Article Text

Download PDFPDF

Heart rhythm disorders and pacemakers
Biphasic energy selection for transthoracic cardioversion of atrial fibrillation. The BEST AF Trial
  1. B M Glover1,2,
  2. S J Walsh1,
  3. C J McCann1,2,
  4. M J Moore1,2,
  5. G Manoharan1,2,
  6. G W N Dalzell1,
  7. A McAllister3,
  8. B McClements3,
  9. D J McEneaney4,
  10. T G Trouton5,
  11. T P Mathew6,
  12. A A J Adgey1,2
  1. 1
    Regional Medical Cardiology Centre, Royal Victoria Hospital, Belfast, Northern Ireland, UK
  2. 2
    Queens University of Belfast, Belfast, Northern Ireland, UK
  3. 3
    Department of Cardiology, Mater Hospital, Belfast, Northern Ireland, UK
  4. 4
    Craigavon Cardiac Centre, Craigavon Area Hospital, Craigavon, Northern Ireland, UK
  5. 5
    Department of Cardiology, Antrim Area Hospital, Antrim, Northern Ireland, UK
  6. 6
    Department of Cardiology, Nottingham City Hospital, Hucknall Road, Nottingham, UK
  1. Professor A A J Adgey, Regional Medical Cardiology Centre, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA, Northern Ireland, UK; jennifer.adgey{at}


Aims: To compare the efficacy and safety of an escalating energy protocol with a non-escalating energy protocol using an impedance compensated biphasic defibrillator for direct current cardioversion of atrial fibrillation (AF).

Methods and results: This prospective multicentre randomised trial enrolled 380 patients (248 male, mean (SD) age 67 (10) years) with AF. Patients were randomised to either an escalating energy protocol (protocol A: 100 J, 150 J, 200 J, 200 J), or a non-escalating energy protocol (protocol B: 200 J, 200 J, 200 J). Cardioversion was performed using an impedance compensated biphasic waveform. First-shock success was significantly higher for those randomised to 200 J than 100 J (71% vs 48%; p<0.01) and for patients with a body mass index (BMI) >25 kg/m2 (75% vs 44%; p = 0.01). In patients with a normal BMI there was no significant difference in first-shock success. There was also no significant difference between subsequent shocks or overall success. The use of a non-escalating protocol (protocol B) resulted in fewer shocks but with a higher cumulative energy. There was no difference in duration of procedure, amount of sedation administered or post-shock erythema between the groups.

Conclusion: First-shock success was significantly higher, particularly in patients with a BMI >25 kg/m2, when a non-escalating initial 200 J energy was selected. The overall success, duration of procedure and amount of sedation administered, however, did not differ significantly between the two protocols.

View Full Text

Statistics from

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in clinical practice, with a recently reported prevalence of 5.5%.1 Although several studies have reported no advantages in a rhythm control strategy over rate control,2 3 restoration of sinus rhythm remains an important goal, particularly in younger and symptomatic patients.4

Transthoracic direct current cardioversion (DCCV) with a biphasic waveform is more efficacious than monophasic at lower energy levels.58 Furthermore, fewer shocks are required, thus potentially reducing procedure times and requirements for intravenous sedation. A lower incidence of skin burns6 and less skeletal muscle damage8 have also been reported.

There are two approaches to energy selection for DCCV. After developing the technique of cardioversion, Lown favoured the escalating energy approach. The rationale for this was to minimise post-shock arrhythmia.9 This approach also allows cardioversion at the lowest energy for each individual patient and may prevent high cumulative doses in some.

Current guidelines for the management of AF, state that the initial energy should be 200 J when using monophasic waveforms,10 although an initial energy of 360 J has been shown to be more efficient, particularly in cardioverting AF of a longer duration.11 It is also suggested that an initial energy of 200 J should be considered when using biphasic waveforms for AF of longer duration.10

Importantly, there is no evidence that the use of higher energy levels results in either arrhythmic complications12 or a rise in cardiac troponin13 as an indicator of myocardial cellular injury.

Some practitioners, however, prefer to begin the procedure at the highest energy in order to minimise the total number of shocks delivered and the duration of the procedure (including exposure to sedation or anaesthesia) for the majority of patients.14

Since there is no overall consensus about initial energy selection and the use of biphasic waveforms for the DCCV of AF, a considerable variation in clinical practice exists.

