Home > Treatment for paroxysmal AF - Atrial...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Chronic Conditions (UK). Atrial Fibrillation: National Clinical Guideline for Management in Primary and Secondary Care. London: Royal College of Physicians (UK); 2006. (NICE Clinical Guidelines, No. 36.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

8Treatment for paroxysmal AF

8.1. Rhythm control for paroxysmal AF

The three main aims of treatment for paroxysmal AF are:

  1. to suppress paroxysms of AF and maintain long-term sinus rhythm
  2. to control heart rate during paroxysms of AF if they occur
  3. to prevent the complications associated with paroxysmal AF, ie stroke- and tachycardia-induced cardiomyopathy.185

This section of the guideline essentially deals with the first aim.

Many patients with paroxysmal AF can be highly symptomatic.31 Paroxysms of AF that are persistent (that is, lasting more than 7 days) can be considered for cardioversion to sinus rhythm (see Chapter 5).

However, the abolition of symptoms of paroxysmal AF does not necessarily mean abolition of the AF per se, as heart rate slowing may abolish symptoms but allow asymptomatic episodes to continue.186 In patients with symptomatic recurrences, it may be appropriate to document the frequency of arrhythmia by Holter monitoring or event recording (see section 4.4 above). If symptoms are abolished by therapy, repeat monitoring to ascertain whether asymptomatic episodes of AF are present is occasionally performed.

If attacks of paroxysmal AF are frequent, in current clinical practice chronic prophylaxis is usually used with drugs to reduce the frequency of paroxysms after removal of precipitating factors such as caffeine, alcohol, stress, and adequate treatment of underlying diseases such as myocardial ischaemia, thyrotoxicosis and heart failure.185 If episodes of paroxysmal AF are infrequent, a pill-in-the-pocket approach can be considered (see section 8.2).

In the long term, few patients achieve complete suppression of paroxysms of AF. In clinical practice, clinicians commonly use beta-blockers or low-dose sotalol as first-line drugs, and if the patient is still getting symptomatic paroxysms, Class Ic and III drugs are used, depending on associated comorbidity and structural heart disease.

In UK clinical practice, the drugs commonly used for paroxysmal AF are beta-blockers, Class Ic agents (flecainide, propafenone) and Class III agents (sotalol, amiodarone). This section addresses the comparative efficacy of these agents.

8.1.1. Methodological introduction

Studies were considered for inclusion in this report if results were reported for a population or subpopulation with paroxysmal AF. Studies were excluded if the treatment drugs were not prescribed as regular medication for paroxysmal AF.

Nine studies were appraised, two187,188 were entirely comprised of paroxysmal AF patients, although the follow-up times were relatively short (12 months or less) compared with other studies in a general AF population. The remaining studies were undertaken in general AF populations and either reported the results for those with paroxysmal AF separately, or reported no significant interaction between AF type and drug efficacy, hence allowing extrapolation of the results to those with paroxysmal AF only.

All of the studies were either single-blinded or double-blinded RCTs with two active treatment arms. Studies that made comparisons with placebo or digoxin were not considered. All of the studies excluded patients with moderate-to-severe heart failure (typically defined as an NYHA grade greater than II).

One study based in Canada compared the costs of amiodarone treatment with propafenone/sotalol treatment.189 The cost analysis was based on 1-year follow-up data of 392 patients randomised to low-dose amiodarone (200 mg/day) or alternative first-line therapy (sotalol or propafenone) in a multicentre open label trial.112

8.1.2. Evidence statements

Propafenone versus sotalol

One study187 found no significant difference in the proportion of patients experiencing a reduction in recurrent AF episodes between propafenone and sotalol (80 to 160 mg bid) over a minimum of 3 months. (1++)

One study136 showed a significant difference in terms of the proportion of patients maintaining sinus rhythm between propafenone and sotalol (up to 480 mg/day) in favour of propafenone after 25 months of administration (1+). However, over shorter periods two other studies134,135 found no significant difference. (1++)

One study187 involving both persistent and paroxysmal AF participants found no significant difference in the rates of either intolerable or tolerable side effects between sotalol (80 to 160 mg bid) and propafenone. (1++)

Amiodarone versus sotalol

Two studies involving subgroups of patients with paroxysmal AF132,133 found amiodarone to be significantly associated with a lower rate of AF recurrence133 and a higher prevalence of sinus rhythm132 at 12 months and at 24 months compared to sotalol (160 to 480 mg/day132 or 160 to 320 mg/day133). (1+)

Beta-blockers versus sotalol

One small, open-label crossover study188 found no significant difference between atenolol and sotalol (80 mg bid) in the frequency of recurrent episodes of AF or the average duration of recurrent episodes. (1+)

