124-3129. 13. Waspe LE, Kim SG, Matos JA, Fisher JD. Role of catheter lead system for transvenous countershock and pacing during electrophysiologic tests: an assessment of the usefulness of catheter shocks for terminating ventricular tachvarrhvthmias. Am T Cardiol 1983:52:477-484, 14. biccone JM, Saksena S, Shah Y, Pantopoulos D. A prospective randomized study of the clinical efficacy and safety of transvenous cardioversion for termination of ventricular tachycardia. Circulation 2985;71:571-578. 15. Saksena S, Chandran P, Shah Y, Boccadamo R, Pantopoulos D, Rothbart ST. Comparative efficacy of transvenous cardioversion and pacing in patients with sustained ventricular tachycardia: a prospective, randomized, crossover study. Circulation 1985;72:253-160. 16. Zipes DP, Heger JJ, Miles WM, Mahomed Y, Brown JW, Spielman SR, Prvstowskv EN. Early experience with an implantable cardioverter. N EngJJ M”ed 1984!311:485-490. 17. Olson WH, Miles WM, Zipes DP, Prystowsky EN. Intracardiac electrical impedance during ventricular tachycardia and ventricular fibrillation in man [abstr). JACC 1985;5:506. 18. Bardy GH, Olson WH, Fishbein DP, Fellows CL, Coltorti F, Weaver WD, Greene HL. Transvenous right ventricular impedance during spontaneous ventricular arrhvthmias in man tabs&l. Circulation 1985;72:suppI 111:111-474.
624
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
19. Holmes HR, Bourland JD, Tacker WA Jr, Geddes LA. Hemodynamic responses to two defibrillating trapezoidal waveforms. Med Instrument 1980:14:47-50. 20. Wessale JL, Bourland JD, Tacker WA, Geddes LA. Bipolar catheter defibrillation in dogs using trapezoidal waveforms of various tilts. J EIectrocardiol 1980;13:359-366.
21. TonesDL, Klein GJ, Kallok MJ, Improved internal defibrillation with twin pulse sequential energy delivej to different lead orientations in pigs. Am J Cardiol 1985;55:821-825. 22. Jones DL, Klein GJ, Guiraudon GM, Sharma AD, Kallok MJ, Bourland JD, Tacker WA. Internal cardiac pulse defibrillation in man: pronounced improvement with sequential pulse delivery to two different lead orientations. Circulation 1986;73:484-491. 23. Kallok MJ, Bourland JD, Tacker WA, Jones DL, Klein GJ, Wessale JL. Effect of epicardial electrode size and implant location on defibrillation
threshold using a new sequential pulse technique (abstr). Proc Am Assoc Adv Med Instrum 1985;20:46. 24. Tacker WA, Mercer J, Foley P. Resistivity of skeletal muscle, skin, fat and lung to defibrillation shocks [abstr]. Proc Assoc Adv Med Instrum 1984;19:81. 25. Jones JL, Jones RE. Decreased defibrillator-induced dysfunction with biphasic rectangular waveforms. Am r Physiol 1984;247:H792-H796, 26. Niebauer MJ, Babbs CF, Geddes LA, Bourland JD. Efficacy and safety of defibrillation with rectangular waves of z to 20 milliseconds duration. Crit Care Med 1983;11:95-98. 27. Chapman PD, Sagar KG, Wetherbee JN, Troup PJ. Relationship of left ventricular massto defibrillation threshold for the implantable defibrillator: a combined clinical and animal study. /ACC 1987, in press. 28. Weaver WD, Cobb LA, Copass MK, Hallstrom AP. Ventricular defibrillation-comparative trial using 175 and 320 Joule shocks. N &gI J Med 1982;307:1101-1106.
