External Cardioversion in Patients With Persistent Atrial Fibrillation

Transcription

1 External Cardioversion in Patients With Persistent Atrial Fibrillation A Reappraisal of the Effects of Electrode Pad Position and Transthoracic Impedance on Cardioversion Success Chien-Jen CHEN, 1 MD, and G. Bih-Fang GUO, 1 MD SUMMARY The optimal methods to perform external cardioversion of atrial fibrillation (AF) have yet to be conclusively determined. This study was performed to examine the relative efficacy of different pad positions on cardioversion success and the relationship between the transthoracic impedance (TTI) and energy requirement for AF cardioversion. Seventy patients with persistent AF undergoing elective cardioversion were randomly assigned to an electrode pad position situated either over the ventricular apex-right infraclavicular area (AL group, n = 31) or over the right lower sternal border-left infrascapular area close to the spine (AP group, n = 39). Energy was delivered at an initial 100 joules (J) and then increased to 150 J, 200 J, 300 J, and 360 J if needed. Energy and TTI readings were recorded. Mean TTI was significantly lower in the AP group than in the AL group. However, the cumulative success rates at each energy level were similar in the two groups (23% vs 19.4%, 41% vs 45.2%, 66.7% vs 74.2%, 79.5% vs 77.4%, and 84.6% vs 83.9% at 100 J, 150 J, 200 J, 300 J and 360 J, respectively). In the AP group, converters showed slightly lower TTI compared to nonconverters. In the AL group, converters showed significantly lower TTI compared to nonconverters. However, for all patients as a group, TTI was the only predictor for cardioversion success and showed a significant relationship to the energy required for cardioversion, which can be described by a quadratic equation. Rather than pad position, TTI is the single factor that significantly affects cardioversion and correlates with energy requirement. The relationship between energy requirement and TTI further allows estimation of energy requirements to achieve a successful cardioversion. (Jpn Heart J 2003; 44: ) Key words: Atrial fibrillation, Cardioversion, Impedance, Electrode pad position EXTERNAL direct current (DC) cardioversion, which is useful for cardioversion of atrial fibrillation (AF), has conventionally been performed by delivering an energy shock starting at 100 joules (J) followed by 200 J, 300 J, and 360 J, From 1 the Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital, Kaohsiung, Taiwan. Address for correspondence: G. Bih-Fang Guo, MD, Division of Cardiology, Department of Internal Medicine, Kaohsiung Chang Gung Memorial Hospital, 123, Ta-Pei Road, Niao-Sung Hsiang, Kaoshiung Hsien 83305, Taiwan, ROC. Received for publication March 5, Revised and accepted May 29,

2 922 CHEN AND GUO Jpn Heart J November 2003 through electrode pads positioned at the ventricular apex and the right infraclavicular area. 1-3) However, optimal methods to perform DC cardioversion have yet to be conclusively determined. Transthoracic impedance, which determines the effective passage of electrical energy through the heart, is a major determinant of the energy requirement for cardioversion. 4) Because transthoracic impedance varies with electrode pad position, 5-6) different pad positions may result in differing cardioversion success rates. Several studies suggest a higher cardioversion success rate when pads are positioned over the right parasternal area or sternal body and the left infrascapular area, rather than when electrode pads are conventionally positioned. 7-10) However, the differences in cardioversion success rate due to differing pad positions have not been confirmed by other studies. 5,6,11) Recently, a small, nonrandomized, and nonprospective study by Mehdiard, et al 11) suggested that improved DC cardioversion efficacy in patients with AF could be achieved by positioning pads under fluoroscopy to encompass as much atrial muscle as possible. In light of differing opinions and the lack of significant conclusive evidence, the relative effectiveness of different electrode positions on cardioversion success remains unclear. We performed our study in a prospective and randomized fashion to compare the success rate for cardioversion of AF using two different positions, namely the conventional pad position and an anteroposterior position that encompassed most of the right and left atria. We also examined the relationship between energy requirements and transthoracic impedance as definite insight has yet to be gained regarding the optimal level of the next shock energy should the starting energy shock of 100 J fail to be adequate for cardioversion. METHODS Patients: Between January 1999 and September 2001, 70 patients undergoing elective cardioversion for AF were enrolled in our study. Patient records were reviewed for clinical data including age, sex, height, weight, AF duration, left atrium diameter and left ventricular ejection fraction determined by echocardiography performed within the previous six months, hypertension, coronary artery disease, idiopathic dilated cardiomyopathy, hyperthyroidism, mitral and aortic regurgitation severity, rheumatic heart disease, lone atrial fibrillation, and history of prior treatment using antiarrhythmic drugs (amiodarone, beta-blockers, calcium channel blockers). Based on echocardiographic studies, a patient was considered to have a rheumatic heart disease if he or she displayed mitral or aortic stenosis and calcification in the presence of another valvular disease. A patient with moderate to severe aortic, mitral, or tricuspid regurgitation but without clear

