Reduced Cardioversion Thresholds for Atrial Fibrillation and Flutter Using the Rectilinear Biphasic Waveform

  • Published on

  • View

  • Download


  • Journal of Interventional Cardiac Electrophysiology 13, 145150, 2005C 2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands.

    Reduced Cardioversion Thresholds for Atrial Fibrillationand Flutter Using the Rectilinear Biphasic Waveform

    Mark J. Niebauer1, Mina K. Chung2, James E.Brewer3, andPatrick J. Tchou21Department of Internal Medicine, Division of Cardiology,University of Nebraska Medical Center, Omaha, NE68198-2265, USA; 2The Section of Cardiac Electrophysiologyand Pacing, the Cleveland Clinic Foundation, Cleveland, OH44195, USA; 3Zoll Medical Corporation, Burlington, MA 01803,USA

    Abstract. Background: The RLB waveform has beenshown to be superior in overall efficacy to the MDSwaveform for cardioversion of AF in one prospectivestudy and one large retrospective analysis. However, lit-tle is known about the efficacy of the RLB waveform atlower energies.

    Objective: This study was undertaken to define thecardioversion thresholds for atrial fibrillation (AF) andflutter (FL) using the rectilinear biphasic (RLB) wave-form and compare these to the cardioversion thresholdusing the conventional monophasic damped sine (MDS)waveform.

    Methods: All patients underwent transthoracic car-dioversion of persistent AF and FL. We performed step-up cardioversion thresholds for AF in 180 RLB patientsand 38 MDS patients and compared those results. Wealso performed cardioversion threshold determinationsin 39 RLB patients with typical right atrial FL. For theRLB patients, an initial energy setting of 5 Joules (J)was selected, with increasing energy steps until success,up to 200J. The MDS energy sequence was 50 up to 360J.

    Results: The average selected energy threshold forAF using the RLB waveform was 70.6 J (median=50 J)versus 193.4 J (median=150 J) for the MDS waveform(p< 0.001). For FL, the average cardioversion thresholdusing the RLB waveform was 33.2 J (median=20 J; p50% efficacy at 70 Joules (J), the low-est energy tested. We acquired this defibrillatorshortly after its release and were interested in de-termining the lowest effective energy of the RLBwaveform in both AF and FL. We had formerlyperformed step-up cardioversions in a group of pa-tients using the MDS waveform in order to deter-mine the cardioversion threshold for AF. We there-fore decided to perform step-up cardioversion test-ing in both AF and FL patients using the RLBwaveform and compare these results to our previ-ous experience with the MDS waveform.


    Patient PopulationThe study is a non-randomized trial to comparethe efficacy of the RLB and MDS waveforms fortransthoracic electrical cardioversion of AF andFL in a general clinical setting. The study was re-viewed by the Cleveland Clinic Institutional Re-view Board and approved.

    From October 1999, to 2001, we performed188 transthoracic cardioversion procedures in173 patients for persistent AF using the RLBwaveform. These patients underwent a sequen-tial step-up shock energy protocol to determine

    Address for correspondence: Mark J. Niebauer, PhD, MD, De-partment of Internal Medicine, Section of Cardiology, 982265Nebraska Medical Center, Omaha, NE 68198-2265, USA.E-mail:

    Received 10 January 2005; accepted 11 May 2005


  • 146 Niebauer et al.

    the cardioversion threshold. We compared theAF threshold data to 38 previously cardiovertedpatients in which we had performed step-upcardioversion thresholds for AF using the MDSwaveform. Additionally, we performed transtho-racic cardioversion in 39 patients with FL usingthe RLB waveform defibrillator. Unfortunately, wehad not performed cardioversion thresholds forFL with the MDS waveform. Hence, we comparedthree clinical groups; RLB AF, MDS AF and RLBFL.

    Patient PreparationThe standard guidelines for anticoagulation of allpatients recommended by the American Collegeof Physicians were adhered to in our laboratory.Specifically, all patients with AF over 48 hoursduration were anticoagulated for a minimum of30 days prior to the cardioversion procedure orhad undergone transesophageal echocardiogra-phy where no left atrial thrombus was visual-ized in the presence of therapeutic anticoagulationwith warfarin or intravenous heparin. Patientswere sedated using either intravenous sodiummethohexitol (44.0 12.4 mg). In all cases theexternal patches were positioned in an anterior-posterior chest orientation [5].