We therefore compared an escalating energy protocol (starting at 100 J) with a non-escalating energy protocol (200 J) to establish the efficacy, safety and clinical implications of both practices for the biphasic DCCV of AF using an impedance-compensated biphasic waveform delivered by a Heartstream XL Defibrillator (Philips Medical Systems, Andover, MA, USA).


Patients with a history of persistent AF who were routinely referred for an outpatient DC cardioversion of AF were invited to take part. They were provided with oral and written information. Patients were recruited in the Royal Victoria Hospital, Mater Infirmorum Hospital (both Belfast), Antrim Area Hospital and Craigavon Area Hospital in Northern Ireland. Ethical approval was obtained from the Queen’s University of Belfast Research Ethics Committee and the study was compliant with the Declaration of Helsinki. Indemnity was obtained in each hospital trust before the start of the study.

Exclusion criteria were inadequate anticoagulation (for 4 weeks before the procedure), known intracardiac thrombus, significant electrolyte imbalance, digoxin toxicity, cardiogenic shock, refractory pulmonary oedema, uncontrolled thyrotoxicosis, pregnancy or women of childbearing potential not using reliable contraceptives, a permanent pacemaker which would result in an altered pad position, any mental condition rendering the subject incapable of giving informed consent or any other condition thought to increase the risk of the procedure.


Patients fasted overnight. All patients were fully anticoagulated with warfarin for a minimum period of 4 weeks before the DCCV. Informed consent was obtained. Data were recorded from the patients, including age, gender, body mass index (BMI), AF duration, aetiology of AF and routine drug treatment. The duration of AF was recorded, based on a combination of the patient’s history and electrocardiograms, where available. Left atrial diameter, left ventricular function and dimensions and the presence or absence of valvular disease were noted for all patients who had undergone a transthoracic echocardiogram within the previous 12 months. Patients were allocated to either one of two possible energy protocols by random number generation. The operator of the defibrillator could not be blinded to the energy protocol.

All shocks were synchronised to the “R” wave of the surface electrocardiogram. A minimum of 1 minute was allowed between each shock. Success was defined as the return of sinus rhythm for at least 30 seconds.

Self-adhesive pads (Agilent Technologies, CA, USA) were placed in the right infraclavicular and apical positions. This position is the most widely used in clinical practice and has been previously shown to be as effective as anteroposterior pad positions when using an impedance-compensated biphasic waveform.15 The transthoracic impedance for each shock was recorded by the defibrillator.

Sedation was given as either intravenous midazolam or intravenous propofol according to standard practice in the enrolling centre.

For calculation of the duration of the procedure the beginning of the procedure was defined as the placement of the pads on the patient’s chest. The end of procedure was defined as either the return of sinus rhythm for at least 30 seconds, or the persistence of AF at the end of the study protocol.

Rhythm strips for all shocks were kept for analysis to verify the result, and to record any post-shock arrhythmias. For successful shocks the duration of any sinus pauses, ventricular premature complexes, ST-segment deviation, recurrence of AF and significant atrioventricular block were recorded.


We hypothesised that starting at a higher initial shock energy would result in a higher success rate with fewer shocks for the DCCV of persistent AF. We further postulated that this higher initial energy protocol might reduce the duration of the procedure and the amount of intravenous sedation required. We calculated that in order to detect a difference in the first-shock success of 15% between both protocols, 187 patients would be required in each group. This was calculated based on a power of 85% and using a significance level of 0.05. The first-shock success was assumed to be 60% as reported by Page et al6 for biphasic shocks at an energy level of 100 J.

Statistical analysis was performed with SPSS version 11 (SPSS, Chicago, IL, USA). Baseline characteristics were compared using the χ2 test or Fisher exact test. Non-parametric testing was performed using the Mann–Whitney U test. Both groups were compared for confounding variables and the overall success was compared using the χ2 test. Confidence intervals were calculated for the overall success for both protocols using Excel software (Excel, Microsoft Corporation, USA). A p value of <0.05 was considered significant.


Baseline characteristics

Three hundred and eighty patients were recruited for this study. Of these, 193 patients were randomised to protocol A, and 187 patients were randomised to protocol B. Of all the patients recruited the mean (SD) age was 67 (10) years, 248 (65%) were male, the median duration of AF was 6 (interquartile range 3–11) months and the mean (SD) BMI was 28 (5) kg/m2. Table 1 summarises the baseline characteristics of each group.

