A comparison of 50-J versus 100-J shocks fordirect-current cardioversion of atrial flutterSergio L. Pinski, MD, Elena B. Sgarbossa, MD, Elizabeth Ching, RN, and Richard G. Trohman, MD Cleveland, Ohio
Background Direct-current cardioversion remains the gold standard for restoration of sinus rhythm in patients withatrial flutter. Although an initial energy of 50 J is recommended, the optimal energy settings have not been evaluated in alarge series of contemporary patients.
Methods We compared the outcome of cardioversion with 50 J versus 100 J in 330 consecutive patients with atrial flut-ter. Initial energy was based on attending physician preference. One hundred sixty patients received 50 J and 170 patientsreceived 100 J.
Results Patients in both groups did not differ significantly in age, sex, weight, body mass index, duration of the arrhyth-mia, postoperative status, presence and type of structural heart disease, or use of antiarrhythmic drugs. Patients in the 100-Jgroup had more first shock conversion (85% vs 70%; P = .001), fewer total shocks (1.2 0.5 vs 1.4 0.7; P = .001), andless induction of atrial fibrillation (2% vs 11%; P = .002). There were no significant differences in overall restoration of sinusrhythm, cumulative energy delivered, anesthetic dose, and procedure room time. On multivariate analysis, delivery of 100 Jwas the strongest predictor of first shock success (odds ratio 2.6, 95% confidence interval 2.13 to 3.16; P < .001).
Conclusion An initial energy of 100 J is more efficient for restoration of sinus rhythm in patients with atrial flutter. (AmHeart J 1999;137:439-42.)
Atrial flutter is a common arrhythmia.1 Despite recentadvances in pharmacologic2 and pacing3 conversion,direct-current (DC) cardioversion remains the mosteffective method for acute termination of atrial flutter.The first reports by Lown et al4,5 35 years ago noted thatcardioversion of atrial flutter could be accomplishedwith less energy than cardioversion of atrial fibrillation.These (and other) investigators6,7 also observed that itwas not uncommon for QRS-synchronized shocks ofenergies
considered postoperative when cardioversion was per-formed within 30 days of cardiac surgery.
DC cardioversionThe technique of electrical cardioversion used in our labora-
tory has been previously described.13 Briefly, maintenancedoses of antiarrhythmic agents were given as usual. Anesthesiawas achieved with intravenous methohexital. Synchronizedshocks were delivered via autoadhesive pads14 (Zoll Stat-Pads,Zoll Medical) placed in an anteroposterior configuration andconnected to an external cardioverter-pacemaker unit (Zoll). Inwomen, placement of the anterior pad directly on the breastwas avoided.15 The initial shock energy (50 J or 100 J) wasselected by the attending physician. Some operators invariablyselected 50 J, others routinely selected 100 J. Subsequent higherenergy shocks were delivered if the initial shock failed torestore normal rhythm. When normal rhythm was restored onlytransiently, at least 1 further shock at the initially successfulenergy level was attempted. The total number of shocks deliv-ered after an unsuccessful first attempt was at the discretion ofthe attending physician. For the purpose of this study, firstshock success was defined when a nontachyarrhythmic rhythmwas documented for at least 30 seconds after shock delivery.Overall success was defined by departure of the patient fromthe procedure room in a nontachyarrhythmic rhythm.
Statistical methodsContinuous variables are presented as mean 1 SD. Base-
line characteristics and outcomes in the 50-J and 100-J groupswere compared by an independent t test for continuous vari-ables and a chi-square test for discrete variables. Univariatepredictors of first shock success were determined by logisticregression for continuous variables and chi-square tests for
American Heart JournalMarch 1999Pinski et al440
discrete variables. Independent predictors of first shock suc-cess were determined by forward stepwise logistic regression.A 2-tailed P value < .05 was considered significant.
ResultsTable I shows the baseline demographic and clinical
characteristics in patients receiving 50 or 100 J initialshocks. There were no significant differences in any ofthe analyzed variables between groups.
