Article Text

Download PDFPDF

External cardioversion of atrial fibrillation: role of paddle position on technical efficacy and energy requirements
  1. G L Botto,
  2. A Politi,
  3. W Bonini,
  4. T Broffoni,
  5. R Bonatti
  1. Department of Cardiology, Ospedale “Sant' Anna”, Via Napoleona 60, 22100 Como, Italy
  1. Dr Botto email: ccaec{at}


AIM To define the effect of defibrillator paddle position on technical success and dc shock energy requirements of external cardioversion of atrial fibrillation.

METHODS 301 patients (mean (SD) age 62 (11) years) with stable atrial fibrillation were randomly assigned to elective external cardioversion using anterolateral paddle position (ventricular apex–right infraclavicular area; group AL (151 patients)) or anteroposterior paddle position (sternal body–angle of the left scapula; group AP (150 patients)). A step up protocol was used, delivering a 3 J/kg body weight dc shock, then a 4 J/kg shock (maximum 360 J), and finally a second 4 J/kg shock using the alternative paddle location.

RESULTS The two groups were comparable for the all clinical variables evaluated. The cumulative percentage of patients successfully converted to sinus rhythm was 58% in group AL and 67% in group AP with low energy dc shock (NS); this rose to 76% in group AL and to 87% in group AP with high energy dc shock (p = 0.013). Thirty seven patients in group AL and 19 in group AP experienced dc shock with the alternative paddle position; atrial fibrillation persisted in 10/37 in group AL and in 10/19 in group AP. Mean dc shock energy requirements were lower for group AP patients than for group AL patients, at 383 (235)v 451 (287) J, p = 0.025. Arrhythmia duration was the only factor that affected the technical success of external cardioversion (successful: 281 patients, 80 (109) days; unsuccessful: 20 patients, 193 (229) days; p < 0.0001). The success rate was lower if atrial fibrillation persisted for > 6 months: 29 of 37 (78%) v 252 of 264 (95%); p = 0.0001.

CONCLUSIONS An anteroposterior defibrillator paddle position is superior to an anterolateral location with regard to technical success in external cardioversion of stable atrial fibrillation, and permits lower dc shock energy requirements. Arrhythmia duration is the only clinical variable that can limit the restoration of sinus rhythm.

  • atrial fibrillation
  • cardioversion
  • electric countershock

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Direct current defibrillation was introduced in clinical practice by Lown in the early 1960s.1 ,2 Cardioversion, the transthoracic application of dc shocks, remains a widely used clinical approach to terminate atrial fibrillation or flutter.3 ,4 For the procedure itself, various types of electrode may be used including traditional hand held paddle electrodes and self adhesive electrodes.5 Various chest placements have been used: the apex-anterior, apex-posterior, and anterior-posterior chest placements have been shown to be equally effective.6-8 The aim of this study therefore was to determine in a prospective fashion the effect of defibrillator paddle position on technical success and dc shock energy requirements of elective external cardioversion of atrial fibrillation.


The research protocol was approved by the locally appointed ethics committee and informed consent was obtained from each patient.

The subjects of this prospective study were 301 consecutive patients who underwent elective external cardioversion for stable atrial fibrillation between August 1993 and December 1997. Standard clinical criteria were used to select patients for cardioversion.9 ,10 We excluded those with: haemodynamically unstable atrial fibrillation in which cardioversion needed to be performed urgently; left atrial dimension > 60 mm measured by M mode echocardiography; arrhythmia duration either > 2 years or of unknown duration; and untreated hyperthyroidism.

The patients were sedated with propofol, and external cardioversion was performed with two self adhesive, preapplied, low impedance, disposable patch electrodes (R2 Medical System Inc, Carlsbad, California, USA). The pads used in this study consisted of foil electrodes covered by stannous chloride pre-gelled pads as the interface between the electrode and the chest wall. The backing was non-conductive and had an adhesive outer ring.5 The dc shocks, which had a damped sinusoidal waveform, were delivered by Lifepack 9 defibrillator (Physio Control Corporation, Redmond, Washington, USA) with the patient in full expiratory phase of ventilation.11 The time interval between each shock was at least three minutes.12

The patients were randomly assigned to undergo external cardioversion using either anterolateral defibrillator paddle position (ventricular apex–right infraclavicular area) (group AL; n = 151) or modified anteroposterior position (right sternal body at the third intercostal space–angle of the left scapula) (group AP; n = 150) (fig1).

Figure 1

Electrode positions: anterolateral = ventricular apex–right infraclavicular area paddle position; (modified) anteroposterior = right sternal body at the third intercostal space–angle of the left scapula paddle position. Front, front view; rear, rear view.

A step up protocol was defined, delivering in a synchronised fashion first a 3 J/kg body weight dc shock, then a 4 J/kg shock (maximum 360 J), and finally a second 4 J/kg shock, relocating the electrode in the alternative position. The end point of the protocol was either the achievement of technical success or the delivery of three shocks. Technical success was defined as interruption of atrial fibrillation for > 10 seconds.