Propafenone versus amiodarone

One study,112 in a population of both persistent and paroxysmal AF patients, found amiodarone to be significantly associated with a higher prevalence of sinus rhythm at a mean follow-up of approximately 15.6 months compared with propafenone or sotalol (69% versus 39%, respectively; p<0.001) (1+). The same study also reported comparable efficacy between propafenone and sotalol, as well as comparable outcomes in each arm of the study for paroxysmal and persistent AF patients (1+). In another study137 involving a subgroup of 92 patients with paroxysmal AF, 40% of those treated with amiodarone had a recurrence of AF at 24 months, compared with 83% of those treated with propafenone (p=0.19) (1+). When the composite endpoint of either AF recurrence or occurrence of side effects was considered, the figures were 58% versus 53%, respectively (p=0.51). (1+)

One study189 found that the average cost per patient for AF-related procedures and hospitalisations was significantly lower in the amiodarone group than the sotalol/propafenone group (US$532 versus US$898, p=0.03). This cost advantage increased when the costs of hospitalisations associated with strokes were included (US$541 versus US$947, p=0.02). However, the total mean hospital costs (US$1,854 versus US$1,834, p=0.96) and the total mean hospital costs plus physician costs (US$2,586 versus US$2,482, p=0.83) were similar in both groups.

8.1.3. From evidence to recommendations

Much of clinical practice in the management of paroxysmal AF has been based on relatively small trials of short duration in (mainly) symptomatic subjects. Most pharmacological studies of paroxysmal AF have concentrated on the reduction of symptomatic recurrences of paroxysmal AF, despite many having asymptomatic episodes. Many studies have also involved mixed populations with paroxysmal AF, paroxysmal atrial flutter or persistent AF.

Those patients who are asymptomatic or have rare paroxysms (eg only a few paroxysms a year) may decide not to take routine medication or to use a pill-in-the-pocket strategy, and the patient’s views need to be considered. Similarly, where paroxysms are induced by precipitants (eg caffeine or alcohol), these need to be avoided by the patient, in which case treatment may not be necessary or a pill-in-the-pocket strategy could be adopted.

An escalating approach to drug therapy in paroxysmal AF could be recommended depending on the frequency of a patient’s paroxysms and the need for increasingly effective antiarrhythmic drugs. For example, a beta-blocker could be prescribed in the first instance and in patients where this did not adequately control the paroxysms a Class Ic drug could be administered followed by amiodarone as a third-line option. This escalation of drug use is in the opposite direction of the evidence on efficacy alone as this did not fully cover the issue of the side effects (especially long-term) of the more efficacious agents.

Propafenone appeared to be at least as effective as sotalol in preventing the recurrence of AF for up to 12 months following administration,134,187 although for longer periods propafenone was more effective.136 The two drugs were comparable in terms of side effects.187 It was noted that Class Ic agents (propafenone and flecainide) should be used with caution in patients with structural heart disease or coronary artery disease.

Amiodarone was more effective than sotalol132,133 and propafenone112 in the prevention of recurrent AF. Due to concerns regarding contraindications of Class Ic agents in patients with left ventricular dysfunction, amiodarone was regarded as the drug of choice in these patients with symptomatic paroxysms despite initial beta-blocker therapy.

The concerns over the long-term toxicity of amiodarone were not addressed in the evidence. Although the clinical evidence demonstrated that amiodarone was the most effective drug, its long-term use in patients with infrequent paroxysms needed to be fully weighed against the risk of side effects, especially since some (eg lung fibrosis) could be serious.

The advice given in the British National Formulary suggested that Class I and III agents should be administered under a hospital physician, but it was stressed that this did not necessarily mean that the patient had to be hospitalised and it was the decision to administer the drug that required the necessary expertise.

The evidence only addressed the suppression of paroxysms and did not cover the control of heart rate or the prevention of complications from paroxysmal AF (eg stroke). It was considered important that patients have their symptoms and medication reviewed regularly to determine whether a patient:

  • had any side effects
  • had a continued need for treatment
  • could have their medication dose reduced.

It was also considered important that patients should not be left on medication long-term without review.

In terms of the cost of AF, amiodarone was cheaper than sotalol or propafenone, although this was based on a one-year follow-up study and costs of side effects may not have been adequately included in the analysis.



Where patients have infrequent paroxysms and few symptoms, or where symptoms are induced by known precipitants (such as alcohol, caffeine), a ‘no drug treatment’ strategy or a pill-in-the-pocket strategy should be considered and discussed with the patient. [D(GPP)]


In patients with symptomatic paroxysms (with or without structural heart disease,* including coronary artery disease) a standard beta-blocker should be the initial treatment option. [D(GPP)]


In patients with paroxysmal AF and no structural heart disease:*

  • where symptomatic suppression is not achieved with standard beta-blockers, either
    • – a Class Ic agent (such as flecainide or propafenone), or [D(GPP)]
    • sotalol** [D(GPP)]
    should be given.
  • where symptomatic suppression is not achieved with standard beta-blockers, Class Ic agents or sotalol, either
    • – amiodarone, or [B]
    • – referral for non-pharmacological intervention (see section 12.3) [A]
    should be considered.

In patients with paroxysmal AF and coronary artery disease:

  • where standard beta-blockers do not achieve symptomatic suppression, sotalol should be given* [D(GPP)]
  • where neither standard beta-blockers nor sotalol achieve symptomatic suppression, either
    • – amiodarone, or [B]
    • – referral for non-pharmacological intervention (see section 12.3) [A]
    should be considered.

In patients with paroxysmal AF with poor left ventricular function:

  • where standard beta-blockers are given as part of the routine management strategy and adequately suppress paroxysms, no further treatment for paroxysms is needed [D(GPP)]
  • where standard beta-blockers do not adequately suppress paroxysms, either
    • – amiodarone or [B]
    • – referral for non-pharmacological intervention (see section 12.3) should be considered. [A]
    should be considered.

Patients on long-term medication for paroxysmal AF should be kept under review to assess the need for continued treatment and the development of any adverse effects. [D(GPP)]

8.2. Treatment strategy for paroxysmal AF

In selected patients with recurrent paroxysmal AF, out-of-hospital initiation of antiarrhythmic drugs may be possible, allowing for earlier treatment, a shorter duration of AF and a presumed likelihood of restoring and maintaining sinus rhythm. A pill-in-the-pocket approach is used in those not taking drugs regularly due to infrequent symptoms/paroxysms, or can be taken as an ‘extra’ drug dose in those already on a low maintenance of that particular drug. This approach is different to the out-of-hospital use of antiarrhythmic drugs in patients with recurrent persistent AF, where the aim may be to achieve pharmacological cardioversion per se or to improve the likelihood of subsequent elective electrical cardioversion (see Chapter 5), or to maintain sinus rhythm (see section 6.1).

The main concern with a pill-in-the-pocket approach is the risk of proarrhythmia often associated with antiarrhythmic drugs. Thus, the pill-in-the-pocket approach has generally been advocated only in those patients with a low risk of proarrhythmia and other adverse side effects. Such patients are typically those with no structural heart disease, absence of heart failure or left ventricular dysfunction, and where there is evidence that the antiarrhythmic drug used has previously worked successfully with no adverse effects (eg after at least one inpatient trial of the drug administered as a single oral dose, under ECG monitoring).

The antiarrhythmic drugs amiodarone and propafenone have both been considered in a number of trials comparing the safety and efficacy of a single oral dose of the drug with the intravenous administration of the same drug.82–85 In all of these trials, patients were selected on the basis of relatively young age and the absence of any severe underlying structural heart disease. There was no incidence of ventricular proarrhythmia reported in either the intravenous or oral administration arms of these trials, and the incidence of successful cardioversion within 8 or 24 hours was comparable in most cases.

The objective of this section is to determine in which patients a single oral-dose antiarrhythmic drug may be safely used as a pill-in-the-pocket approach.

8.2.1. Methodological introduction

Studies were included if a comparison was made in terms of the safety, efficacy and impact on healthcare resources between the out-of-hospital self-administration of pharmacological cardioversion and the supervised, in-hospital administration in a well-defined patient cohort with either AF or supraventricular tachycardia (SVT). Studies were not included if the pharmacological agents were administered prophylactically.

Both of the included studies were based in Italy and compared the rates of hospital admission and emergency room treatment in a single cohort of patients in the period before and the period after the self-administration of antiarrhythmic drugs for the termination of either paroxysmal AF190 or paroxysmal SVT.191

Neither study specified the treatment protocol during the period before the self-administration of antiarrhythmic drugs, or made a comparison between the two periods in terms of safety and efficacy.

8.2.2. Evidence statements

One study190 found that the average number of admissions per month for emergency treatment was significantly lower during treatment of paroxysmal AF using a pill-in-the-pocket approach with Class Ic drugs than during conventional treatment (4.9 versus 45.6, p<0.001) in a population of patients with the following criteria: (2+)

Another study found a similar result using similar selection criteria among patients with paroxysmal SVT.191 (2+)

8.2.3. From evidence to recommendations

The limited evidence suggested that pill-in-the-pocket treatment was associated with a lower incidence of inpatient and emergency hospital admissions than conventional treatment.190,191 It was uncertain whether the pill-in-the-pocket strategy was associated with more adverse events, or reduced episode duration when compared to in-hospital treatment.