Table II shows the outcomes in both groups. Therewas a significantly higher first shock success rate inthe 100-J group. Patients in the 100-J group requiredfewer shocks and had less shock-induced atrial fibrilla-tion. Overall success rates, cumulative energies deliv-ered, procedure duration, and methohexital dose werenot significantly different. For patients who failed theinitial shock, subsequent cumulative energy deliverywas similar in both groups (260 207 J in the 50-Jgroups versus 281 190 J in the 100-J group).
Only 2 complications occurred. A patient developedatrial fibrillation with a very slow ventricular responseafter a 50-J first shock and required intravenousatropine. A subsequent shock restored sinus rhythm.One patient with chronic obstructive pulmonary dis-ease who received a 100-J shock developed respira-tory distress. The total methohexital dose was 1.03mg/kg. The patient was treated with subcutaneousepinephrine, diuretics, and respiratory therapy. Allsymptoms subsided within 45 minutes.
Univariate correlates of first shock success were 100-Jshock (P = .001), lower height (P = .004), lower weight(P < .001), lower body mass index (P < .001), typical
50 J (n = 160) 100 J (n = 170) P value
Age (y) 65 13 64 14 .71Men, n (%) 131 (82) 135 (79) .57Duration of flutter (d) 34 175 13 22 .18Weight (kg) 81 18 83 18 .22Height (cm) 173 9 172 9 .70Body mass index kg/m2 26.8 5 27.9 5 .06Cardiac disease, n (%)
None 20 (12) 26 (15) .46Coronary 76 (47) 74 (44) .47Valvular 44 (27) 50 (29) .70Idiopathic cardiomyopathy 17 (11) 9 (5) .07
Postoperative 33 (21) 45 (26) .21Antiarrhythmic drugs 101 (63) 107 (67) .97
Class I A 55 (34) 51 (30) .39Class I C 16 (10) 22 (13) .40Amiodarone 29 (18) 30 (18) .90Sotalol 6 (4) 9 (6) .50
Table I. Baseline characteristics
atrial flutter (P = .02), presence of coronary artery dis-ease (P = .04), and absence of idiopathic dilated car-diomyopathy (P = .003). When the 26 patients withidiopathic dilated cardiomyopathy were excluded fromanalysis, 100-J shocks remained significantly associatedwith first shock success (86% versus 73%, P = .005).
Delivery of 100-J shock remained the strongest inde-pendent predictor of first shock success in the multivari-able analysis (odds ratio 2.6, 95% confidence interval [CI]2.13 to 3.16; P < .001). Atypical flutter (odds ratio 0.23,95% CI 0.06 to 0.8; P = .02) and idiopathic dilated car-diomyopathy (odds ratio 0.34, 95% CI 0.14 to 0.8; P =.01) were independent predictors of first shock failure.
DiscussionPatients with atrial flutter constitute approximately
25% of the population in several recent large series ofcardioversion for supraventricular arrhythmias.13,16,17
DC cardioversion remains the gold standard for acutetermination of atrial flutter. The potential benefits ofprompt restoration of sinus rhythm in patients withatrial flutter include prevention of atrial electricalremodeling,18 decreased risk of thromboembolism,19
the restoration of atrial transport function,20 andimprovement in left ventricular function.21
Because electrical cardioversion is efficacious andsafe, there has been limited effort to refine this tech-nique. The current emphasis on cost- and time-effi-ciency makes reappraisal of even the most time-hon-ored techniques worthwhile. In this study, we foundthat an initial energy of 100 J was significantly moreeffective than 50 J for elective DC cardioversion ofatrial flutter. This increased first shock effectivenesswas not associated with adverse events or increasedanesthetic requirements. Because fewer patients wereconverted with the initial 50-J shock, the cumulativeenergies delivered with both strategies were not signif-icantly different. Our findings suggest that, contrary toprevious recommendations, 100 J (and not 50 J)
should be used as initial energy for elective cardiover-sion of atrial flutter. Our findings cannot be explainedby a lower than expected success rate for the 50-Jshock. Our 70% success rate for the 50-J shock com-pares favorably with previous reports. Although theinitial energy was not randomized, operators followedrote personal preferences, which created nearly identi-cal patient groups. Therefore, selection bias wasunlikely to influence our results.