3 Vol 44 No 6 CARDIOVERSION OF PERSISTENT ATRIAL FIBRILLATION 923 evidence of mitral or aortic stenosis and calcification was considered to have a nonrheumatic valvular disease. Inclusion criteria: Patients with persistent AF (duration 1 month) without a previous history of DC cardioversion for AF were included in this study, regardless of etiology, age, gender, and prior use of antiarrhythmic agents. Exclusion criteria: Patients with a new onset of AF (duration of AF less than 1 month), paroxysmal AF, pregnancy, unstable congestive heart failure (New York Heart Association functional class III to IV), spontaneous restoration to sinus rhythm before a cardioversion attempt, or rheumatic heart diseases for which a percutaneous transluminal mitral commisurotomy was not performed were excluded. Protocol for cardioversion: A DC cardioversion protocol for AF, approved by a local Research and ethics committee, was already in place at our hospital. Informed written consent was obtained from all patients prior to the study. The procedure was performed in the electrophysiology laboratory of our hospital. All enrolled patients were administered warfarin to ensure at least 4 weeks of adequate anticoagulation to the international normalized ratio between 2.0 and 3.0. Patients fasted through the night prior to treatment and were administered intravenous midazolam or diazepam to induce sedation and meperidine 50 mg for analgesia prior to DC cardioversion. The electrodes were self-adhesive, preapplied, low impedance, disposable electrode pads (R2 Medical System Inc, Carlsbad, California, USA) and were randomly positioned either over the ventricular apex and right infraclavicular area (designated as the AL position) or over the right lower sternal border and the left infrascapular area close to the spine (designated as the AP position). The latter electrode pad position was able to encompass most of the right and left atria, as confirmed by fluoroscopy. The patients received DC shocks from a Hewlett-Packard defibrillator (model 1722A/B Code Master XL+), which produced a damped sinusoidal waveform current. Shocks were delivered to patients at the moment of maximum expiration. Each patient received a step-up protocol of cardioversion, once at each level, with an initial shock energy setting of 100 J, followed by, if necessary, 150 J, 200 J, 300 J, and maximum 360 J shocks. A minimum of a 3-minute interval was allowed before delivering a shock at the next energy level. The Hewlett- Packard defibrillator used in our study was equipped with a built-in design that could automatically measure and annotate transthoracic impedance (TTI) on rhythm strips recorded for shocks at each energy level. A cardioversion attempt was deemed successful if sinus rhythm was restored and maintained throughout the whole observation period of at least 60 minutes in our electrophysiology laboratory.