During defibrillation threshold determination, voltage and current waveforms were recorded and integrated to determine delivered energy. Average defibrillation threshold leading-edge voltage for the sequential pulse technique was 496 f 140 V, compared with 365 f 157 V for the single-pulse technique (p <0.005). Defibrillation threshold leadingedge current for the sequential-pulse technique was 6.0 f 2.3 A, compared with 10.6 f 5.1 A for the single-pulse method (p
Sixteen survivors of cardiac arrest underwent intraoperative comparison of the effectiveness of sequentiaf-pulse and single-pulse defibrillation. Defibrillation was tested alternately with the single-pulse or sequential-pulse technique 10 seconds into an episode of ventricular fibrillation that was induced with alternating current. The sequential-pulse defibrillation technique using truncated exponential pulses was performed with a right ventricular endocardial catheter and a left ventricular epicardial patch electrode. The first pulse was delivered between the right ventricular apical and the superior vena caval electrode on the right ventricular endocardial catheter. The second pulse was delivered between the right ventricular apical electrode and the left ventricular patch electrode 0.2 ms after termination of the first pulse. Single-pulse defibrillation was performed with a standard intracardiac defibrillation system in which a single truncated exponential pulse was delivered across 2 epicardial patch electrodes positioned over the anterolateral right ventricle and the posterolateral left ventricle.
(Am J Cardiol 1967;60:616-624)
A
lthough the clinical value of the automatic implantable defibrillator for treatment of life-threatening arrhythmias has been demonstrated,l-g problems with the device persist. Defibrillator size, electrode config-
From the Division of Cardiology, Department of Medicine and Division of Cardiovascular Surgery, Department of Surgery, University of Washington, Seattle, Washington. This work was supported in part by grants from the American Heart Association, Washington Chapter, Seattle, Washington; Grant HL3617001 from the National Institutes of Health, Bethesda, Maryland; and Medtronic, Inc., Minneapolis, Minnesota. Manuscript received February 10, 1987; revised manuscript received May 18, 1987, accepted May 24,1987. Address for reprints: Gust H. Bardy, MD, Harborview Medical Center, 325 Ninth Avenue, Seattle, Washington 98104. 818
uration, battery longevity, arrhythmia, detection, defibrillation effectiveness, antitachycardia pacing, lowenergy defibrillation alternatives, and the requirement for major surgery for implantation are aspects of the present defibrillation system that need improvement.@lOOne step toward improving some aspects of the automatic defibrillator concept would be to have an endocardial pacing lead as part of the system. Some efforts have been made to design a defibrillator that delivers its energy by way of a right ventricular endocardial lead system only, but this system has been only partially effective, making such a method too unreliable for clinical use in patients requiring defibrillation.lAll-16
Sequential-pulse defibrillation using a right ventricular catheter in combination with a left ventricular patch electrode bridges part of the gap between the
September
TABLE I
Pt
Clinical
Age (~0 & Sex
Data in 16 Cardiac
Heart Disease
1
72M
CAD, DMI
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
39M 79M 64M 71M 70M 67M 72M 76M 63M 67M 78M 59M 47M 65F 66M
HC MR, CAD CAD, AMI, CABG CAD, AMI CAD, AMI CAD CAD, GABG X 4 CAD CAD, IDC IDC HC IDC, CAD IDC CAD, IDC CAD, AMI
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
BW
Arrest Survivors
LVEF
Drugs at Time of Defib. Testing
VF
0.29
0
VF VT VF VF VF VF VF VF VT VF, AF VF VF VF VF VF
0.57 0.51 0.39 0.51 0.35 0.77 0.62 0.23 0.16 0.14 0.70 0.20 0.59 0.33 0.30
Clinical Arrhythmia
11
CABG, endocardial resection AICD MVR, CABG AICD CABG, AICD patches CABG, AICD patches CABG CABG, AICD patches CABG, AICD patches CABG, AICD patches AICD AICD CABG, AICD AICD CABG, AICD AICD, endocardial resection
Mean
66f
AF = farction: coronary infarction; ejection
atrial fibrillation; AICD = automatic implantable cardioverter-defibrillator; AMI = anterior myocardial inAS = aortic stenosis; AVR = aortic valve replacement; CABG = coronary artery bypass grafting; CAD = artery disease: CHF = congestive heart failure: Defib. = defibrillation; DMI = diaphragmatic myocardial HC = hypertrophic cardiomyopathy; IDC = idiopathic dilated cardiomyopathy; LVEF = left ventricular fraction; MVR = mitral valve replacement; VF = ventricular fibrillation; VT = ventricular tachycardia.