    DefinitionsShock success is defined as conversion to a si-nus, AV junctional or paced atrial rhythm im-mediately after the shock in our laboratorydatabase. Transiently successful procedures aretallied differently as those with initially successfulcardioversions with spontaneous recurrence of fib-rillation after at least one organized atrial depolar-ization prior leaving our laboratory (typically 1530 minutes). However, the target arrhythmia re-curred in only a small percentage of study patients(8 patients or 4%) in the RLB AF group and nonein the RLB FL group or the MDS AF group. There-fore, we excluded those RLB patients in which AFrecurred, in order to eliminate any controversyover combining threshold results from patientsmaintaining sinus rhythm and those experienc-ing early recurrences. Our cardioversion resultstherefore only pertain to those without very earlyrecurrence. The final RLB AF comparison groupconsisted of 180 patients.

    Shock WaveformsThe RLB waveform was delivered from an externalRLB defibrillator (Zoll M-Series Biphasic Defibril-lator, Burlington, MA). The waveform was gener-ated by a 115 F capacitive discharge, comprisingan essentially constant-current 6 ms first phaseand a truncated, exponential 4 ms second phase(Fig. 1). The amplitude of the phases varied withthe selected energy, while both phase and total

    Fig. 1. Representation of a 120 J rectilinear biphasic (RLB)waveform shock and a 200 J monophasic damped sine (MDS)waveform shock both delivered across 75 ohms, which iscomparable to a patients transthoracic impedance. Shockamplitude is expressed as current (Amperes) and the durationof the shocks is measured in milliseconds.

    pulse width remained fixed. The MDS waveformshocks were delivered from a Hewlett-Packarddefibrillator (Model M1722A, Hewlett-Packard,Andover, MA). Both defibrillators were capableof reporting selected energies (from defibrillatorcontrol) and delivered energies as well as peak cur-rent and patient impedance for each shock.

    Shock ElectrodesAdhesive pads for external cardioversion (ZOLLCardiology Specialty Pad) were used to deliver theRLB shocks. The anterior electrode was circularand had an active surface area of 78 cm2. Theposterior electrode was rectangular and had anactive surface area of 113 cm2. Hewlett PackardMDS shocks were likewise delivered throughself-adhesive electrodes (Model M1749A, Hewlett-Packard, Andover, MA). As noted above, thepatches were initially positioned in a right ante-rior, left posterior chest wall orientation. If failureoccurred at maximum energy of either defibrilla-tor, another maximum energy shock was deliveredafter pressure was applied over the anterior patchand/or repositioning of that patch.

    Step-Up Cardioversion ProtocolThe step-up energy protocol for determining atrialcardioversion threshold consisted of shocks of in-creasing energy that were sequentially delivered,until success or maximal energy was reached.When using the RLB waveform defibrillator, wedelivered the following sequence of selected en-ergies: 5, 10, 20, 50, 75, 100, 150, and 200 J.The previously determined step-up cardioversionthresholds in 38 patients with AF using the MDSwaveform utilized energy steps that started with

  • Biphasic Cardioversion Thresholds 147

    Table 1. Clinical characteristics of all atrial fibrillation and flutter group patients

    Atrial flutter RLB waveform Atrial fibrillation RLB waveform Atrial fibrillation MDS waveform

    Sex (M/F) 31/8 140/40 30/8Age (y) 63.8 11.4 68.0 10.3 66.8 11.6Ejection fraction 45.8 17.4 46.1 14.9 50.2 10.5LA diameter (mm) 49.8 8.2 47.8 8.8 47.8 11.1Arrhythmia duration (D) 8.0 8.7 271.0 813.9 273.8 518.8Body weight (kg) 91.9 23.3 89.5 24.8 88.3 16.9No structural disease 4 (10.3%) 34 (18.9%) 12 (31.6%)*Coronary artery disease 11 (28.2) 64 (35.6) 14 (36.8)Valvular heart disease 16 (41.0) 48 (26.7) 8 (21.1)Hypertensive 3 (7.7) 34 (18.9) 5 (13.2)Dilated cardiomyopathy 2 (5.1) 19 (10.6) 0 (0)Hypertrophic 1 (2.6) 1 (

  • 148 Niebauer et al.