4 924 CHEN AND GUO Jpn Heart J November 2003 Statistics: Data are expressed as the mean ± standard deviation (SD). Student's t test was used to compare the continuous variables between groups. The Chisquare test and Fisher's exact test were used to determine differences in categorical data between groups. The data of the energy level at successful conversions in relation to the TTI for each group of patients had a nonlinear shape and were fitted to a quadratic equation: y = ax 2 + bx + c, where x is the TTI and y is the energy required for successful cardioversion. Because transthoracic impedance (TTI) tended to decrease with each subsequent cardioversion, TTI at the level of 100 J shock only was used for statistical analysis. A probability value of less than 0.05 was considered to be statistically significant. RESULTS Clinical characteristics of study patients: The clinical characteristics of the study patients with atrial fibrillation are shown in Table I. Of the 70 patients undergoing DC cardioversion, 39 patients were treated at the AP electrode pad position Table I. Patient Characteristics and Clinical Demographic Data Group AP (n = 39) Group AL (n = 31) P Value Age (years) Gender (male) Height (cm) Weight (kg) Body mass index (kg/m 2 ) Duration of AF (months) Associated diseases Valvular heart disease Hypertension Rheumatic heart disease Idiopathic cardiomyopathy Coronary artery disease Lone atrial fibrillation Echocardiographic data Left atrial diameter (cm) Left ventricular ejection fraction Antiarrhythmic agents Amiodarone Beta-blockers Calcium channel blockers 57.6 ± (69.2%) ± ± ± ± (46.2%) 15 (38.5%) 11 (28.2%) 7 (17.9%) 3 (7.7%) 1 (2.7) 40.2 ± ± 13.6% 25 (64.1%) 8 (20.5%) 7 (17.9%) 59.1 ± (64.5%) ± ± ± ± (35.5%) 10 (32.3%) 8 (25.8%) 5 (16.1%) 1 (3.2%) 5 (16.1%) 40.8 ± ± 9.1% 18 (58.1%) 7 (22.6%) 4 (12.9%) Unless otherwise indicated, data presented are mean ± SD or number (%) of patients. AP: Electrode pads are positioned anteriorly at the right lower sternal border and posteriorly at the left infrascapular area close to the spine; AL: Electrode pads are positioned at the ventricular apex and right infraclavicular area; AF = atrial fibrillation.

5 Vol 44 No 6 CARDIOVERSION OF PERSISTENT ATRIAL FIBRILLATION 925 (group AP) and 31 patients were treated at the AL electrode pad position (group AL). The two groups were similar in terms of variables such as age, sex, height, weight, AF duration, left atrium diameter and left ventricular ejection fraction, associated diseases and prior treatment using antiarrhythmic drugs. Cardioversion results: TTI was significantly lower in the AP group than in the AL group (65.5 ± 14.5 vs 73.8 ± 16.8 Ohms, P = 0.033), although the success rate achieved at each energy level was similar (Figure 1). The combined cumulative success rates for cardioverisons performed were 21.4% (15/70), 42.9% (30/70), 70% (49/70), 78.6% (55/70), and 84.3% (59/70), at energy levels of 100 J, 150 J, 200 J, 300 J, and 360 J, respectively. Apart from transient bradycardia below 60 beats/min in 4 patients (for which none required temporary cardiac pacing), we recorded no complications following cardioversion procedures. Comparisons between successful and unsuccessful cardioversions: After considering all of the clinical variables that could potentially influence cardioversion effectiveness, we found TTI to be the only one to have a significant impact upon cardioversion success (Table II). In the AL group, TTI was significantly lower in Figure 1. Cumulative cardioversion success rate at each energy level according to pad positions. AP: Electrode pads positioned over the right sternal border-left infrascapular area close to the spine; AL: Electrode pads positioned over the ventricular apex-right infraclavicular area.