presently available defibrillation system and the ideal one. Using a catheter electrode offers the potential for antitachycardia pacing, bradycardia pacing and, theoretically, low-energy cardioversion. In addition, it can be used for noninvasive programmed stimulation, given appropriate generator design. It also has the potential for combining a hemodynamic arrhythmia detector, right ventricular endocardial impedance, with more conventional rate detection and QRS morphologic detection algorithms.17J” Sequential-pulse defibrillation has been effective in animal studies and in patients with no underlying structural heart disease.1g-2zHowever, it has not been shown to be effective in survivors of cardiac arrest. Furthermore, the transvenous sequential-pulse technique has not been shown to terminate VF with the same reliability as the single-pulse method used in the clinically available automatic implantable defibrillator. The purpose of this investigation was to compare the effectiveness of sequential-pulse defibrillation intraoperatively to the standard single-pulse system in clinically relevant patients and to record electrical measures of defibrillation success for each method.
Methods Sixteen patients gave informed consent for intraoperative determination of defibrillation thresholds during cardiac surgery through a median sternotomy. Fourteen patients survived out-of-hospital VF and 2 had a cardiac arrest from ventricular tachycardia. Thirteen patients underwent implantation of the standard cardioversion/defibrillation system (Cardiac Pacemakers Inc.) in combination with coronary artery bypass grafting; 3 patients underwent coronary artery
0.42 f
Amiodarone 0 0 0 0 0 0 0 0 0 0 0 0 0 Amiodarone
Surgical Procedure
0.20
bypass grafting, endocardial resection or mitral valve replacement as primary therapy. Clinical data for patients who underwent defibrillation threshold testing are listed in Table I. Single-pulse defibrillation with the standard automatic implantable defibrillator used 2 identical rectangular patch electrodes (CPI model L67) placed over the anterior right ventricle and posterolateral left ventricle (Fig. 1, top, and 2, left]. The area of each of these electrodes is 27.9 cm2. The anterior right ventricular patch electrode served as the anode. The sequential-pulse defibrillation system [Medtronic Inc.] consisted of a 10.5Fr defibrillation electrode (Medtronic model 6886) inserted into the innominate vein and positioned manually into the right ventricular apex. Appropriate catheter electrode lead position also was confirmed by demonstrating a left bundle branch block, left-axis QRS morphology during pacing in conjunction with a pacing threshold less than 2.0 V, a pacing resistance of less than 200 Q and an R-wave amplitude greater than 5.0 mV. Lead position was reconfirmed after each defibrillation episode. The sequential-pulse defibrillation catheter has 2 electrode pairs with each element of the pair separated by a pliable joint to improve catheter flexibility and maneuverability (Fig. 1, bottom, and 2, right]. Each electrode has a surface area of 125 mm2. The distal right ventricular apical electrode pair served as the common cathode for both pulses. The proximal electrode pair, positioned at the superior vena cava-right atria1 junction, was separated from the right ventricular apical cathode by 150 mm and served as the anode for the first pulse. The second pulse in the sequentialpulse system was delivered to a 10 cm2 surface area
620
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
oval patch electrode (Medtronic #TX7, Fig. 1, bottom and 2, right). The patch served as the second anode and was placed along the lateral margin of the left ventricle midway between the apex and the base with the long axis of the electrode parallel to the long axis of the left ventricle. The pulse generator for the standard single-pulse defibrillation system was an external battery-powered pulse generator (Intec ECVD] in which pulse amplitude was varied by using energy settings of 1 to 40 J, with 1-J increments possible from the l- to 5-J settings and 5-J increments possible above the 5-J setting. The pulse generator for the sequential-pulse defibrillation system was a battery-powered pulse generator (Medtronic model 23763in which pulse amplitude was varied according to leading-edge voltage settings. Pulse increments or decrements for the sequential-pulse system were made in 100-V steps. All defibrillation threshold determinations were conducted either without hemodynamic support or while the patient was on partial normothermic cardiopulmonary bypass. VF was induced with alternating current applied to the anterior right ventricle. After 10 seconds of VF, a defibrillation pulse was delivered. Before repeat induction of VF, surface electrocardiographic signals and hemodynamic values were allowed to return to baseline.