Table 1 Baseline characteristics

No statistically significant difference in the demographic variables was found between the two groups. The commonest cause of AF was hypertension, followed by coronary artery disease.

Echocardiographic data were available for 214/380 (56%) patients. The mean (SD) left atrial diameter for all patients was 46 (7) mm (protocol A 47 (7) mm, protocol B 46 (7) mm, p = 0.19). There were no significant differences between the two groups in the mean (SD) left ventricular ejection fraction (protocol A 52 (27), protocol B 49 (29), p = 0.30). Presence of mitral valve disease (protocol A 14/104 (13%), protocol B 15/110 (14%), p = 0.78) and aortic valve disease (protocol A 6/104 (6%), protocol B 5/110 (5%), p = 0.64) were uncommon.

Shock success

For patients randomised to protocol A (100 J/150 J/200 J/200 J) the success was 92/193 (48%) after the first shock, 148/193 (77%) after the second shock, 169/193 (88%) after the third shock and 174/193 (90%) after the fourth shock. For those randomised to protocol B (200 J/200 J/200 J) the success was 132/187 (71%) after the first shock, 155/187 (83%) after the second shock and 165/187 (88%) after the third shock. There was a significantly higher number of successes after the first shock for those randomised to an initial higher energy (48% A vs 71% B; p<0.01). For subsequent shocks there was no statistically significant difference in the success rates (fig 1). No cases of immediate recurrence of AF were seen after any shock in both groups within the first 30 seconds (time considered as success).

Figure 1 Shock energy protocols used and successes observed.

A difference was observed between the two protocols in patients with a normal body mass index (BMI ⩽25 kg/m2) and those who were overweight or obese (BMI>25 kg/m2).

First-shock success did not differ significantly between protocol A and protocol B in patients with a normal BMI (protocol A 27/46 (59%); protocol B 17/34 (50%)). In patients who were overweight or obese first-shock success was significantly higher for protocol B versus protocol A (protocol A 65/147 (44%); protocol B 115/153 (75%)). Logistic regression performed using the protocol and BMI as categorical variables, confirmed that the difference in success rates between the two protocols depended on the patients’ BMI (p = 0.001 for test of interaction).

First-shock success was found to be higher for protocol B in patients with an initial transthoracic impedance ⩾79 Ω compared with protocol A (protocol A 60/141 (43%); protocol B 97/140 (69%), p<0.001). There was no significant difference between the success of subsequent shocks or the overall success in these patients with a high transthoracic impedance. For patients with an initial transthoracic impedance <79 Ω there was no difference in shock efficacy between either protocol.

Overall success showed no significant difference, with 90% of those in protocol A and 88% of those in protocol B achieving sinus rhythm by the end of the procedure (p = 0.56, CI 4.4 to 8.2). Figure 1 shows a summary of the success rates for each shock. When multivariable analysis was used looking at all the baseline characteristics the only factor found to influence overall shock success (in both groups) was the duration of AF (p = 0.04). No significant difference was found between the two groups for the first-shock transthoracic impedance (mean (SD): protocol A 89 (19), protocol B 88 (17), p = 0.08) or subsequent change in impedance between shocks.

Number of shocks, cumulative energy, sedation and duration of procedure

Fewer shocks were required for those patients randomised to an initial energy of 200 J (protocol B 1.46 (0.76) vs protocol A 1.88 (1.04), p<0.01). However, despite fewer shocks the cumulative energy for the procedure was significantly higher using the initial higher-energy protocol compared with the escalating energy protocol (protocol A 202 (135) vs protocol B 251 (110), p<0.01).

The mean (SD) duration for the procedure (protocol A 8 (4) minutes, protocol B 7 (4) minutes) and mean (SD) dose of intravenous sedation (protocol A 7 (3) (midazolam), 94 (36) (propofol); protocol B 7 (3) (midazolam), 94 (38) (propofol)) were not significantly different between the two groups.

Post-shock arrhythmias

Episodes of post-shock arrhythmias were recorded on a rhythm strip and those following a successful shock (339) were analysed. There was no significant difference between the two groups in the mean (SD) sinus pause duration (protocol A 1.87 (0.94) seconds, protocol B 1.70 (0.87) seconds). Only three shocks resulted in a sinus pause greater than 5 seconds; two of these followed a single 200 J shock, and one followed a single 100 J shock. There was no haemodynamic compromise with any of these episodes and pacing was not required for any patient. Transient ST-segment elevation was noted after successful cardioversion in 21/174 (12%) of patients in protocol A versus 16/165 (10%) in protocol B (p = 0.35).

Ventricular premature complexes were noted in 157/339 (46%) cases. Although these complexes were more common in protocol B than protocol A, this difference was not significant (p = 0.12). No cases of second- or third-degree atrioventricular block were seen.


The use of an escalating energy protocol has been regarded as conventional practice for the DCCV of AF with monophasic waveforms. The original rationale for this was proposed by Lown in order to minimise post-shock arrhythmias.9

It has been well established that biphasic waveforms are more successful in the cardioversion of AF than monophasic waveforms with higher efficacy at lower energies. Additionally biphasic waveforms result in fewer post-shock arrhythmias, fewer skin burns6 and a shorter period of myocardial stunning.16

Our results demonstrate that there is a significant increase in the first-shock success for the biphasic transthoracic DCCV of AF if the initial energy selected is 200 J rather than 100 J. There was no significant difference in the success of subsequent shocks or overall success between either protocol. In patients who were overweight or obese (BMI >25 kg/m2) first-shock success was significantly greater if a higher-energy shock was selected. In patients with a normal or low BMI there was no difference in the first-shock success regardless of whether 100 J or 200 J was used. Obesity has been shown to reduce the success of transthoracic DCCV of AF.17 Although transthoracic impedance did not directly influence shock success, patients with a transthoracic impedance ⩾79 Ω had a higher first-shock success if the initial shock was 200 J rather than 100 J. Again there was no significant difference in success between subsequent shocks.

A previous study has shown that higher-energy shocks are more successful than lower-energy shocks in patients with a raised transthoracic impedance18 and can easily be measured before delivery of the shock. It has been shown that about 4% of current, delivered by a transthoracic defibrillator in the right infraclavicular to apical configuration, actually reaches the heart.19 For DCCV to be successful a shock field gradient of approximately 5 V/cm is required.20 Despite the use of impedance-compensated waveforms, the amount of energy which traverses the atria is probably lower in patients with a high BMI than in those with a normal BMI. This explains the benefit of an initial higher-energy protocol in these patients. Of interest, subsequent shock and overall shock success were not significantly different.

The baseline characteristic in this study found to influence overall shock success was the duration of AF. This has previously been shown in a study looking at the predictors of success of DCCV of atrial fibrillation.21

Starting at 200 J (protocol B) resulted in significantly fewer shocks for successful cardioversion but a higher cumulative energy compared with starting at 100 J (protocol A). The number of shocks and cumulative energy were lower than those reported in a previous study comparing an escalating energy protocol with a high-energy protocol using monophasic waveforms.11 Protocol B was on average 1 minute shorter than protocol A. However, this did not achieve statistical significance.

Although a previous trial comparing first-shock success at various energy levels for a biphasic waveform reported higher success rates than our study (63% at 100 J, 83% at 200 J22), the definition of success was one or more sinus beats within 30 seconds of the shock. We decided that the presence of sinus rhythm for 30 seconds was more clinically applicable.

There were no significant arrhythmias using either energy protocol. There was no significant difference in the duration of post-shock sinus pauses or the incidence of transient ST elevation after a successful DCCV which is thought to represent premature repolarisation of the subepicardium and is not associated with an elevation of cardiac markers.23 No cases of second- or third-degree atrioventricular block were seen.


Fewer shocks were required for an energy selection protocol that begins at 200 J. Although this approach is safe and not associated with any additional complications, no significant difference in the duration of the procedure or amount of sedation required was demonstrated in this study. It is therefore reasonable to recommend higher biphasic energies as a first DCCV shock for patients, particularly in patients who are overweight or have a higher transthoracic impedance.


Shervin Ayati of Philips Medical provided defibrillators and pads for this study. Statistical analysis was performed by Dr Chris Patterson, Department of Statistics, Queens University of Belfast, Northern Ireland.


View Abstract


  • Competing interests: None.

  • Funding: BMG is a recipient of a grant from Northern Ireland Chest Heart and Stroke Association.

  • Ethics approval: Approval was obtained from the Queen’s University of Belfast Research Ethics Committee.

    See Editorial, p 830

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.

Linked Articles

  • Editorial
    Guy Haywood Bogdan Nuta