    Table 3. Cardioversion thresholds for the atrial fibrillation and atrial flutter groups. Data are presented as group means andstandard deviations

    Atrial flutter RLB waveform Atrial fibrillation RLB waveform Atrial fibrillation MDS waveform

    Selected energy (J) 41.7 29.5 70.6 49.3 193.4 93.2Delivered energy (J) 47.0 34.3 81.6 56.2 201.4 98.3Peak current (Amps) 7.3 3.4 9.7 3.8 33.4 9.0Peak voltage (Volts) 586 259 815 342 2288 641Impedance (Ohms) 83.8 22.9 86.7 21.4 70.1 15.7

    p < 0.0001 atrial flutter vs. atrial fibrillation with the RLB waveform. p < 0.0001 atrial fibrillation with the MDS waveform vs. atrial fibrillation with the RLB waveform.

    group who failed cardioversion with maximum(360J) energy. All patients treated with the RLBwaveform were successfully cardioverted. Themean AF cardioversion selected energy thresholdusing the RLB waveform was 70.6 49.3 J witha median of 50 J, whereas the average cardiover-sion threshold using the MDS waveform was sig-nificantly higher (p < 0.0001) at 193.4 93.2 Jwith a median of 150 J. The distribution of thresh-old energies within each of the three groups areshown in Figure 2, with the median thresholdnoted for each group. Delivered energy, peak cur-rent and peak voltage (calculated from the peakcurrent and impedance using Ohms law) thresh-olds were also significantly different between thetwo groups as shown in Table 3. The averageshock impedance was greater for the RLB AFpatients (86.7 ohms) than for the MDS AF pa-tients (70.1 ohms; p < 0.0001). The average se-lected energy cardioversion threshold for the FLpatients using the RLB waveform was 33.2 28.8 J with a median of 20 J. This was significantlylower (p < 0.01) than the threshold for AF, us-

    Fig. 2. Distribution of cardioversion selected energy thresholds in each of the three study groups, displayed as percentages of thatgroups population. In addition, the selected energy corresponding to the median cardioversion thresholds for each group are shownin asterisks.

    ing the RLB defibrillator. The corresponding deliv-ered energy, peak current and voltages at thresh-old were also significantly lower than that for AF(Table 3).

    Adverse EventsThere were no adverse events recorded duringall cardioversion procedures and no documentedstrokes or transient ischemic cerebral events re-ported within 30 days after cardioversion. Skinburns were not systematically followed in ourdatabase, however, there were no reported clini-cally significant burns on follow-up (generally 35 weeks after cardioversion).


    To our knowledge, this is the only study that hasevaluated the atrial cardioversion threshold of theRLB defibrillator for both AF and FL. It is also thefirst study to compare the cardioversion thresh-old of the RLB waveform for AF to that of theMDS waveform. Our threshold results support the

  • Biphasic Cardioversion Thresholds 149

    prior randomized study showing that the rectilin-ear biphasic waveform is superior to the monopha-sic damped sinusoidal waveform for external car-dioversion [6]. These data are also consistent withthe improved efficacy seen with the RLB waveformfor ventricular defibrillation [7]. We have also con-firmed the superior overall efficacy of the RLBwaveform to the MDS waveform for cardioversionof atrial fibrillation in over 3000 patients [8]. Thestep-up cardioversion threshold for RLB shocksestablished in this study is less than half that ofthe MDS waveform as measured by selected en-ergy, delivered energy as well as peak current.

    Surprisingly, few studies have attempted to de-termine low energy cardioversion efficacy, usingthe MDS waveform. Ricard, et al. demonstrateda 96% cardioversion efficacy for a monophasicdamped sine (MDS) defibrillator with a step-upprotocol using steps of 40/50 J, 80/100 J, 160/200 J,and 360 J [9]. In a second MDS cardioversionstudy, a randomized protocol of 100 J, 200 J, and360 J was followed, in which patients were ran-domly assigned to start at 100 J, 200 J, or 360 J,for up to five shocks in sequence [10]. In that study,21 patients were assigned to start at 100 J, 23 pa-tients to start at 200 J, and 20 patients to start at360 J, with a cardioversion efficacy of 90, 90, and100%, respectively. The cardioversion rate for theentire study population was 94%.

    Previous studies have shown variable effectsof antiarrhythmic drugs on atrial cardioversion.Sotalol has been reported to reduce cardioversionthresholds when the MDS waveform is used tocardiovert using transvenous electrodes [11].However, in our group of patients, sotalol hadno significant effect on the external biphasiccardioversion threshold.