FIGURE 1. Top, epicardial patch electrodes used In single-pulse defibrillation technique (CPI model L67). Bottom, catheter-patch electrodes used in sequential-pulse defibrillation technique (Medtronic models 6660 and 7X7).
Defibrillation measurements were made in a random sequence with either the single-pulse or the sequential-pulse system tested first. The standard singlepulse method used a truncated exponential waveform where pulse width automatically varied to maintain a constant pulse tilt of 60%. The sequential-pulse method used two 5-ms truncated exponential-pulses with an interpulse separation interval of 0.2 ms. Defibrillation using the sequential-pulse method was first attempted with a 500-V setting for each pulse. If the 500-V setting was successful, pulse amplitude was then decreased by a 100-V decrement and VF was reinduced. This process was repeated until VF could not be terminated with the initial discharge. If the initially used 500-V pulse was unsuccessful, pulse amplitude was increased sequentially until sinus rhythm was restored. After restoration of sinus rhythm, the starting voltage was increased by 100 to 600 V and VF was reinduced. The starting voltage was increased by 100-V increments with each subsequent VF episode until the first pulse in an episode terminated VF. The defibrillation threshold was the lowest voltage that successfully terminated VF with 1 discharge delivered 10 seconds after VF initiation. Determination of the defibrillation threshold for the single-pulse method began by using a 10-J pulse. Pulse amplitude increments or decrements were made by varying energy settings in 5-J steps for pulses between 5 and 40 J and in 1-J steps for pulses between 1 and 5 J. If the initial 10-J setting was successful, pulse amplitude was decreased by a 5-J decrement to 5 J and VF was reinduced. This process was repeated, with subsequent pulse amplitude decreases being in 1-J steps until VF could not be terminated with the initial discharge.
FIGURE 2. retf, standard single-pulse defibrillation technique using 2 large defibrillation patch electrodes placed over the posterolateral left ventricle and the anterolateral right ventricle. Righf, electrodes used for sequential-pulse defibrillatlon technique. An endocardial lead is shown with a right ventricular cathodal electrode and a superior venal caval anodal electrode for delivery of the first pulse. A left ventricular patch serves as the anode for the second pulse. See text for further discussion.
September
TABLE II
Defibrillation
Threshold Voltage
Sequential Puke
Pt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Mean f SD p Value
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
Data
(V)
Current (A) Single Plhe
500 480 490 680 278 365 280 480 477 570 752 380 745 470 470 520 496 f 140
1, 1987
Sequential Pulse
575 590 660 360 185 215 275 220 455 430 420 265 465 145 360 220 365 z!z 157
Single Pulse
7.8 6.7 5.2 5.8 1.6 3.0 2.8 7.5 5.7 6.2 8.5 4.6 11.2 5.6 5.8 7.2 6.0 f- 2.3
<0.005
Impedance
21.0 14.0 13.1 11.2 3.6 5.8 10.4 6.0 10.2 11.3 13.0 5.6 18.2 4.5 15.6 6.3 10.6 f 5.1
current; impedance
If the initial 10-J setting was unsuccessful, a rescue shock was delivered and, upon repeat fibrillation, the starting amplitude was increased by 5 to 15 J. This process of increasing the initial energy setting by 5-J increments was repeated until the first pulse in an episode terminated VF. The defibrillation threshold was the lowest energy that successfully terminated VF with 1 discharge delivered 10 seconds into VF. For both methods of defibrillation, the voltage and current waveforms of the defibrillator pulse were recorded on a Tektronics 2354 digital oscilloscope. Delivered energy was determined by integrating the voltage and current waveforms. Impedance was determined by the equation:
Energy(J)
Single Pulse
70 89 98 97 111 84 75 78 101 111 102 97 84 101 94 72 92 f 13
<0.0005
Voltage is leading-edge voltage: current is leading-edge NS = not significant: SD = standard deviation.