The cellular mechanism of DC cardioversion is notknown, but it appears that achievement of a critical cur-rent density causing a rapid change in local myocardialvoltage (ie, the potential gradient) is its best determi-nant. The current density delivered to the myocardiumdepends on the energy delivered, transthoracic (andmyocardial) impedance,22 plus electrode position, size,contact, and pressure. Identification of the optimal car-dioversion energy is clinically important because thecurve relating success rate to energy (or current) deliv-ered has an inverted U-shape.23 If the current density istoo large or not large enough, atrial flutter will not ter-minate. Additionally, very high current densities mayproduce myocardial damage.24 Chalasani et al25 com-pared the efficacy of shocks of
Other factors appear less important in determiningshock success. In agreement with randomized dataobtained in patients with atrial fibrillation, we did notfind a significant independent effect of treatment withclass I drugs on the energy requirements for cardiover-sion.28 The cardioversion energy requirements for typeII and other forms of atrial flutter have not been stud-ied. In our study, atypical atrial flutter was an indepen-dent predictor of first shock failure. As curative catheterablation for typical atrial flutter becomes more wide-spread, the proportion of patients presenting with atyp-ical forms of flutter will increase and will providestronger justification for our recommendation of 100 Jas initial the energy for cardioversion.
A review of recent clinical trials suggests that manycommonly held beliefs and long-established clinicalpractices need to be challenged.29 The simplicity ofour recommendation makes it very appealing. Reluc-tance to be more aggressive is not justified. Car-dioversion of atrial flutter with 100 J is more effica-cious, less proarrhythmic, and well suited to theevolving arrhythmia substrates that are byproducts ofadvances in interventional electrophysiology.
References1. Waldo AL, Touboul P, editors. Atrial flutter: advances in mechanisms
and management. Armonk, New York: Futura Publishing; 1996.2. Foster RH, Wilde MI, Markham A. Ibutilide. A review of its phar-
macological properties and clinical potential in the acute managementof atrial flutter and fibrillation. Drugs 1997;54:312-30.
3. Tucker KJ, Wilson C. A comparison of transesophageal atrial pacingand direct current cardioversion for the termination of atrial flutter: aprospective, randomised clinical trial. Br Heart J 1993;69:530-5.
4. Lown B, Amarasingham R, Newman J. New method for terminatingcardiac arrhythmias: use of synchronized capacitor discharge. JAMA1962;182:548-55.
5. Lown B, Kleiger R, Wolff G. The technique of cardioversion. Am HeartJ 1964;67:282-4.
6. Mah GK. Cardioversion in the treatment of atrial flutter. SingaporeMed J 1969;10:164-9.
7. Neuman J, Wigutov J, Pesce R, Valero E, Drajer S. Desencade-namiento de fibrilacin auricular durante la cardioversin de pacientescon aleteo auricular. Medicina (Buenos Aires) 1973;33:110-6.
8. Lown B. Electrical reversion of cardiac arrhythmias. BMJ 1967;29:469-89.
9. Emergency Cardiac Care Committee and Subcommittees, AmericanHeart Association. Guidelines for cardiopulmonary resuscitationand emergency cardiac care. Part III: adult advanced cardiac lifesupport. JAMA 1992;268:2199-241.
10. Josephson ME, Buxton AE, Marchlinski FE. The tachyarrhythmias. In:Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, KasperDL, editors. Harrisons principles of internal medicine, 13th edition.New York: McGraw-Hill; 1994. p. 1019-36.
11. Olshansky B, Kerber RE. Cardioversion of atrial flutter. In: Waldo AL,
Touboul P, editors. Atrial flutter: advances in mechanisms and man-agement. Armonk, New York: Futura Publishing; 1996. p. 387-409.
12. Waldo AL, Biblo LA. AV nodal-independent supraventricular tachy-cardias. In: Topol EJ, editor. Textbook of cardiovascular medicine.Philadelphia: Lippincott-Raven; 1997. p. 1695-725.