All patients with an arrhythmia duration of > 72 hours received anticoagulation with warfarin for at least three weeks before attempted cardioversion and continued for at least four weeks after restoration of sinus rhythm.13 Pharmacological antiarrhythmic treatment was not randomised, and some physicians began this treatment before cardioversion while others started immediately after cardioversion. All antiarrhythmic treatment before and during cardioversion was given in the steady state condition.

Data are expressed as mean (SD). Continuous variables are compared by using Student's t test for independent samples. A χ2 test was used to determine the two tailed statistical significance of associations in 2 × 2 tables. A probability (p) value < 0.05 was considered statistically significant.


The mean age of the study population was 62 (11) years (range 26 to 80 years). The two groups were well matched, with no significant differences in age, male to female ratio, body weight, left atrial dimension, left ventricular ejection fraction, number of previous atrial fibrillation episodes, arrhythmia duration, and aetiology of heart disease (table 1). The numbers of patients who were on antiarrhythmic treatment before and during external cardioversion were also equally distributed between the two groups (the total presented in table 1 does not include patients who had been started on antiarrhythmic treatment following cardioversion). There was no significant difference in the number of patients in the two groups with no identified organic heart disease.

Table 1

Patient population   

Overall, electrical cardioversion was effective in interrupting the arrhythmia in 281 of 301 patients (93.4%). There were no complications related to cardioversion and in particular all the cases of bradycardia after cardioversion were transient and none required temporary pacing.

The main results are given in fig 2. Atrial fibrillation was interrupted by low energy dc shock in 87 of 151 patients (58%) in group AL and 100 of 150 patients (67%) in group AP (NS). With high energy dc shock, the rate of conversion rose to 114 of 151 cases (76%) in group AL and 131 of 150 cases (87%) in group AP (p = 0.013). Thus 56 patients experienced dc shock with the alternative paddle site: atrial fibrillation persisted after anteroposterior dc shock in 10 of 37 patients in group AL and after anterolateral dc shock in 10 of 19 patients in group AP.

Figure 2

Success rate for each treatment group after 3 J/kg body weight dc shock, then 4 J/kg shock, and finally a second 4 J/kg shock after crossover of the electrode paddle position. Group AL, anterolateral paddle position; Group AP, anteroposterior paddle position

The mean dc shock energy requirements were lower for group AP patients (383 (235) J) than for group AL patients (451 (287) J) (p = 0.025).

Univariate analysis showed that the duration of atrial fibrillation episode was the only factor that affected both the technical success of external cardioversion (table 2) and the need to deliver the third dc shock. In 281 patients with successful cardioversion, arrhythmia duration was 80 (109) days, while in 20 patients with unsuccessful cardioversion it was 193 (229) days (p < 0.0001). In 56 patients who needed the delivery of three shocks in the attempt to stop the arrhythmia, atrial fibrillation duration was 133 (145) days, while in 245 patients in whom arrhythmia was interrupted with one or two shocks atrial fibrillation duration was 77 (115) days (p = 0.003).

Table 2

Predictors of unsuccessful external cardioversion   

When patients with arrhythmia duration longer than 180 days were compared with those with a shorter duration of arrhythmia, the conversion rate rose from 78% to 95% (29/37v 252/264; p = 0.0001). The conversion rate was higher in patients with an arrhythmia duration of less than 180 days (95%) than in those with a duration of more than 180 days (78%) (252/264 v 29/37 patients).


When the capacitor of the defibrillator or cardioverter discharges, the amount of current delivered depends on the impedance or resistance between the electrodes. The higher the impedance, the lower the current delivered.14 ,15 It is the current density that traverses the muscle of the chamber to be defibrillated which determines the success of cardioversion16 ,17; thus, to be successful, a critical muscle mass of the atria must be defibrillated. The current that flows is determined by the energy level selected by the operator and, because the maximum output of present defibrillators in Europe is limited to 360 J, the principal factor that determines the success of electrical cardioversion of atrial fibrillation is transthoracic impedance to dc discharge.18 The determinants of transthoracic impedance to dc defibrillator or cardioverter discharge mainly include the interface between the electrode and skin,5 ,14 ,19 the electrode size and placement,14 ,16 ,17 and the delivered energy.14


In the study by Kerber et al on the same self adhesive pads that we used in the present study, the transthoracic impedance was 75 (21) ohms,5 while in Ewy's study, using metal electrodes and paste (not cream or gel) between them and skin, the transthoracic impedance was significantly lower, at 56 (16) ohms.20 Despite this theoretical limitation, we obtained a very high rate of technical success using self adhesive, preapplied pads: the cumulative proportion of patients successfully converted was 93.4%. This tallies well with those obtained in other series of patients (ranging from 85% to 96%) where metal paddles were used,2 ,8 ,21 ,22 and is even better than the result obtained by Levy et al,23 who reported a cardioversion rate of only 67% using external delivery shock. Despite this apparent advantage of metal pads, the technique of metal electrode application may also have some disadvantages: the operator requires unimpeded access to the patient's chest to apply the paddle electrode, which must be well coated with a coupling agent before use. If the gel, cream, or paste used for this purpose is inadvertently spread across the chest, the current may follow this low resistance pathway rather than traversing the thorax, shunting the current away from the heart and thus leading to failure to achieve cardioversion.24 Finally, the use of metal hand held paddle electrodes limits their use to the apex-anterior position, while with self adhesive pads any other location is obtainable.