6 926 CHEN AND GUO Jpn Heart J November 2003 Table II. Comparison of Successful and Unsuccessful Cardioversions Unsuccessful (n = 11) Successful (n = 59) P Value Age (years) Gender (male) Height (cm) Body weight (kg) Body mass index (kg/m 2 ) AF duration (months) Impedance at 100 J shock (ohms) Impedance in Group AP Impedance in Group AL Associated diseases Valvular heart disease Hypertension Rheumatic heart disease Idiopathic cardiomyopathy Coronary artery disease Lone atrial fibrillation Echocardiographic data Left atrial diameter (cm) Left ventricular ejection fraction (%) Antiarrhythmic agents Amiodarone Beta-blockers Calcium channel blockers 62.0 ± (54.5%) ± ± ± ± ± ± ± (54.5%) 5 (45.5%) 2 (18.2%) 1 (9.1%) 2 (18.2%) 1 (9.1%) 41.1 ± ± 7.4% 4 (36.4%) 3 (27.3%) 4 (36.4%) 57.6 ± (69%) ± ± ± ± ± ± ± (39%) 19 (33.3%) 17 (28.8%) 11 (18.6%) 2 (3.4%) 5 (8.5%) 40.4 ± ± 12.3% 39 (66.1%) 12 (20.3%) 7 (11.9%) Unless otherwise indicated, data presented are number (%) of patients. Abbreviations as in Table I. converters than in nonconverters (70.1 ± 13.7 Ohms versus 96.5 ± 17.9 Ohms, P = 0.002). In the AP group, the difference in mean TTI values, while lower in converters than in nonconverters, was not statistically significant (64.0 ± 13.5 Ohms versus 73.5 ± 18.3 Ohms, P = 0.142). TTI was found to be significantly lower in patients with cardioversion success than in those without success among patients in both groups (66.4 ± 13.8 Ohms and 82.7 ± 20.8 Ohms, P = 0.003). We did not find other factors, such as body weight, height, body mass index (BMI, kg/m 2 ), associated diseases, left atrial diameter, left ventricular ejection fraction, duration of AF, and use of antiarrhythmic agents to affect the outcome of cardioversion (Table II). Even when patients were divided into subgroups based on AF duration (either shorter than or longer than 6 months), we did not find AF duration to be a cardioversion success predictor. The success rate was 82.6% (18/23) in patients with AF duration of shorter than 6 months and 85.1% (40/47) in patients with AF duration of 6 months or longer (P = 0.787). Of the

7 Vol 44 No 6 CARDIOVERSION OF PERSISTENT ATRIAL FIBRILLATION 927 A B Figure 2. Relationship between energy requirement and transthoracic impedance in patients with successful cardioversion in the AP group (A) and AL group (B). Regression equation, where x is impedance and y is energy threshold. Correlation coefficients (r, P = 0.001) are shown. Abbreviations as in Figure 1. antiarrhythmic agents previously used by our patients, only amiodarone was shown to have a tendency to increase the success rate (P = 0.063). Energy requirement for successful cardioversion: Age, gender, associated diseases, AF duration, left atrial diameter, left ventricular fraction, or prior use of antiarrhythmic agents did not affect the energy required for successful cardioverion in each pad position. In each group, however, the energy required for successful cardioversions correlated significantly with TTI. The relationship between energy requirement for successful conversion and TTI, based on electrode pad position, is shown in Figure 2. In each pad position, a quadratic equation well describes this relationship with a statistically significant correlation coefficient (r = 0.55, P = in the AP group and r = 0.66, P = in the AL group). DISCUSSION Our results indicate that electrode pad position is not a critical determinant of successful cardioversion, even though impedance was found to vary with different pad positions. However, TTI was found to be the single predictor for cardioversion success and to correlate significantly with the amount of energy required. This relationship between energy requirement and TTI further allows estimation of energy requirements to achieve a successful cardioversion. Pad position and cardioversion: The ventricular apex-right infraclavicular area has been the standard pad placement position for cardioversion, although it is known that the optimal electrode position should permit current passage through the entire atrial mass to attain cardioversion success. Previous studies comparing