    LimitationsThe primary limitation of our study is that thewaveform selections were not prospectively ran-domized. Rather, we compared the RLB thresholdsfor the AF cardioversions to those of previouslycardioverted patients with AF who had under-gone step up cardioversion using the MDS wave-form. Because of the differing ranges of efficacy, aswell as the different energy steps available in therespective defibrillators, the RLB energy shockswere begun at a lower level than the MDS shocks.Finally, the lack of MDS threshold data for the FLpatients is unfortunate, but there is a high efficacyof DC cardioversion for this arrhythmia, regard-less of waveform used [8].


    Our results show that the threshold for externaldirect current cardioversion of AF is markedly

    lower for the rectilinear biphasic waveform defib-rillator compared to the conventional monopha-sic damped sine waveform. Our results are con-sistent with the higher success rate for AF forthe RLB waveform seen in the prospective, ran-domized trial compared to the MDS waveform [6],as well as our own clinical experience in overallRLB efficacy [8]. In addition, we report the aver-age threshold for the external cardioversion of FL,which is significantly lower than that for AF, asmight be expected. These results provide the firstmeasurement of cardioversion threshold for AFand FL using the rectilinear biphasic waveformas well as further proof of the higher effectivenessof transthoracic cardioversion using the rectilin-ear waveform over the conventional damped sinewaveform.


    The authors would like to express gratitude to Ms. SawnsuraeMarion-Phillips for her expert assistance in manuscript prepa-ration.


    1. Lown B, Amarasingham R, Newman J. New method forterminating cardiac arrhythmias: Use of synchronized ca-pacitor discharge. JAMA 1962;182:548555.

    2. Van Gelder IC, Crijns HJ, Van Gilst WH, Verwer R, Lie KI.Prediction of uneventful cardioversion and maintenanceof sinus rhythm from direct current electrical cardiover-sion of chronic atrial fibrillation and flutter. Am J Cardiol1991;68:4146.

    3. Hagemeijer F, Van Houwe E. Titrated energy cardiover-sion of patients on digitalis. British Heart J 1975;37:13031307.

    4. Van Gelder IC, Crijns HJ, Tieleman RG, Brugemann J,De Kam PJ, Gosselink AT, Verheugt FW, Lie KI. ChronicAF: Success of serial cardioversion therapy and safetyof oral anticoagulation. Arch Intern Med 1996;156:25852592.

    5. Botto GL, Politi A, Bonini W, Broffoni T, Bonatti R. Exter-nal cardioversion of atrial fibrillation: Role of paddle posi-tion on technical efficacy and energy requirements. Heart1999;82:726730.

    6. Mittal S, Ayati S, Stein KM, Schwartzman D, CavlovichD, Tchou PJ, Markowitz SM, Slotwiner DJ, Scheiner MA,Lerman BB. Transthoracic cardioversion of atrial fibril-lation: Comparison of rectilinear biphasic versus dampedsine wave monophasic shocks. Circulation 2000;101:12821287.

    7. Mittal S, Ayati S, Stein KM, Knight BP, Morady F,Schwartzman D, Cavlovich D, Platia EV, Calkins H, TchouPJ, Miller JM, Wharton JM, Sung RJ, Slotwiner DJ,Markowitz SM, Lerman BB. Comparison of a novel rec-tilinear biphasic waveform with a damped sine wavemonophasic waveform for transthoracic ventricular defib-rillation. Am J Cardiol 1999;35:15951601.

    8. Niebauer MJ, Chung MK, Tchou PJ, Tchou PJ. Compari-son of the rectilinear biphasic waveform to the monophasic

  • 150 Niebauer et al.

    damped sine waveform for external cardioversion ofatrial fibrillation and flutter. Am J Cardiol 2004;93:14951499.

    9. Ricard P, Levy S, Trigano J, Paganelli F, Daoud E, ManKC, Strickberger SA, Morady F. Prospective assessmentof the minimum energy needed for external electrical car-dioversion of atrial fibrillation. Am J Cardiol 1997;79:815816.

    10. Joglar JA, Hamdan MH, Ramaswamy K, Zagrodzky JD,Sheehan CJ, Nelson LL, Andrews TC, Page RL. Initial en-ergy for elective external cardioversion of persistent atrialfibrillation. Am J Cardiol 2000;86:348345.

    11. Lau CP, Lok NS. A comparison of transvenous atrial de-fibrillation of acute and chronic atrial fibrillation and theeffect of intravenous sotalol on human atrial defibrillationthreshold. Pacing Clin Electrophysiol 1997;20:24422452.


View more >