YACC plot. 24, No. 1 871

(I ‘i-15). The development of an implantable atria1 defibrilla- tor should therefore be considered in patients with sy tomatic recurrent atrial fib~l~ation.

Although patients may develop sig~i~ca~t symptoms at the onset of each episode of paroxysmal atrial fibrillation, they do remain fully conscious. Acceptability of atrial

t would therelare be critically dependent al waveform. In ~~~~ti~~,

t waveform for the atrium

ogy. Physics, Cardiac Surgery and Pub , London, England, United Kingdom.

1993; revised manuscript received Feb

: Dr. David Keane, Department of Interventional Cardiology, Thoraxcenter, Erasmus University, Room Ee 2332, P.O. Box 1738, Rotterdam 3000, The Netherlands.

61394 by the American Coliegc oP Cardiology

would ~aci~~~a~e a reduction in t

rovement of elective tr

brillators, extensive

than a monophasic waveform, at prcsenl the most widely adopted waveform for defib~l~atio~ of the ventricle in ex inental and clinical use appears to be biphasic (2%=24). report our study corn a bip waveform in human ep ial aWi

m The two waveforms were compared in 21 pa- tients undergoing routine coronary artery bypass grafting (14 men, 7 women; mean [rSD] age 62.3 1 7.7 years). The

073%1097/94/$7.00

172 KEANE ET AL. BIPHASIC WAVEFORM IN ATRIAL DEFIBRILLATION

JACC Vol. 24. No. 1 July 1894:141-6

study WAS approved by the hospital ethics committee, and all patients gave informed written consent.

Patients were excluded if 1) they had a left ventricular ejection fraction <40% as assessed by contrast ventriculog- raphy at the time of cardiac catheterization; 2) they were undergoing repeat cardiac surgery; 3) they were undergoing concomitant valve replacement; or 4) they had chronic atrial fibrillation (no patient entered in the study had a history of either paroxysmal or chronic atrial fibrillation). Patients with low ejection fraction were excluded to maintain a more uniform study group and to limit any potential risk of a n ropic effect on the ventricles from low voltage a

venous pressure and tempera ously on a cardiac monitor (He from the cardiac monitor was c

CGs), a~e~al pressure,

for R wave synchronization of atria1 shocks. An WI3 output from the cardiac monitor was also recorded on rna~~et~c tape

tronic model 2394 external cardioverterdefibrihator to deliver both the monophasic and the biphasic

shocks as follows: I) a monophasic truncated exponential, 8-ms waveform; 2) a biphasic truncated exponential, g-ms, dual-capacitor bidirectional waveform with equal first- and second-phase duration and leading-edge voltage. Energy for the external cardioverter-defibrillator was stored in a series of four external capacitors providing a total capacitance of 130 fiF.

The delivered energy of each shock was controlled by the leading-edge voltage. For monophasic sh the iwedge voltage settings were 70, 100, 150. and

250 V, and for biphasic ks the five leading-edge voltage settings were 50, 70, V. One hundred shocks were randomiz 0 shocks at each vo f both waveforms.

cardioverter-defibrillator was connected to atria! defibrillation paddles via a junction box, which pro- vided a measurement of the vol and current delivered to the patient. The measurements vided by the junction box were in the form of two low voltage outputs for recording on magnetic tape. The junction box contained a small series resistance of 0.5 ohms through which the patient current flowed and a high impedance attenuator to reduce the defibrillation voltage to <IO V. Output voltages from the junction box proportional to the patient’s applied current and voltage were connected to two univerd isolation am- plifiers to maintain patient safety. The frequency response of the amplifiers was direct current (DC) to IO kHz, and their Outputs, together with the surface ECG, were recorded at two sensitivities on half-inch magnetic tape at a speed of

fidelity, frequency-modulated, RACAL Store 14 recorder with a frequency response of DC to 20 kHz.

Because no specific atrial defibrillating paddies or con- toured epicardial patches are commercially available, we designed atrial paddles from modified pediatric and infant

contoured ventricular tached an 1 l-cm2 cone

with the anterior

atrial free walls withou

c monitor was read . Success or failure

y the investigator and

of a successful shock (i.e., co sion to sinus rhythm),

mum of 30 s of established atrial ~bri~~at~o~. A~~ys~. After each study the signals stored on magneti:.

tape (ECC, delivered voltage and delivered current) were replayed at a reduced speed of 1% in/s into a Mingograph ink jet recorder (Siemens 34T). This gave a paper recording 64 times slower than real time and allowed a full bandwidth representation of the data.

The data-recording and playback system was calibrated using recordings with dummy resistive loads. A range of shocks was delivered across a nd recorded on magnetic tape. Printout on the corder allowed calibration of the recorded voltage and current si

The leading and trailing edges of the current and voltage of each phase of all shocks were measured and entered on a data base. The delivered energy of each atrial shock was calculated using the formula E = O.SC$VLE’ - VTE2), where E = delivered energy, C = capacitance, VLE = leading-edge voltage, VTE = trailing-edge voltage. The probability of success of the moncphasic and biphasic wave- forms was modeled as a function of the delivered energy and

0.39 (0.36 to 0.38) NS

v 50 153.8 (46 to 261) 71*9 (6i.9 to 81.9) NS V 80 327.9 (I 19 to 537) 86.1 (71.1 to 97.1) < 0.05

delivered energy and voltage expressed as mean vahres (and 95% coniidence intervals of the mean).

The data points and etwee~~ delivered energ and percentage of succ I defib~l~atio~ for eat waveform are shown in Figure 1. The biphasic shocks were associated with a steep dose-response relation and narrow 95% confidence intervals of the mean, in contrast to the monophasic shocks, which were ed with a shallow dose-response curve with wide 9 dence intervals of the mean (data scatter). The vdues ivered enesgy and

“ated with 50% and 8 basic waveforms a

was found to he si~~ificaat~y more ic wavefo~ at tbe m ess (80%). The data on

delivered voltages and outcomes are given in Table 2. Mean and standard error of impedance at the leading edge

for all 100 shocks was 52.19 -+ 1.24 ohms. Impedance at tbe

varie cy of bi tih, the

wave- on alad

Monopbasic Biphaie

Voltage Success success

(VI Outcome (%I Outcome (%I

44-55 2,‘lO 29%

3110 30% 3110 3@%

5110 50% 10110 1 511 I 4.5% la/lo 1

168-192 5110 50% 10110 100%

219-239 919 78%

Results of 100 randomized atrial shocks. Voltage has been grouped in five bands for eack waveform (ranging from 64 to 239 V for tbe mosophasic waveform and from 44 to 192 V fur the biphasic waveform), reflecting the measured leading-edge voltage of all shocks. Outcome is expressed for each waveform as the number of successful shocks over the total number of shocks within each voltage band. Success is derived from the corresponding outcome for each voltage band.

174 KEANE ET AL. JACC Vol. 24. No. I

BIPHASIC WAVEFORM IN ATRIAL DEFIBRILLATION July 1994:171-d

the amplitude of both the first and second phases (26). Indeed, at critical values of these three variables certain biphsic waveforms can require more energy to defibril!ate thar monophasic pulses of equal duration (27). With three such interdependent variables the optimal characteristics for the ideal biphasic waveform become diticult to determine. However, in general it appears that biphasic waveforms are most efficient when the duration of the first phase is in the region of 4 ms, and the duration of the second phase is less than or equal to that of the first phase (28).

Two recent studies deployed a biphasic wavef3rm in atria! &ibrillation in a sheep model. In the

to be etfective and deliv associated with 50% and

5 and 2.5 J were

r et al. (31) presented results of a study on rms and catheter electrod

pace-induced atrial fibrillation in sheep w Biphasic waveforms of 3 + 3 and 6 + 6 ms in duration we compared with monophasic waveforms of 6 and 12 ms duration. Biphasic waveforms of 3 + 3 ms in duration were found to be most effective and were associated with a success rate at a delivered energy of 1.3 J, whereas the effective of the two monophasic waveforms (6 ms) had a success rate at 2.2 J, Our study is the first to assess a biphasic waveform in atrial defib~liation in humans and is consistent with the results of the studies in the sheep model

“excitation pulse” (31.32); 2) the first phase may shorten Ceh refractoriness, enabling more cells to be excited second phase (26); 3) the biphasic waveform has a

lower “upper limit of vulnerability”; i.e., the myocardium is less able to reinitiate fibrillation when partially replarized (33,34); 4) the first phase opens sodium channels on one half

te (the half nearer the cathode), and the ns channels on the other half without

ose on the first half (thus the energy of the second phase becomes critical) (27).

Of these four proposed mechanisms, that of a “condition- ing” effect of the first phase appears to be gaining the most suPport; however, whether the mechanism of improved efficiency in the atria is identical to that of the ventricles remains to be determined.

~ro~bilistic function of that the slope and data larger R~mbcr of shocks intercept of the

increase in energy reqMirem~~

pulmonary bypass (28°C). Of note, we had found in a pilot study that atrial fibrillation was more easily maintained after 20 min of hypothermic cardiopulmonary bypass and a period of aortic cross-clamping compared to the period before or immediately after commencing cardiopulmonary bypass. In the abzen- of previous studies on atriai defib~I~atio~ in this model it is difficult to determine what effect the combination of cardiopu~mon~ bypass, hypothermia, aortic cross- clamping and a right atriotomy may have had on the ene

In previously pMblisbed brillation, a “rescue

shock” has often been used to maintain period of ventricular fibrillation before each study ta avoid prolongee hemodynamic embarrassment. In our study a “rescue shock” was not administered to the atria in the case of an unsuccessful study shock for the following reasons: 1) ventricular studies have shown that delaying a defibrillat-

t receive an im

atria! defib~llat~on by surface area of

e in a single-chamber implant- evice, kctwever, would have se~s~~g-~ac~~g kad, which

ave signal of constant amplitude

ventricular defibrillation (i.e., a dual-chamber defibrillator). The incorporation of an efficient biphasic waveform in an

implantable atria1 defibrillator device would facilitate a re- duction in battery size and an increase in Mery longevity.

We ~~ate~ul~y acknowledge the statistical advice of Eric Boersma, Department of Epidemiology. Thoraxcenter. Erasmus University Rotterdam, The Netherlands.

I. Feld 6. Atrial fibrillation: is there a safe and highly effective pharmaco- logical treatment? Circulation 1990;82:2248-50.

2. Lewis R. Atrial fibrillation: the therapeutic options. Drugs 1996;40:841- 53.

3. Bauemfeind R. Welch W. New hope in atrial fibrillation. J Am Coil

Van Gelder 1, Van Gist W, Hillege H. Gosselink A, Lie K. Serial adarrhythmic drug treatment to maintain sinus rhythm after electrical cardioversion for chronic atrial fibrillation or atria! flutter. Am J Cardiol 1991;68:335-41.

5. Coplen S, Antmann E, Berlin J, Hewitt P, Chalmers T. Efficacy and safety of quinidine therapy for maintenance of sinus rhythm after cardio- version: A metaanalysis of randomised control trials. Circulation 1990; 82:1106-16.

6. Antman E, Beamer A, Cantillon C, McGowan N. Goldman L, Friedman P. Long term oral propafenone therapy for suppression of refraclory symptomatic atrial tibrillation and atrial flutter. J Am Coil Cardiol 1988,12:1OOo.c-11.

7. Karlsson B, ‘fortensson 1. Abjom C, Jansson S, Peterson L. Disopyra- mide in the maintenance of sinus rhythm after electroversioa of atrial fibrillation. Em Heart J 1988;9:284-90.

8. Gold R. Haffajee C, Charos K, Sloan K, Baker S, Alpert J. Amiodarone for refractory atrial fibrillation. Am J Cardiol 1986;57:124-7.

9. Feld 6, Chen P, Nicod P, Meyer D, Fleck R. Possible atrial proarrhyth- mic effects of encainidt and Recainide. Am J Cardiol 1990:66:378-83.

IO. Holt P, Boyd E. Long term outcome of His bundle ablation: benefits and drawbacks [abstr). J Am Coil Cardiol 1991;17:367A.

I I. Lehch J, Klein G, Yee R, Guiraudon G. Sinus node-atrioventricular node

176 KEANE ET AL. BIPHASIC WAVEFORM IN

JACC Vol. 24, No. I July I :171-b

12.

13.

14.

IS.

16.

17.

18.

19.

20.

21.

22.

jJolation: long-term results with the conidor operation for atrial fibrilla- tion, J Am Coil Cardio11991;17:970-5. DiMarco I. Sutgical therapy for atrial fibrillation: a first step on what may be a long road. J Am Coil Cardiol1991;17974-7. Gwoy S, De hIIyI% B, Wellens F, Cuiraudon G. Bruguda P. InIeratrial dissociation following the conidor operation: role of attial contraction in thrombogenisis [abstri. PACE 1992;15: Part II:532. Cox J, Boineau J, Schuessler R, et al. Successful surgicaI treatment of atrial fibriI&ion. JAMA 1991;266:19%-gfl. Lin F, Lo H, Cheng J, Jong Y, Tseng C. Tseng Y. A new surgery based on Me-AIlessii’s hypothesis for cbmnic atrial IibriIBtion in mitral valve disease [abstract]. PACE 1992;15: Part II:532. Jones D, RIem G, Guiraudon G, Sluuma A. Sequential pulse defibrillation in bumans: sequential Pulse defibrillation with epicardii ~Ieetrodes. J Am CoU CardioI I Jones D, KIein G, Guiruadon G, et al. Internal cardiac deIlbrillation in

riced improvement wilh sequential pulse delivery to two orientations. Circulation I

Jones D, #kin G, KaIlok hf. Improved interneI defibrillation with twin nergy delivery to different kad orientations in pi

M, lnoue H, Kallok M, Zipes D reduce ventricular defibrillation in and without myocar-

dial infarction. J Am Coil Catdiil I!&%& Bardy G, Ivey T, Allen M. JohnsonG, Mehra R, Greene II. A prospective

form, Am Heart J Iwis;lI1:I22-7. FIaker G, Schuder 1, McDaniel W, Stoeckle K. Dbeis hi. Superiority of biphasi~ shocks in the defibrillation of dogs by epicardial patches and catheter electrodes. Am Heatt J 1989,118:28R-91.

23. Fain E, Sweeney M. Franz M. Improved internal defibrillation elcacy with a biphasic waveform. Am Heart J 1389,117:358-64.

24. Chapman P. Vetter J. Souxa J, Tmup P, Wetberbee J. HoBTman R. Comparative e&acy of monophasic and biphasic truncated exponential shocks for nonthoracotomy internal defib~llation in dogs. I Am Coil CardioI 198%i2:739-45.

25. Davy JM, Fain E, Dorlan P. Winkle R. The relationship between suc- cessM defibrillation and delivered energy in opeucbest dogs: ~a~~~ of the “defibrillation threshold’* concent. Am Heurt J 1987:I 13:77-83.

26. Tang A, Yabe S, Wharton M. Doiker hi, Smith W. ldeker R. Ventricular defibrillation using biphasic waveforms: The imoortance of nhasic dura- tion. J Am Coil CTardkl 1989;13:207-14. . .

thduralion and probability

of success curves for defib~uatio~ with bipbasic waveforms. Circulation l99&82:2128-41.

26. Hagler J, Alfemess C, ~~on-TuM~b E, W~le~us T, Sm ldeker R. Defibrillation efficacy of long and short duration biphasic o~o~basic waveforms [abstr]. PACE 199I;I4:II-7

29. Powell A. Garan H, McGovern B, Fal n S, Ruskin J. Low energy conversion of atriaI fibrillation J Am Coil Cardiol I992;20:707-II.

39. Jones J. Jones R. Baliasky G. I symmetrical biphasic defibrillator Wl418-24.

31. Cooper RA. Alfemess CA, Comparison of multiple w cardioversion of attial fibrillation in sheep ~abst~~t~. PACE I99%l5 Part

obinson K, Giles W,

33. Wharton &I, Rickard V, Murry C, et al. ~le~tro~b~s~olo~~~a~ eIfecls of monophasic and biphasic stimuli in normal and infarcted dogs. PACE I ;I3:1158-72.

34. Swaru J, Jones J, Jones . Fletcher R. ~ondit~~~in~ prepulse of biphasic defibrillator waveforms enhances refractoriness to fib~llation w~v~fronts. Circ Res I~i;~43~-4~.

35. Lim C. Doursounian M. Valeri C. ~~d~o~~rno~ary ~fib~l~atiao energy requirements (abstract]. PACE

36. IX., Kleim G. Effects of time to d subthnshold preshocks OR defibrillation success in pigs. 358-65.

37. Keane D, Sulke N, Cooke Jackson 6. Sowton E. End cardioversion of atrial flutter d ~b~I~atio~ [abst~ct~. Part IL928.

38. Keane D. A review of experimental and clinical studies of atrial de:Rbril- letion: imp1 design of an implantable atrial defibrillator. Eur J CPE

39. #cane D. E rations for atrial defibrillation. la: Camm AJ, Liudemans F, editors. New Waves in A~hytbmi~ Therapy. Mouat Kisco (NYl. In press.

40. Levy S, Lacombe P, Cointe R, Bru P. sion of chronic atrial fibrillation. J Am

41. Kumagl K, Yamanouchi Y. Hiroki T, Arak cardiovetsion on chronic lone atrial fibrillation. PACE 1991;14:1571-5.

42. Levy S, Laurlbe P. Ila E. el al. A randomised comparison of external and internal cardioversion of chronic atrial fibrillation. Circulation 1992; &1415-20.