8 928 CHEN AND GUO Jpn Heart J November 2003 the relative efficacy of pad positions on AF cardioversion have yielded divergent results, with several studies indicating that position has no bearing on cardioversion success 5,6,11) and other studies showing that anteroposterior positioning achieves significantly improved results. 7-10,12) In the study by Mehdiard, et al 11) 360 J DC shocks with the electrode pads positioned anteroposteriorly (under fluoroscopic guidance) to cover the atrial mass as much as possible were able to successfully induce sinus rhythm in 8 of 15 patients who had failed cardioversion using a conventional pad position. The recent randomized study by Kirchhof, et al 12) also supported the advantage of an anterior-posterior position over the anterior-lateral position for external cardioversion of persistent atrial fibrillation. However, the results of our study using electrode pad positions similar to those recommended by Mehdiard, et al do not show that pad position is a critical determinant for cardioversion. The differences in research results may be related to different patient characteristics, such as AF duration, underlying cardiovascular diseases, or the use of drugs. In addition, the difference may also be due to individual variations in heart axis, which may affect the effectiveness of electric current through the critical mass of atria. Therefore, more current expected from a lower TTI at a particular pad position may not translate into more effective current across the atria. It should also be noted that the study of Mehdiard, et al 11) involved a very small number of patients and was not carried out in a prospective and randomized fashion that would permit a comparison of the success rates of cardioversion using a fluoroscopy-guided pad position and cardioversion employing conventional electrode pad positioning. It is also difficult to compare our results with those of Kirchhof, et al 12) because the criteria of a successful cardioverison were not clearly defined and TTI was not measured in their study. In addition, the lack of discussion about the distribution of outcome according to gender raised the possibility of uncertainty regarding their conclusion, as the female breast might affect the success rate with anterior-lateral electrode positioning. 13) Clinical predictors of cardioversion success: Several investigators concluded that AF duration was an important factor affecting the technical success of external cardioversion 9,11,14,15) with a lower success rate achieved if atrial fibrillation persisted for longer than 6 months. 3,9,15) In contrast, our data revealed that even in patients with relatively long AF duration, eg, > 6 months (median 9 months), success rates were similar to those in patients of AF duration < 6 months. Therefore, our data suggest that DC treatment administered to control AF rhythm should not be overlooked when treating patients with prolonged AF duration. A previous study 16) suggested that left atrial size was related to cardioversion success. In contrast, our study failed to demonstrate that left atrial size based

9 Vol 44 No 6 CARDIOVERSION OF PERSISTENT ATRIAL FIBRILLATION 929 on M-mode measurement in AF patients, was a predictor of successful cardioversion. Pretreatment of AF patients with amiodarone for DC shock when AF duration was 16.3 weeks 17) or less 18) was shown to deliver a higher cardioversion success rate compared to pretreatment with digoxin and diltiazem. In our study, there was a tendency toward a higher success rate of cardioversion in patients who had undergone amiodarone pretreatment than those who did not (P = 0.063). It should be noted that our patients, on average, had a relatively long AF duration (mean 27.7 months). Overweight patients (BMI > 25 kg/m 2 ) undergoing external cardioversion were at twice the risk of having an unsuccessful external cardioversion compared to patients with a body mass index < 25 kg/m 2. 15,19) Specifically, patients with a body weight of less than 80 kg had a good chance of successful AF cardioversion. We found that neither weight and height, nor body mass index influenced cardioversion success. Our results may be explained by the relatively smaller body build of East Asians as compared to Caucasians. Only 18.8% of our patients weighed 80 kg. Therefore, our study suggested that regardless of a large atrial size, prolonged AF duration, excessive body weight, and the presence of underlying heart diseases, all patients with AF, particularly those who are symptomatic, should be given the chance to undergo cardioversion as a method to restore sinus rhythm. The administration of amiodarone may be considered prior to cardioversion for patients with long AF duration. Impedance and cardioversion success: Upon defibrillator capacitor discharges, the amount of current delivered depends on the impedance between the electrodes. The higher the impedance, the lower the current delivered. It is the density of current, which traverses the muscle of the chamber to be defibrillated, that determines cardioversion. Thus, to be successful, a critical muscle mass of the atria must be defibrillated. Our results indicate that TTI significantly affected the success of the external cardioversion technique for AF in patients with AL pad positions, but did not significantly affect the success of the technique in patients with AP pad positions, although a trend toward such was observed in the latter group (P = 0.142). The failure to have a significant TTI effect on AP group cardioversion success most likely is attributable to a type 2 error caused by the small number of study patients who failed cardioversion. Energy required for successful cardioversion: A shock of 100 J has been suggested as the starting level of energy for AF cardioversion. 1-3) However, cardioversion using 100 J only achieves a very modest success rate (50% or less). This figure rises to about 75-85% when a 200 J shock is then administered. 20,21) Other

