Comparison of Monophasic and Biphasic Defibrillating Pulse Waveforms for

Transthoracic Cardioversion H. Leon Greene, MD, John P. DiMarco, MD, PhD, Peter J. Kudenchuk, MD,

Melvin M. Scheinman, MD, Anthony S.L. Tang, MD, Michael J. Reiter, MD, PhD, Debra S. Echt, MD, Peter D. Chapman, MD, Mohammad R. Jaza eri, MD,

Fred W. Chapman, PhD, Mahboob Ahmed, MSEE, Jeffrey L. Jo ii nson, Robert A. Niskanen, MSEE, and The Biphasic Waveform Defibrillation Investigators

All transthoracic defibrillators on the US market use nominally monophasic shock waveforms. However, bi- phasic waveforms have a lower defibrillation thresh- old than monophasic waveforms for transthoracic de- fibrillation of animals and for defibrillation of humans b i

implantable cardioverter defibrillators. The relative e icacies of Edmark monophasic and Gurvich biphasic transthoracic cardioversion waveforms (200 J into 50 Q) were compared for transthoracic cardioversion in 171 patients undergoing electro evaluation of ventricular arrh YtR

hysiologic study for mias. Patients were

randomized in a blinded fashion to receive either a monophasic or a biphasic waveform for the initial shock for conversion of induced ventricular arrhythmias (ven- tricular fibrillation [VFI = 53, monomorphic ventricular tach cardia [VT] = 80, polymorphic VT = 30, ventricu- lar fktter = 8). D e rvered energies for the Edmark and I’ Gurvich waveforms were 215 f 11 and 171 + 11 J,

respectively. There were no significant differences in patient characteristics, use of antiarrh rhythmia cycle len th, or duration o

f P

mic agents, ar- arrh

r hmia prior

to shock for monop asic and biphasic wave orm groups. The first shock for all arrhythmias was successful in 75 of 88 patients (85.2%) for the monophasic waveform compared with 81 of 83 patients (97.6%) for the bipha- sic waveform, p = 0.0054. The first shock for VF was successful in 22 of 28 patients (78.6%) for the mono- phasic waveform compared with 25 of 25 (100%) for the biphasic waveform, p = 0.0241. The Gurvich bipha- sic waveforms deliverin rior to Edmark

a mean of 171 J were supe- monop asic waveforms delivering a ii

mean of 215 J for transthoracic cardioversion of ar- rhythmias of short duration. This finding may have im- portant implications for the development of future trans- thoracic defibrillators.

(Am J Cardiol 1995;75:1135-1139)

A 11 transthoracic defibrillators used in the United States deliver shocks that are nominally monopha-

sic. Damped sinusoidal waveforms delivered into a low impedance have a small terminal negative component, but they are effectively monophasic. Recent studies have shown that biphasic shocks with rectangular and trun- cated exponential waveforms, in which the polarity is reversed part way through the pulse, are more effective than monophasic shocks for extemal14 and intema15-lo defibrillation of animals and for internal defibrillation of humans.n,12 In the former Soviet Union, defibrillators have incorporated a quasi-sinusoidal biphasic waveform since its development there decades ago.13J4 Our study evaluates the efficacy and safety of a quasi-sinusoidal biphasic waveform in patients needing cardioversion or defibrillation for ventricular tachycardia (VT) or ven-

From the University of Washington, Seatile, Washington; University of Virginia, Charlottesville, Virginia; University of California, San Fran- cisco, California; University of Ottawa, Ottawa, Ontario, Canada; University of Colorado, Denver, Colorado; Vanderbilt University, Nashville, Tennessee; St. Joseph’s Hospital, Milwaukee, Wisconsin; Mt. Sinai Medical Center, Milwaukee, Wisconsin; and Physic&Ion- trol Cor oration, in part E

Redmond, Washington. This study was supported y a grant from the PhysioControl Corporation, Redmond,

Washington. Manuscript received November 22, 1994; revised manuscript received and accepted March 16, 1995.

Address for reprints: H. Leon Greene, MD, Division of Cardiol- ogy, Harbon/iew Medical Center, 325 9th Avenue (Z&3.5), Seat- tle, Washington, 98 104-2499.

tricular fibrillation (VF) induced during evaluation in the electrophysiology laboratory.

METHODS Patient selection: This evaluation was a multicenter,

prospective, randomized comparison of monophasic ver- sus biphasic waveform cardioversion/defibrillation in the electrophysiology laboratory. All patients were under- going electrophysiologic study because of a history of VT or VF, or both. All patients signed written, informed consent, and the study was approved by the institutional human subjects review committee for each of the 8 par- ticipating study sites (see Appendix). Enrollment was limited to adults (aged 218 years). Patients with implant- ed epicardial, pericardial, intracardiac, or subcutaneous defibrillator patch electrodes were excluded.

Arrhythmia induction: Patients were studied in the elec- trophysiology laboratory in the postabsorptive, nonse- dated, or lightly sedated condition. Up to 3 multipolar 5Fr or 6Fr electrodes were inserted by standard tech- nique, and ventricular arrhythmias were induced by pro- grammed right ventricular pacing in 22 sites, at 2 cycle lengths, and with up to 3 extrastimuli.15 Alternating cur- rent was not used to induce arrhythmias in this study.

Cardioversion/defibrillation: If, during the course of electrophysiologic testing, cardioversion/defibrillation be- came necessary, shocks were delivered by a LIFEPAK@ 7 monitor/defibrillator (Physio-Control Corporation, Red- mond, Washington), modified for research under an In-

ARRHYTHMIAS AND CONDUCTION DISTURBANCES/BIPHASIC WAVEFORM DEFIBRILLATION 1135

TABLE I Patient Characteristics

Waveform

Edmark Gurvich Monophasic Biphasic

Number of patients 88 83 Age (rr) 64 iz 11 63 zt 12

Men/women (no.) 78/10 76/7 Height (in) 69 * 4 70 f 3 Weight (lb) 181 * 31 179 f 32

Coronary artery disease (no.) 65 65 Other heart disease (no.) 23 18 Clinical arrhythmia type (no.)

Ventricular tachycardia 67 54

Ventricular fibrillation 14 13 Both 6 9

Other 7 Impedance (a) 82 +l21 80~ 19 induced arrhythmia type (no.)

Monomorphic VT 42 38 Polymorphic VT 15 15 Ventricular flutter 3 5

VF 28 25 Induced arrhythmia cycle length (ms) 226 * 53 228 f 54 Arrhythmia duration (s) 44 f 66 37 f 30

Values are expressed as mean * SD unless otherwise indicated. All differences, NS. VF = ventricular fibrillation; VT = ventricular tachycardio.

vestigational Device Exemption approved by the Food and Drug Administration. The delivered energy of the first shock was nominally 200 J, and was either the con- ventional Edmark monophasic waveform or a Gurvich biphasic quasi-sinusoidal waveform (Figure l), as ran- domized by the defibrillator. A code number printed on the strip chart by the defibrillator/monitor after energy delivery was later decoded for waveform identity, so that the type of waveform delivered was always blinded to the entire clinical team. Defibrillation pulses were deliv- ered through FAST-PATCH@ disposable defibrillation/ electrocardiographic electrodes positioned anterolateral- ly. With a patient impedance of 50 a, both circuits deliv-

TABLE II Number of Antiarrhythmic Drugs Taken at Time of First Episode, First Shock

Edmark Gurvich

Type Monophasic Biphasic

Number of patients 88 83 Quinidine 6 5 Procainamide 5 5 Disopyramide 2 0 Moricizine 0 1

Mexiletine 1 2 Propafenone 0 3 Amiodarone 6 8 Sotalol 4 2 Dofetilide 1 0 Verapamil 1 0 Digoxin 18 14

None 53 50 Total 97 90

Owing to the use of multiple ontiarrhythmic drugs, the total is higher than number of patients.

All comparisons, p = NS.

er 200 J; the amplitude of the second phase of the Gur- vich waveform has one half the amplitude of the tirst phase. The duration of the first phase is approximately 50% of the total waveform duration. Actual delivered energy was expected to vary with patient impedance. After any unsuccessful discharge, subsequent “rescue” _ cardioversion/defibrillation shocks were all of the Ed- mark waveform selectable at nominal energies of either 200 or 360 J.

Data collection: Twelve-lead electrocardiograms were recorded before induction, during the induced arrhythmia if clinical condition permitted, 30 seconds after defibril- lation discharge, and every minute thereafter until any shock-induced changes were restored to baseline. Elec- trocardiograms of the induced arrhythmias and cardio- version/defibrillation attempts were documented addi- tionally on printout from the strip-chart recorder and on a multichannel physiologic recorder with intracardiac tracings.

curletlt(Amps)

0 5 10 15

Time (ms)

Data analysis: The principal inves- tigator, a cardiologist, at each site read the electrocardiograms to diagnose pre- shock rhythm, measure time from onset of arrhythmia to shock, measure ar- rhythmia cycle length, and measure QRS and ST-T responses to the shock. Induced rhythms were classified as monomorphic VT, ventricular flutter (~200 ms cycle length), polymorphic VT, and VF based on the regularity, consistency of QRS morphology, and rate. Shock waveform identity was masked to all physicians. One cardiol- ogist (HLG) read all rhythms without knowledge of the classification made by the primary cardiologist. Any dis- crepancies in classification were re- solved by consensus review with a third cardiologist (PJK).

Each arrhythmia induction that re- quired a defibrillator shock was de- fined as an episode. Except where

Voltage (Volts)

FIGURE 1. Voltage and current of Edmark and Gurvich waveforms delivered into a 50 ti load.

1136 THE AMERICAN JOURNAL OF CARDIOLOGY@ VOL. 7.5 JUNE 1, 1995

specifically stated, analyses include only data from the lirst shock of the first episode from each patient.

Statistics: Efficacies of the 2 waveforms were com- pared using Fisher’s exact test and chi-square analysis. Patient variables, arrhythmia characteristics, and other covariates were analyzed using chi-square analysis, Fish- er’s exact test, and Student’s t test as appropriate. A p value co.05 was considered significant.

RESULTS Patient population: One hundred seventy-one of 176

natients initiallv enrolled (154 men and 17 women, age range 23 to 83 years [mean & SD 64 + 111) were entered into the study (Table I). One patient was excluded from analysis because his arrhythmia lasted >15 minutes before delivery of a shock, and intravenous procainamide was be- ing given during that time. One patient had spontaneous conversion from VT to sinus rhythm immediately before the shock was delivered. One patient who had a permanent pacemaker was enrolled erroneously. Two patients were excluded because technical problems with the defibrillator precluded analy- sis of the shock waveform type and en- ergy. A variety of cardiac diagnoses were present in patients with sustained ventricular arrhythmias. One hundred thirty patients had coronary artery dis- ease, and 19 patients had dilated car- diomyopathy. The remaining 22 pa- tients had a variety of diagnoses: con- genital heart disease, coronary spasm, hypertensive cardiomyopathy, hyper- trophic cardiomyopathy, idiopathic car- diomyopathy, right ventricular dyspla- sia, and pulmonary disease. Patient characteristics, diagnoses, arrhythmia characteristics (Table I), and concur- rent drugs (0 blockers, bronchodila- tors, calcium channel blockers, angio- tensin-converting enzyme inhibitors, and digoxin) in the 2 groups were similar. Many patients were taking an- t&rhythmic drugs at the time of elec- trophysiologic study (Table II). Some patients underwent >l ventricular tachyarrhythmia induction, usually on different days, when patients were tak- ing different antiarrhythmic drugs. The study data include 171 first episodes (Figure 2) and 61 repeat episodes for a total of 232 episodes (Figure 3).

% FAILURE /

S EDMARK: Monophasic Waveform 6128 q GURVICH: Giphasic Waveform

VF

FIGURE 2. First-shock failure rates for all patients, first episode only. The value was determined by Fisher’s exact test. V = ventricular; VF = ventricular ft nllation; -B* VT = ventricular tachycardia.

% FAILURE

Covariate analysis was performed to determine whether there were sig- nificant differences between the first epi- sode, first-shock groups receiving bipha- sic and monophasic shocks. There was no significant difference between the 2 groups in the length of time from onset

of the arrhythmia to shock (Table I). The preshock rhythm was VF in 53, monomorphic VT in 80, poly- morphic VT in 30, and ventricular flutter in 8 patients. For measurement of average cycle length, mean VF cycle lengths were included and averaged with VT cycle lengths. For measurement of duration of arrhythmias, 1 patient’s arrhythmia duration was not measurable.

Delivered energy: The mean delivered energy for the Edmark waveform (215 + 11 J) was greater than the ener- gy for the Gurvich waveform (171 f 11 J), p <O.OOOl.

Efficacy: The Gurvich biphasic waveform was equal or superior to the Edmark monophasic waveform in all

Polymorphic Monomorphic ..- V Flutter

VF Polymorphic Monomorphic V Flutter All

FIGURE 3. First-shock failure rates for all patients, all episodes. The p value was determined by Fisher’s exact test. Abbreviations as in Figure 2.

ARRHYTHMIAS AND CONDUCTION DISTURBANCES/BIPHASIC WAVEFORM DEFlBRltlATlON 1137

circumstances (Figures 2 and 3). Shock efficacy was unrelated to concomitant drug administration. Analysis of the tirst-shock efficacy results for lirst episodes of all induced arrhythmias revealed that 97.6% of arrhythmias were converted with the biphasic waveform compared with 85.2% with the monophasic waveform, p = 0.0054. Analysis of all episodes, first shock, likewise revealed that the biphasic waveform was superior to the mono- phasic waveform, p = 0.0048. This latter analysis includ- ed repeat inductions with the same antiarrhythmic drug or different antiarrhythmic drugs, so, that some patients are represented more than once in Figure 3. The bipha- sic waveform was also statistically superior to the monophasic waveform for VF (p = 0.0241, tist episodes; p = 0.0053, all episodes).

Complications: No complications occurred as a result of this study. Although some patients required a second shock for conversion of the arrhythmias, there were no adverse effects, and all induced arrhythmias underwent successful cardioversion or defibrillation: 156 on the first shock, 5 on the second, 3 on the third, and 7 on greater than the third shock. Conversely, it would appear that because the biphasic waveform defibrillation is more effi- cacious, it is likely that fewer patients required a second shock than would have been expected had only mono- phasic shocks been used in accordance with common clinical practice.

DISCUSSION This study demonstrates that transthoracic shocks,

using the biphasic waveform tested, are more success- ful in converting sustained ventricular arrhythmias than conventional monophasic shocks. These results confirm findings from the animal laboratory. Similar outcomes have been observed for epicardial shocks delivered both in the operating room in patients, and with implantable cardioverter-defibrillators. Even though the success rate of transthoracic cardioversion and defibrillation with monophasic waveforms is high, the biphasic waveform further improved success rates for transthoracic car- dioversion in our study.

Because the iirst shock of the Gurvich waveform suc- cessfully defibrillated all of the patients with VF, the observed difference in efficacy of the 2 waveforms could be viewed as a minimal estimate of the difference. It may therefore be possible to reduce the energy of the Gur- vich waveform without compromising its efficacy. Had the study used lower energies, it may have uncovered an even larger difference in efficacy of the 2 waveforms.

The defibrillators used in this study were calibrated to deliver 200 J into a 50 Q impedance. In our study, the average patient impedance was 81 + 20 fi, slightly high- er than seen in other studies, perhaps because adhesive electrodes were used rather than handheld paddles. Although there was no significant difference in patient impedance between the 2 groups, there was a significant difference in delivered energy. An impedance of 81 Q caused the Edmark-delivered energy to increase to 215 J and the Gurvich-delivered energy to decrease to 171 J. It also caused the amplitude of the second phase of the

Gurvich waveform to decrease from 50% to 45%. These differences were due to fundamental differences in the responses of the 2 electronic circuits to increases in impedance. If one assumes that a 215 J Edmark wave- form is more effective than a 171 J Edmark waveform, as data from Kerber et all6 suggest, then the differences between our 2 groups are more striking. The Edmark waveform delivered approximately 45 J (26%) more energy than the Gurvich waveform, yet it was less effec- tive. An even more striking difference in efficacy may have been observed if both defibrillation waveforms delivered 171 J shocks, although it is possible that there could be no difference in efficacy at other energies.

The duration of sustained ventricular arrhythmias in this study was quite short compared with patients with spontaneous arrhythmias who must be resuscitated and defibrillated in the field. External defibrillators need pulse waveforms that also work well on patients who experience a much longer duration of VF than the VF duration of electrophysiology lab patients studied. Fur- thermore, when VF occurs out-of-hospital, there may be a precipitating event (e.g., ischemia, infarction, or con- gestive heart failure) that could also adversely affect defibrillation success. In a study of prehospital resusci- tations, Weaver et ali7 estimated a delay of 10.6 f 5.5 minutes between patient collapse and delivery of the first defibrillator shock. In the present study, time from ar- rhythmia induction to tirst shock was 40 f 52 seconds. Yakatis et all8 demonstrated a reduction in defibrillation success rates as fibrillation time of dogs increased from 1 to 9 minutes. Our success rates probably differ from those observed by ‘Weaver et a1,17 primarily because of the difference in arrhythmia duration.

The type of arrhythmia, as well as duration, may also influence the success of cardioversion/defibrillation. In our study, most patients initially had VT, rather than VF, induced in the electrophysiology laboratory. It was often difficult to distinguish the time at which the rhythm de- generated from polymorphic VT to early VF, since both can be rapid and irregular; the only difference was main- tenance of some degree of organization of electrical activity in polymorphic VT External cardioverter-defib- rillators are most often used in the field for VF rather than VT Thus, our results demonstrating the superiori- ty of the biphasic waveform on VF are most relevant to the field use of these medical devices.

An improved survival rate for patients experiencing an out-of-hospital cardiac arrest is expected as external defib- rillators become less expensive, smaller, lighter, more portable, and therefore more widely available. However, a greater fundamental understanding of waveforms, effi- cacy, and mechanisms of defibrillation is needed. Bipha- sic waveforms, by defibrillating effectively with less deliv- ered energy, hold promise for reducing defibrillator cost, size, and weight, and use of biphasic waveforms may facilitate development of more effective defibrillators.

Acknowledgment: We thank Janice L. Jones, PhD, and Raymond E. Ideker, MD, PhD, for their helpful advice during this study.

1138 THE AMERICAN JOURNAL OF CARDIOLOGYa VOL. 75 JUNE 1, 1995

APPENDIX Lit of study sites (in order of number of patients enrolled): (1) University of Virginia Hospital: Principal Investigator-John P. DiMarco, MD, PhD; Nursing Coordinator-Russ Gallop, RN. (2) University of Washington Medical Center: Prin- cipal Investigator-Peter J. Kodenchuk, MD; Nursing Coordinator-Donna Nyen- huis, RN. (3) University of California/San Francisco: Principal Investigator- Melvin M. Scheinman, MD; Nursing Coordinator-Marilyn Wang, RN. (4) University of Ottawa Hem Institute: Principai Investigator-Anthony S.L. Tang, MD; Nursing Coordinator-Marilyn Lute, RN. (5) University of Colorado Health Science Center: Principal Investigator-Michael .I. Reiter, MD, PhD; Nursing Coor- dinator-Jane Williams, RN. (6) Vanderbilt University Medical Center: Principal Investigator-Debra S. Echt, MD; Nursing Coordinator-Diane Crawford, RN. (7) St. Joseph’s Hospital: Principal Investigator-Peter D. Chapman, MD; Nursing Coordinator-Jan Veseth-Rogers, RN. (8) Mt. Sinai Medical Cetiter: Principal Investigator-Mohammad Jazayeri, MD; Nursing Coordinator-Carol Gilbert, RN.

1. Schuder JC. Defibrillation of 100 kg calves with asymmetrical, bidirectional rec- tangular pulses. Cardiovasc Res 1984;18:419-426. 2. Schuder JC, Gold JH, Stoeckle H, McDaniel WC, Cheung KN. Trarlsthoracic defibrillation in the 100 kg calf with symmetrical one-cycle bidirectional rectan- gular wave stimuli. IEEE Trans Biomed Eng 1983;30:415422. 3. Schuder JC, McDaniel WC, Stoeckle H, Yerkovich D. Comparison of effec- tivedess of relay-switched, one-cycle quasisinusoidal waveform with critically damped sinusoid waveform in transthoracic defibrillation of lOOkilogram calves. Med Instrumentation 1988;22:281-285. 4. Schuder JC, McDaniel WC, Stoeckle H. Transthoracic defibrillation of 100 kg calves with bidirectional truncated exponential shocks. Tram Ain Sot Artif Intern Org 1984;30:520-525. 5. Chapman PD, Vetter JW, Souza JJ, Wetherbee JN, Troop PJ. Comparison of monophasic with single and dual capacitor biphasic waveforms for nonthoracoto- my canine internal defibrillation. J Am Coil Cardiol 1989; 14242-245. 6. Fain ES, Sweeney M, Franz MR. Improved internal defibrillation efficacy with a biphasic waveform. Am Heart J 1989;117:358-364.

7. Flaker GC, Schuder JC, McDaniel WC, Stoeckle H, Dbeis M. Superiority of biphasic shocks in the defibrillation of dogs by epicardial patches and catheter elec- trodes. Am Heart J 1989;118:288-291. 8. Feeser SA, Tang AS, Kavanagh KM, Rollins DL, Smith WM, Wolf PD, Idek- er RE. Strength-duration and probability of success cUrYes for defibrillation with biphasic waveforms. Circulation 1990;82:2128-2141. 9. Kavanagh KM, Tang AS, Rollins DL, Smith WM, Ideker RE. Comparison of the internal defibrillation thresholds for monophasic and double and single capac- itor biphasic waveforms. J Am Coil Car&l 1989;14:1343-1349. 10. Wyse DG, Kavanagh KM, Gillis Aq/l, Mitchell LB, Duff HJ, Sheldon RS, Kieser TM, Maitland A, Flanagan P, Rothschild J, Mehra R. Comparison of bipha- sic and monophasic shocks for defibrillation using a nonthoracotomy system. Am .l Car&l 1993;71:197-202. 11. Winkle AW, Mead RH, Ruder MA, Gaudiani V, Buch WS, Pless B, Sweeney M, Schmidt P. Improved low energy defibiillation efficacy in man with the use of a biphasic truncated exponential waveform. Am Heart J 1989;117:122-127. 12. Bardy GH, Ivey TD, Allen MD, Johnson G, Mehra R, Greene HL. A prospec- tive randomized evaluation of biphasic verses monophasic waveform pulses on defibrillation efficacy in humans. JAm Coil Cardiol 1989;14:72&733. 13. Gurvich NL, Yuniev GS. Restoration of heart rhythm during fibrillation by a condenser discharge. Am Rev Soviet Med 1947;4:252-256. 14. Negovsky VA, Smerdov AA, Tabak VY, Venin IV, Bogushevich MS. Criteria of efficiency and safety of the defibrillating impulse. Resuscitation 1980;8:53-67. 15. Waldo A, Akhtar M, Brugada P, H&horn R, Scheinman MM, Ward D, Wellens HJJ. NASPE Policy Statement: the minimally appropriate electrophysiologic study for the initial assessment of patients with documented sustained monomorphic ven- tricular tachycardia. PACE 1985;8:918-922. 16. Kerber RE, Kienzle MG, Olshansky B, Waldo AL, Wilber D, Carlson MD, Aschoff AM, Birger S, Fugatt L, Walsh S, Rockwell M, Charbonnier F. Ventricu- lar tachycardia rate and morphology determine energy and current reqhirements for transthoracic cardioversion. Circulation 1992;85:158-163. 17. Wekver WD, Cobb LA, Copass MK, Hallstrom AP. Ventri~ulular defibrillation- a comparative trial using 175-J and 320-J shock. NEngl JMed 1982;307:1101-1106. 18. Yakatis RW, Ewy GA, Otto CW, Taren DL, Mooti TE. Influence of time and therapy on ventricular defibrillation in dogs. Crit Care Med 1980;8:157-163.

ARRHYTHMIAS AND CONDUCTION DISTURBANCES/BIPHASIC WAVEFORM DEFIBRILLATION 1139