13. Tobin MG, Pinski SL, Tchou PJ, Ching EA, Trohman RG. Cost-effec-tiveness of intravenous anesthetics for direct-current cardioversion bynonanesthesiologists. Am J Cardiol 1997;79:686-8.
14. Kerber RE, Martins JB, Kelly KJ, Ferguson DW, Kouba C, Jensen SR,et al. Self-adhesive preapplied electrode pads for defibrillation andcardioversion. J Am Coll Cardiol 1984;3:815-20.
15. Pagan-Carlo LA, Spencer KT, Robertson CE, Dengler A, Birkett CA,Kerber RE. Transthoracic defibrillation: importance of avoidingelectrode placement directly on the female breast. J Am CollCardiol 1996;27:449-52.
16. Sanchez Diaz CJ, Gonzalez Carmona VM, Ruesga Zamora E,Monteverde Grether CA. Electric cardioversion in the emergencyservice. Experience in 1000 cases. Arch Inst Cardiol Mex 1987;57:387-94.
17. Arnold AZ, Mick MJ, Mazurek RP, Loop FD, Trohman RG. Role of pro-phylactic anticoagulation for direct current cardioversion in patientswith atrial fibrillation or atrial flutter. J Am Coll Cardiol 1992;19:851-5.
18. Franz MR, Karasik PL, Li C, Moubarak J, Chavez M. Electrical remod-eling of the human atrium: similar effects in patients with chronic atrialfibrillation and atrial flutter. J Am Coll Cardiol 1997;30:1785-92.
19. Lanzarotti CJ, Olshansky B. Thromboembolism in chronic atrial flutter:is the risk underestimated? J Am Coll Cardiol 1997;30:1506-11.
20. Jordaens L, Missault L, Germonpr E, Callens B, Adang L, Vanden-bogaerde J, et al. Delayed restoration of atrial function after con-version of atrial flutter by pacing or electrical cardioversion. Am JCardiol 1993;71:63-7.
21. Karnegis JN, Matoole JJ, Bjorling VG. Immediate improvement inleft ventricular function after cardioversion of atrial flutter. Am J Car-diol 1989;64:1043-7. (Published erratum appears in Am J Cardiol1990;66:772.)
22. Kerber RE, Kouba C, Martins J, Kelly K, Low R, Hoyt R, et al. Advanceprediction of transthoracic impedance in human defibrillation andcardioversion: importance of impedance in determining the successof low-energy shocks. Circulation 1984;70:303-8.
23. Kerber RE, Martins JB, Kienzle MG, Constantin L, Olshansky B,Hopson R, et al. Energy, current, and success in defibrillation andcardioversion: clinical studies using an automated impedance-basedmethod of energy adjustment. Circulation 1988;77:1038-46.
24. Kerber RE. Transthoracic cardioversion of atrial fibrillation and flutter:standard techniques and new advances. Am J Cardiol 1996;78:22-6.
25. Chalasani P, Cambre S, Silverman ME. Direct-current cardioversionfor the conversion of atrial flutter. Am J Cardiol 1996;77:658-60.
26. Guiney TE, Lown B. Electrical conversion of atrial flutter to atrialfibrillation. Flutter mechanism in man. Br Heart J 1972;34:1215-24.
27. Florin TJ, Weiss DN, Peters RW, Shorofsky SR, Gold MR. Induction ofatrial fibrillation with low-energy defibrillator shocks in patients withimplantable cardioverter defibrillators. Am J Cardiol 1997;80:960-2.
28. Bianconi L, Mennuni M, Lukic V, Castro A, Chieffi M, Santini M.Effects of oral propafenone administration before electrical car-dioversion of chronic atrial fibrillation: a placebo-controlled study. JAm Coll Cardiol 1996;28:700-6.
29. Califf RM. Considerations in the design, conduct and interpretation ofquantitative clinical evidence. In: Topol EJ, editor. Textbook of cardio-vascular medicine. Philadelphia: Lippincott-Raven; 1997. p. 1063-81.
American Heart JournalMarch 1999Pinski et al442