Optimal electrode position is important to defibrillate a critical mass of the atrium for cardioversion to occur. A variety of chest placements has been used to defibrillate the atria and although the standard placement is ventricular apex–right infraclavicular area, other placements have also been advocated.9 For example Lown reported that the anteroposterior paddle position was more effective for cardioversion of atrial fibrillation than the anterolateral position.1-4 In contrast, other investigators have reported that electrode position made only minor differences to cardioversion success.5 ,7 ,8 We used a modified anteroposterior paddle position, locating the front paddle just to the right (instead of the left) of the sternal body, and our data are in complete agreement with those previously published by Lownet al. The modified anteroposterior paddle location is superior to the anterolateral position not only in terms of technical success of cardioversion of atrial fibrillation (87%v 76% with high energy dc shock), but also concerning the energy requirements to achieve successful cardioversion (383 (235) v 451 (287) J). For successful cardioversion, the current vector must transverse a critical mass of atrial muscle and the anteroposterior paddle position best fulfils this criterion; the anterolateral paddle position probably does not provide the optimal current vector and flow through the atria. However, in nine patients in whom the anteroposterior paddle position was ineffective, the arrhythmia was terminated using an anterolateral location. Therefore, depending on the individual patient, one or other of these positions may turn out to be more effective, so if the initial shocks are unsuccessful in terminating the arrhythmia the operator should consider relocating the electrode and repeating the shocks.


Various studies have examined the energy required for successful dc cardioversion.1-3 ,5 ,21 ,25 ,26Cardioversion using 100 J will only be successful in about 50% of patients with atrial fibrillation,5 ,21 while this figure rises to about 75–85% with 200 J.1-3 ,21 ,25 ,26

Though it is desirable to use the lowest energy shock likely to restore sinus rhythm, in order to avoid deleterious myocardial and haemodynamic effects,27 we decided that the initial energy delivered in the present study had to be in the middle energy range (that is, 200 J for a 66 kg patient) because our aim was to reduce the total number of shocks. The next shock was a high energy shock (maximum 360 J) because it has been shown that almost all patients in whom external cardioversion can be achieved will respond to the maximum setting.1-3 If the first high energy shock was ineffective, we delivered a final high energy shock with a different electrode paddle location. This protocol allowed an increase in the overall success rate from 81% after the second shock to 93% after the third shock, an increase of 15%. However, our finding of increased efficacy with the third shock contrasts with the results of a previous study in a smaller number of patients in which the delivery of a third shock did not increase the success rate of cardioversion.22 In this latter study the electrode position was unchanged between the second and the third shock.


Of all the variables studied, duration of atrial fibrillation was the best predictor of technical failure of external cardioversion. The mean arrhythmia duration was 80 (109) days in patients with successful cardioversion, and 193 (229) days in those with unsuccessful cardioversion, while the success rate was reduced by 18% (78% v 95%) if atrial fibrillation had persisted for more than six months before cardioversion. This confirms the results of a previous report.9 It has been shown that paroxysms of atrial fibrillation lead to changes in atrial electrophysiological properties in animals28 and humans.29 Prolonged maintenance of atrial fibrillation by a fibrillatory pacemaker leads to long term shortening of the atrial refractory period in the absence of any significant effects on intra-atrial conduction velocity.28 ,30 This electrical remodelling is thought to promote the progressive persistence of this arrhythmia. It is thus possible that the modification of atrial electrophysiological parameters could make the interruption of the arrhythmia more difficult.

In contrast to a previous study,31 we failed to demonstrate any value of M mode measurement of atrial size as a predictor of unsuccessful cardioversion, suggesting that patients with larger atrial sizes (up to at least 60 mm) should still be considered for cardioversion.


This study did have a limitation. Although it was prospective and included a relatively large number of patients, the antiarrhythmic drug treatment was not randomly assigned, but prescribed individually according to the patient's history. This might have accounted for the lack of any difference in cardioversion success between patients who were on antiarrhythmic drugs at the time of cardioversion and those who were not.


The anteroposterior (right anterior, left posterior) defibrillator paddle position is superior to the anterolateral location in terms of the technical success of elective external cardioversion of stable atrial fibrillation, and allows lower dc shock energy requirements. However, in some patients the anterolateral position could also prove to be more effective. Thus if initial shocks are unsuccessful in terminating the arrhythmia, the operator should consider relocating the electrode and repeating the shocks. The high success rate cardioversion obtained using this protocol should minimise the need for internal cardioversion for converting atrial fibrillation,32limiting this procedure to those patients with atrial fibrillation who have failed conventional external cardioversion.33

The data from this study could be used when considering cardioversion in patients with long standing atrial fibrillation.