Z = (- t/C)(l/ln
Sequential Pulse
(a)
36 52 59 60 59 42 36 43 59 49 57 56 43 53 36 35 48 f 10
Sequential Pulse 13.1 12.4 11.0 18.8 2.0 4.1 2.5 12.7 11.6 76.2 28.5 7.2 31.1 11.1 11.2 14.9 13.0 + 8.1
<0.005 is mean impedance
calculated
Single Pulse 24.3 23.8 29.5 13.3 2.2 3.0 5.9 3.3 15.2 13.1 17.3 4.7 18.9 1.9 11.2 3.9 12.0 * 8.9 NS
from leading- and trailing-edge
voltages
pulse technique was 496 f 140 V. Mean leading edge voltage for the sequential-pulse system was considerably higher than that for the single-pulse technique (p <0.005] (Fig. 31. The leading-edge current for the single-pulse technique was 10.6 f 5.1 A. That for the sequential-pulse technique was 6.0 f 2.3 A. The mean leading-edge current defibrillation threshold for the sequentialpulse technique was significantly higher than that for the single-pulse technique (p
VT/VL)
where t = pulse width, C = capacitance, Vr = trailingedge voltage and Vt, = leading-edge voltage. Capacitance for the sequential-pulse system is 50 yF and for the single-pulse system, 133 yF. Single-pulse and sequential-pulse leading-edge voltage, leading-edge current, impedance and energy were compared using paired t tests. The 2 values for voltage, current, impedance and energy for the sequential-pulse technique were averaged before comparing with single-pulse values,
DFT Voltage eoo-
Voltage (Volts)
4oo
Results Defibrillation data are summarized in Table II and Figures 3 through 6. The number of VF episodes induced for determining the defibrillation threshold was similar for each method. With the single-pulse system 3.9 f 0.8 VF episodes were induced. With the sequential-pulse system, 4.0 f 0.9 VF episodes were induced. Leading-edge voltage defibrillation threshold for the single-pulse technique was 365 f 157 V. Leading-edge voltage defibrillation threshold for the sequential-
0,
Sequential Pulse
Single Pulse
FIGURE 3. Comparison of defibrillation threshold edge voltage for sequential pulse catheter-patch pulse patch-patch defibrillation methods.
(DFT) leadingand for single-
622
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
technique differed significantly from that for the single-pulse technique (p X0.005) (Fig. 5). The minimum energy needed for defibrillation using the single-pulse technique was 12.0 f 8.9 J and that for defibrillation using the sequential-pulse technique was 13.0 f 8.1 J (both pulse energies combined). There was no statistical difference in energy requirements for either defibrillation method (Fig. 6).
Discussion The standard automatic implantable cardioverterdefibrillator has been available for 7 years without significant alterations in its pulsing method. Although numerous canine studies have been performed in an attempt to improve on the presently available defibrillation system, few human studies have validated the
DFT Current 25
. 20-
. .
.rf + 8 :l
15-
Current (Amps)
;
.
:
lo-
5-
.
0’
I
Sequential Pulse
p40.005
I
Single Pulse
FIGURE 4. Defibrillation threshold (DFT) leading-edge current for sequential pulse catheter-patch and for single-pulse patch-patch defibrillation methods.
,DFT Impedance
..
120 1 100
80
Impedances0
m
DFT Energy
2
H
: .
. a
*
(Ohms)
strengths and weaknesses of any of the newly proposed methods. Those human studies that have been performed have, for the most part, been undertaken in patients with structurally normal hearts.2zWe believe that our study is the first to investigate alternative defibrillation methods in humans with serious structural heart disease and with clinically relevant arrhythmias while also providing information on defibrillation threshold voltage, current, impedance and delivered energy for the standard defibrillation systemas well as for sequential-pulse defibrillation. As such, these data should offer some insight into methods for improving implantable automatic defibrillators. The successor failure of defibrillation is multifactorial and includes lead configuration,8,21-23 lead surface area,8 system impedance,24 position,8y22pulse waveform25r28and anatomic characteristics of the patient’s heart disease.27Defibrillation successor failure is further complicated by the observation that defibrillation is probably a statistical phenomenon, that is, the same defibrillation pulse may terminate VF on some occasionsbut not others.28We believe that in practice, the term “defibrillation threshold” is best considered a liberal approximation of the amount of energy necessaryto restore normal rhythm. A more specific statistical representation of defibrillation effectiveness would require a large number of inductions of VF, which may be difficult in clinical investigation, given the safety constraints in humans. Therefore, as a clinically reasonable alternative approach, we choseto define the “defibrillation threshold” as the lowest amplitude pulse observed to restore sinus rhythm delivered 10 seconds after VF induction. This definition of defibrillation threshold, although incomplete, is useful on a practical level for discriminating relative effectiveness of 2 defibrillation systems. Another problem in defining the parameters of defibrillation effectiveness derives from our lack of understanding of whether energy, voltage, current or some other factor is the key determinant of defibrillation success.Our study illustrates this point. The defibrillation threshold from the perspective of total delivered energy is similar for the 2 methods. However,
.T
t
.T
4r”
40-
20-
0
p
Pulse
Sirlgle Pulse
FIGURE 5. Defibrillation threshold (DFT) impedance for sequentialpulse catheter-patch and for single-pulse patch-patch defibrillation methods.
Pulse
FIGURE 6. Defibrillation threshold pulse and single-pulse defibrillation
Pulse
(DFT) energy methods.
for sequential-
September
there is considerable disparity in leading-edge voltage and in leading-edge current between the single- and sequential-pulse methods. For the most part these differences are dependent on lead system impedance, with the lower impedance, single-pulse patch-patch system yielding higher leading-edge currents and lower leading edge voltages than the higher-impedance, sequential-pulse system. It therefore becomes difficult to explain defibrillation success or failure merely on the basis of how many amperes or volts were applied, especially when both defibrillation methods are effective at similar energies. An additional difficulty in comparing the 2 methods is that the distribution of current with the 2 techniques is probably considerably different. Differences in current flux would render oscilloscopic current waveform values only partially useful when trying to understand the role of current in defibrillation, Considerations of differences in potential gradient with the two techniques would also be difficult without actually knowing tissue electric field strength in a 3-dimensional manner. Such an analysis is not yet possible. Despite the lack of precise tissue measurements of electric field intensity, some more practical considerations can be drawn upon to help assessthe relative effectiveness of the two techniques. For example, previous work suggests that encompassing more ventricular mass within the defibrillation pulse increases the likelihood of restoring sinus rhythrnz3 This would minimize the persistence of islands of fibrillating tissue that could serve as a refibrillation source for tissue that had been defibrillated. Anatomically, an area of particular interest in this regard may be the ventricular septum. The complex relation of septal myocardium to both ventricles and to the Purkinje system may make it more likely to retrigger VF if a small segment of septum continues to fibrillate in the short period immediately after the defibrillation pulse. Such anatomic considerations may account for the sequential-pulse technique’s relative effectiveness despite lower currents. Alternately, one might explain effectiveness for the sequential-pulse method on the basis of orthogonal currents. Despite the lower current with the sequential-pulse technique than with the single-pulse method, the orthogonality of the 2 currents may promote defibrillation more readily than a similar, more unidirectional current. Patients with different forms of heart disease may have different responses when subjected to electrically dissimilar pulses. For example, patients with diaphragmatic infarcts may respond more favorably to a particular pulse than patients with anterior infarcts even though comparable current densities occur throughout the myocardium. Marked disparity in total energy needed for defibrillation in some of our patients suggests that anatomic factors may indeed alter responses to different pulses. By implication, one may infer that the optimal defibrillation method might vary between patients. From the practical vantage of patient care, the catheter-patch sequential-pulse defibrillation method has several advantages over the standard single-pulse
1, 1987
THE AMERICAN
JOURNAL
OF CARDIOLOGY
Volume 60
623
method. Endocardial bradycardia and antitachycardia programmed pacing would be available. This might obviate the need for 2 devices in patients with associated bradycardia. Such an option may also be used to interrupt monomorphic ventricular tachycardia in patients who manifest a relatively stable ventricular arrhythmia before development of VF, perhaps avoiding the need for defibrillation. In addition, this system may be used to perform noninvasive programmed stimulation for drug evaluation or for testing device effectiveness thereby eliminating the need for recurrent catheterizations. Furthermore, an endocardial right ventricular lead system would provide the opportunity to use a hemodynamic sensor such as impedance,17J8Such an addition would improve arrhythmia detection algorithms, Finally, the right ventricular catheter-epicardial patch configuration would facilitate further progress toward the goal of a totally transvenous defibrillator.
References 1. Mirowski M, Mower MM, Gott VL, Brawley RK, Denniston R. Transvenous automatic defibrillator-preliminary clinical tests of its defibrillating subsvstem. Trank Am Sot Artif Intern~Organs 1972;18:520-525. i Mirowski M. Mower MM. Gott VL. Bmwlev RK. Feasibihtv and effectiveness of low-energy catheterhefibrillation in man. Circulation 1973;h9-85. 3. Mirowski M. Treatment of malignant ventricular tachyarrhythmias with the automatic implantable defibrillator. Int J CardioJ X983:2:409-413. 4. Mirowski M. Reid PR. Mower MM. Watkins L, Gott VL, Schauble IF, Langer A, Heilman MS, Kolenik SA. Fischell RE, Weisfeldt ML. Termination of malignant ventricular arrhythmias with an implanted automatic defibrillator in human beings. N EnnJ I Med 1989:303:322-324. 5. Reid PR, Miroiski M, Mower MM, Platia EV, Griffith LSC, Watkins L, Bach SM, Imran M, Thomas A. Clinical evaluation of the internal automatic cardioverter/defibrilJator. Am J CardioJ 1983;51:1608-1613, 6. Winkle RA, Bach SM, Echt DS, Mead RH, Schmidt P. Practical aspects of automatic cardioverter/defibriJJator implantation. Am Heart r 1984;108: 1335-1346.
7. Echt DS, Armstrong K, Schmidt P, Oyer PE, Stinson EB, Winkle RA. Clinical exuerience. comolications, and survival in 70 patients with the automatic impjantable cardidverter/defibriJJator. Circulation 1985;71:289-296. 8. Troup PJ, Chapman PD, Olinger GN, Kleinman LH. The implanted defibrillator: relation of defibrillating lead configuration and clinical variables to defibrillation threshold. JACC 1%35;6:1315-i321. 9. Marchlinski FE, Flores BT, Buxton AE, Hat-grove WC, Addonizio VP, Stephenson LW, Harken AH, Doherty JU, Grogan EW, Josephson ME. The automatic implantable cardioverter-defibriJJator: efficacy, complications, and device failures. Ann Jntern Med 3986;104:481-488. 10. Bardy GH, Ivey TD, Stewart RB, Graham E, Greene HL. Failure of the automatic implantable defibrillator to detect ventricular fibrillation. Am J CardioJ 1986;58:1107-1108. 11. Tackman WM. Zincs DP. Low-enernv svnchronous cardioversion of ventricular tachycardia king a catheter electrode in a canine model of subacute myocardial infarction. Circulation 1982;66:187-194. 12. Yee R, Zincs DP, Gulamhusein S, Kallok MJ, Klein GJ. Low energy countershock &ing an intravascular catheter in an acute cardiac care setting. Am J CardioJ 1982;50:1124-3129. 13. Waspe LE, Kim SG, Matos JA, Fisher JD. Role of catheter lead system for transvenous countershock and pacing during electrophysiologic tests: an assessment of the usefulness of catheter shocks for terminating ventricular tachvarrhvthmias. Am T Cardiol 1983:52:477-484, 14. biccone JM, Saksena S, Shah Y, Pantopoulos D. A prospective randomized study of the clinical efficacy and safety of transvenous cardioversion for termination of ventricular tachycardia. Circulation 2985;71:571-578. 15. Saksena S, Chandran P, Shah Y, Boccadamo R, Pantopoulos D, Rothbart ST. Comparative efficacy of transvenous cardioversion and pacing in patients with sustained ventricular tachycardia: a prospective, randomized, crossover study. Circulation 1985;72:253-160. 16. Zipes DP, Heger JJ, Miles WM, Mahomed Y, Brown JW, Spielman SR, Prvstowskv EN. Early experience with an implantable cardioverter. N EngJJ M”ed 1984!311:485-490. 17. Olson WH, Miles WM, Zipes DP, Prystowsky EN. Intracardiac electrical impedance during ventricular tachycardia and ventricular fibrillation in man [abstr). JACC 1985;5:506. 18. Bardy GH, Olson WH, Fishbein DP, Fellows CL, Coltorti F, Weaver WD, Greene HL. Transvenous right ventricular impedance during spontaneous ventricular arrhvthmias in man tabs&l. Circulation 1985;72:suppI 111:111-474.
624
SEQUENTIAL-
AND SINGLE-PULSE
DEFIBRILLATION
19. Holmes HR, Bourland JD, Tacker WA Jr, Geddes LA. Hemodynamic responses to two defibrillating trapezoidal waveforms. Med Instrument 1980:14:47-50. 20. Wessale JL, Bourland JD, Tacker WA, Geddes LA. Bipolar catheter defibrillation in dogs using trapezoidal waveforms of various tilts. J EIectrocardiol 1980;13:359-366.
21. TonesDL, Klein GJ, Kallok MJ, Improved internal defibrillation with twin pulse sequential energy delivej to different lead orientations in pigs. Am J Cardiol 1985;55:821-825. 22. Jones DL, Klein GJ, Guiraudon GM, Sharma AD, Kallok MJ, Bourland JD, Tacker WA. Internal cardiac pulse defibrillation in man: pronounced improvement with sequential pulse delivery to two different lead orientations. Circulation 1986;73:484-491. 23. Kallok MJ, Bourland JD, Tacker WA, Jones DL, Klein GJ, Wessale JL. Effect of epicardial electrode size and implant location on defibrillation
threshold using a new sequential pulse technique (abstr). Proc Am Assoc Adv Med Instrum 1985;20:46. 24. Tacker WA, Mercer J, Foley P. Resistivity of skeletal muscle, skin, fat and lung to defibrillation shocks [abstr]. Proc Assoc Adv Med Instrum 1984;19:81. 25. Jones JL, Jones RE. Decreased defibrillator-induced dysfunction with biphasic rectangular waveforms. Am r Physiol 1984;247:H792-H796, 26. Niebauer MJ, Babbs CF, Geddes LA, Bourland JD. Efficacy and safety of defibrillation with rectangular waves of z to 20 milliseconds duration. Crit Care Med 1983;11:95-98. 27. Chapman PD, Sagar KG, Wetherbee JN, Troup PJ. Relationship of left ventricular massto defibrillation threshold for the implantable defibrillator: a combined clinical and animal study. /ACC 1987, in press. 28. Weaver WD, Cobb LA, Copass MK, Hallstrom AP. Ventricular defibrillation-comparative trial using 175 and 320 Joule shocks. N &gI J Med 1982;307:1101-1106.