Within the UK, the number of patients managed in this way is currently thought to be small, and patients need to be made more aware of this treatment option, although strict selection criteria are deemed necessary. In particular, it was considered that patient education in its use is vital.

Therapy for paroxysmal AF should be tailored to the patient. For example, episodes of AF for 1 to 2 minutes once a year or for 10 hours twice a day are both paroxysmal AF, but their impact on the patient’s quality of life, if symptomatic, would be quite different. In patients with infrequent and brief paroxysms, the regular use of antiarrhythmic therapy may not be necessary (and is commonly not prescribed in current clinical practice). Such patients may be suitable for the pill-in-the-pocket approach. However, for infrequent but protracted and symptomatic paroxysmal AF, rapid cardioversion of each event and/or antiarrhythmic drug prophylaxis may be considered.



In patients with paroxysmal AF, a pill-in-the-pocket strategy should be considered in those who:

8.3. Antithrombotic therapy for paroxysmal AF

Patients with paroxysmal AF appear to carry the same risk of stroke and thromboembolism as those with persistent AF.192 Whether the risk is dependent upon the frequency of the paroxysms or their duration is unclear.193 In addition, patients frequently get asymptomatic paroxysms31 and may still be at risk of thromboembolism. There is evidence of some clustering of thromboembolic events around the time of onset of AF167,194 and following successful cardioversion.139 Compared with permanent AF, patients with paroxysmal AF tend to be younger18 and have a lower prevalence of associated clinical risk factors.

8.3.1. Methodological introduction

No studies were found that considered the efficacy of antithrombotic therapy compared with placebo in a population comprising only paroxysmal AF patients.

However, subgroups of patients with paroxysmal AF were reported in two meta-analyses. One meta-analysis192 pooled data from the aspirin treatment arms of three serial trials and compared ischaemic stroke rates in the paroxysmal AF subgroup with those of non-paroxysmal AF patients. The second study174 compared the effects of anticoagulation and antiplatelet therapy on rates of ischaemic stroke and major bleeding in a subgroup of participants with paroxysmal AF.

8.3.2. Evidence statements

One meta-analysis174 found that in patients treated with aspirin, the incidence of stroke was similar in those with paroxysmal AF and those with non-paroxysmal AF. (2++)

One meta-analysis192 found adjusted-dose warfarin to be associated with a reduced incidence of ischaemic stroke compared with aspirin (1.5% versus 4.7%, respectively, p<0.05), although no significant difference was found in terms of the incidence of bleeding events. The study also found the efficacy of warfarin in reducing the incidence of stroke to be comparable between those with paroxysmal and those with non-paroxysmal AF. (1+)

8.3.3. From evidence to recommendations

The data from clinical trials were limited by the relatively small number of patients with paroxysmal AF (approximately 12% of subjects in five randomised trials).192 There was some evidence that patients with paroxysmal AF receiving aspirin experience a similar rate of ischaemic stroke as patients with non-paroxysmal AF who receive aspirin.192

The effectiveness of anticoagulation compared with antiplatelet therapy in reducing the rate of ischaemic stroke is similar for patients with either paroxysmal AF or non-paroxysmal AF.174 The rate of adverse bleeding events associated with anticoagulation is also similar between the two groups.174

In current clinical practice, the indicators for antithrombotic therapy are highly variable (often depending on the clinical presentation of symptoms). For example, the more frequent a patient’s paroxysms the more likely they will be prescribed anticoagulants, despite the risk of stroke being determined by the presence of associated risk factors. It was concluded that patients with paroxysmal AF required a similar degree of anticoagulation as patients with permanent AF, but that further studies would be required to confirm this.



Decisions on the need for antithrombotic therapy in patients with paroxysmal AF should not be based on the frequency or duration of paroxysms (symptomatic or asymptomatic) but on appropriate risk stratification, as for permanent AF (see section 11.6). [B]

8.4. Rhythm-control treatment algorithm for paroxysmal AF

Flowchart Icon

Figure 8.1. Rhythm-control treatment algorithm for paroxysmal AF (PDF, 56K)



Progressively titrated from 80 mg twice daily up to 240 mg twice daily.

Copyright © 2006, Royal College of Physicians of London.

All rights reserved. No part of this publication may be reproduced in any form (including photocopying or storing it in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright owner. Applications for the copyright owner’s written permission to reproduce any part of this publication should be addressed to the publisher.

Cover of Atrial Fibrillation
Atrial Fibrillation: National Clinical Guideline for Management in Primary and Secondary Care.
NICE Clinical Guidelines, No. 36.
National Collaborating Centre for Chronic Conditions (UK).

NICE (National Institute for Health and Care Excellence)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...