10 930 CHEN AND GUO Jpn Heart J November 2003 studies have suggested that 360 J, or 200 J at a minimun, should be used as an initial energy level for elective cardioversion of persistent AF. 22,23) Gallagher, et al 23) even reported that, for patients with AF for more than 30 days, shocks of < 200 J simply were a waste of time. They observed that in patients with AF for > 180 days, an initial setting of 300 J reduced the total energy used but achieved a modest success rate of 56%. For patients in our study, the cumulative success rate was 21.4% at 100 J and 42.9% at 150 J, and rose to 70% at 200 J of shock. With 360 J of shock, we achieved an 84% success rate. We did not find, as Gallagaher, et al 23) did in their study, that AF duration affects the energy requirement. Thus, based on their and our results, it is reasonable to start with a 200 J shock, as this level of energy is capable of achieving a reasonable AF conversion success rate, supporting the recent recommendations for electrical cardioversion of atrial fibrillation. 24) To minimize the total number of shocks delivered and the discomfort to patients, 360 J should be the next level used if the 200 J shock fails because there is no evidence of cardiac damage induced by shocks at this high level of energy. 25,26) In our study, TTI was the only predictor of DC shock success. TTI correlated significantly with the energy required to achieve a successful shock. Previous studies also indicated that, for patients with lower TTI values (< 70 Ohms), low energy shocks were more often successful than in patients with TTI > 70 Ohms. 6,11,21) Kerber, et al 21) demonstrated that a modified Hewlett-Packard defibrillator, which could automatically double the energy in response to a high TTI (> 70 Ohms), could improve the rate of shock success. In our study, the relationship between energy level required for successful cardioversion and TTI should allow us to adjust the energy level required to deliver a successful cardioversion in each pad position (Figure 2A and 2B). Therefore, an alternative strategy targeted to minimize the total number of shocks and energy delivered would be to start with a 100 J shock and adjust the next energy level based on the relationship between the energy requirement for conversion and TTI. Of note, this relationship as suggested from our data should not be interpreted as an assurance of successful cardioversion if TTI is low, because other factors, such as underlying disease processes and atrial pathology, may be potentially important in determining cardioversion success. On the other hand, cardioversion failure may be reasonably predicted if TTI is excessively high. Limitations: There are two major limitations to our study. First, while the study was carried out in a prospective and randomized fashion, the number of participants was relatively small. Small group size may limit our statistical power in detecting and defining differences between the groups. Second, our patient population was heterogeneous in terms of associated diseases, AF duration, and the use of antiarrhythmic agents. However, this heterogeneity may, in fact, also

11 Vol 44 No 6 CARDIOVERSION OF PERSISTENT ATRIAL FIBRILLATION 931 reflect patients in the broader population that we actually encounter in daily practice. Defibrillators producing a biphasic waveform current are increasingly available. 27) Future studies using this new modality for cardioversion of AF are warranted to examine the relationship between impedance and energy requirement for successful cardioversion. Conclusion: Electric pad position may not be critical to successful AF cardioversion, even though impedance varies with paddle positions. The clinical characteristics of the patients, including AF duration, echocardiographic data on left atrial size and left ventricular function, associated diseases, and the use of antiarrhythmic agents, do not affect the success of DC cardioversion. TTI is the single factor associated in a significant manner with cardioversion success and energy requirement. Our data also suggest that a 200 J shock, as the starting level of energy, followed by a 360 J shock if needed, is recommended in order to reduce the total number of shocks administered. If a 100 J shock is employed in the initial attempt and fails to attain normal sinus rhythm, a second shock at a higher energy level, based on the relationship between energy requirement and TTI, should be adopted. REFERENCES 1. Yurchak PM, Williams SV, Achord JL, et al. Clinical competence in elective direct current (DC) cardiversion. A statement for physicians from AHA/ACC/ACP task force on clinical privileges in cardiology. Circulation 1993; 88: Lip GYH, Watson RD, Singh SP. Cardioversion of atrial fibrillation. Brit Med J 1996; 312: Kerber RE. Transthoracic cardioversion of atrial fibrillation and flutter: Standard techniques and new advances. Am J Cardiol 1996; 78: Kerber RE, Grayzel J, Hoyt R, et al. Transthoracic resistance in human defibrillation: Influence of body weight, chest size, serial shocks, paddle size and paddle contact pressure. Circulation 1981; 63: Kerber RE, Jensen SR, Grayzel J, et al. Elective cardioversion: Influence of paddle-electrode location and size on success rates and energy requirements. N Eng J Med 1981; 305: Dalzell GW, Anderson J, Adgey AAJ. Factors determining success and energy requirements for cardioversion of atrial fibrillation. Q J Med 1991; 78: Lown B. Electrical reversion of cardiac arrhythmias. Brit Heart J 1967; 29: Morris JJ Jr, Kong Y, North WC, et al. Experiences with cardioversion of atrial fibrillation and flutter. Am J Cardiol 1964; 14: Botto GL, Politi A, Bonini W, et al. External cardioversion of atrial fibrillation: role of paddle position in technical efficacy and energy requirements. Heart 1999; 82: Mathew TP, Moore A, McIntyre M, et al. Randomized comparison of electrode positions for cardioversion of atrial fibrillation. Heart 1999; 81: Mehdirad AA, Clem KL, Love CJ, et al. Improved clinical efficacy of external cardioversion by fluoroscopic electrode positioning and comparison to internal cardioversion in patients with atrial fibrillation. PACE 1999; 22: Kichhof P, Eckardt L, Loh P, et al. Anterior-posterior versus anterior-lateral electrode positions for external cardioversion of atrial fibrillation: a randomized trial. Lancet 2002; 360:

12 932 CHEN AND GUO Jpn Heart J November Pagen-Carlo LA, Spencer KT, Robertson CE, et al. Transthoracic defibrillation: importance of avoiding electrode placement directly on the female breast. J Am Coll Cardiol 1996; 27: Waris E, Kreus KE, Salokannel J. Factors influencing persistence of sinus rhythm after DC shock treatment of atrial fibrillation. Acta Med Scand 1971; 189: Fick M, Frykman V, Jensen-Urstad M, et al. Factors predicting success rate and recurrence of atrial fibrillation after first electrical cardioversion in patients with persistent atrial fibrillation. Clin Cardiol 2001; 24: Henry WL, Morganroth J, Pearlman AS, et al. Relation between echocardiographically determined left atrial size and atrial fibrillation. Circulation 1976; 53: Villani GQ. Effects of diltiazem pretreatment on direct-current cardioversion in patients with persistent atrial fibrillation: A single-blinded, randomized, controlled study. Am Heart J 2000; Capucci A, Villani GQ, Rosi A, et al. Oral amiodarone increases the efficacy of direct-current cardioversion in restoration of sinus rhythm in patients with chronic atrial fibrillation. Eur Heart J 2000; 21: Alt E, Ammer R, Lehmann G, et al. Patient characteristics and underlying heart disease as predictors of recurrent atrial fibrillation after internal and external cardioversion in patients treated with oral sotalol. Am Heart J 1997; 134: Ricard P, Levy S, Trigano J, et al. Prospective assessment of the minimal energy needed for external electrical cardioversion of atrial fibrillation. Am J Cardiol 1997; 79; Kerber RE, Martins JB, Kienzle MG, et al. Energy, current, and success in defibrillation and cardioversion: clinical studies using an automated impedance-based method of energy adjustment. Circulation 1988; 77: Joglar JA, Hamdan MH, Ramaswamy K, et al. Initial energy for elective external cardioversion of persistent atrial fibrillation. Am J Cardiol 2000: 86; Gallagher MM, Guo XH, Poloneiecki JD, et al. Initial energy setting, outcome and efficacy in direct current cardioversion of atrial fibrillation and flutter. J Am Coll Cardiol 2001; 38: Fuster V, Ryden LE, Asinger RW, et al. ACC/AHA guidelines for the management of patients with atrial fibrillation: executive summary. Circulation 2001; 104: Neumayr C, Hagn C, Ganzer H, et al. Plasma levels of troponin T after electrical cardioversion of atrial fibrillation and flutter. Am J Cardiol 1997; 80: Rao ACR, Naeem N, John C, et al. Direct current cardioversion does not cause cardiac damage: evidence from cardiac troponin T estimation. Heart 1998; 80; Page RL, Kerber RE, Russell JK, et al. Biphasic versus monophasic waveform for cardioversion of atrial fibrillation. J Am Coll Cardiol 2002; 39: