Article Text

Download PDFPDF

Atrial fibrillation
The management of atrial fibrillation after cardiac surgery
  1. Robert W Rho
  1. Robert W Rho, MD, Associate Professor, Director, Cardiac Electrophysiology Catheter Ablation Program, University of Washington School of Medicine, HSB AA 121C, 1959 NE Pacific Street, Seattle, WA 98195, USA; rrho{at}

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.

Atrial fibrillation (AF) is the most common arrhythmia complicating cardiac surgery. Patients who develop AF following cardiac surgery are at increased risk of stroke, congestive heart failure, and haemodynamic instability. Postoperative AF is invariably associated with increased intensive care unit stays and prolonged hospitalisations and is responsible for significant patient morbidity and health care expenditures. Management of AF after cardiac surgery can be guided by knowledge of what is known about clinical postoperative AF and the evidence available on the efficacy of treatment strategies. The following is a concise review of postoperative AF as well as an evidence based approach to the management of AF after cardiac surgery.


Postoperative AF occurs in approximately 30–40% of patients undergoing coronary artery bypass grafting (CABG) surgery and in as many as 64% of patients with concomitant valve surgery. In a prospective observational study of 4657 patients undergoing CABG surgery at 70 centres, the peak incidence of AF occurred on postoperative day 2 and 3. Over 60% of AF episodes occurred by day 2. Fifty-seven per cent of patients had only one episode of AF during their hospitalisation.1


Several clinical factors have been implicated as risk factors for postoperative AF (box 1). These risk factors include advanced age, concomitant valve surgery, a prior history of AF, congestive heart failure, left atrial enlargement, decreased left ventricular function, postoperative withdrawal of β-adrenergic blockade, chronic obstructive pulmonary disease, and P wave duration on a 12 lead ECG. Among these risk factors, age is the most consistently reproduced independent risk factor for postoperative AF. In a study of patients undergoing CABG, Zaman et al reported that only 3.4% of patients aged 50–54 years developed postoperative AF, compared to 42.2% of patients aged 70–74 years. The odds of developing AF increased 1.48-fold for each 5 year increase in age.2 Similarly, Matthew et al reported that for every 10 year increase in age, there was a 75% increase in the odds of developing AF. Among patients over the age of 70, 47% had postoperative AF.1 With the ageing of the population and the increased number of elderly patients undergoing cardiovascular surgery, it is likely that postoperative AF will remain a frequently encountered arrhythmia.

Box 1 Clinical risk factors for postoperative atrial fibrillation

  • Advanced age

  • Male sex

  • Concomitant valve surgery

  • History of atrial fibrillation

  • Left ventricular dysfunction

  • Chronic obstructive pulmonary disease

  • Withdrawal of β-adrenergic blocking agents

  • P duration >116 ms


AF results from multiple wavelets of re-entry initiated by focal triggers (atrial tachycardia or atrial premature beats) that may occur frequently in the postoperative period. These multiple wavelets of re-entry rapidly and irregularly bombard the atrioventricular (AV) node, which may conduct rapidly to the His/Purkinje system. Rapid irregular contraction of the ventricles may result in congestive heart failure, hypotension, and ischaemia.

In contrast to AF encountered outside the post-surgical period, multiple acute transient factors contribute to the development of AF after cardiac surgery. An increase in circulating catecholamines, heightened sympathetic and parasympathetic tone, atrial stretch, transcellular fluid and electrolyte shifts, metabolic abnormalities, inflammation, and pericarditis are among the many factors implicated in contributing to the development of AF following cardiac surgery.


In most cases the diagnosis of AF is made easily by telemetry monitoring and by 12 lead ECG. Common cardiac rhythms observed postoperatively that may obscure the diagnosis of AF include “atrial”-ventricular sequential pacing (undersensing of AF resulting in a pacing output in the atrium in the DDD mode), ventricular pacing (pacemaker in the VVI mode), and accelerated junctional rhythm. Due to the high frequency of AF following cardiac surgery and the consequences of a missed diagnosis, it is important to maintain a high level of vigilance to the cardiac rhythm, especially under these circumstances (fig 1).

Figure 1 Ventricular pacing through a permanent pacemaker on postoperative day 2 after an aortic valve replacement and coronary artery bypass grafting. This patient is in atrial fibrillation (AF) with complete heart block. Ventricular pacing can make the diagnosis of AF challenging. A high degree of vigilance to the atrial rhythm should be maintained in this setting.


The initial management of a patient who develops AF will depend on a rapid assessment of the patient’s haemodynamic status. If AF is accompanied by ischaemia, hypotension, or congestive heart failure, immediate cardioversion should be attempted. The initial shock energy should start at 300–360 J (monophasic waveform) or 200 J (biphasic waveform). Direct current (DC) cardioversion is associated with a >95% success rate in converting AF to sinus rhythm.3 It is important to differentiate failure to cardiovert and early reinitiation of AF (ERAF). In the former, DC cardioversion does not result in a single beat of sinus rhythm, and in the latter, DC cardioversion is successful but AF is immediately reinitiated (sometimes after one or two beats of sinus rhythm). This distinction is important because in the former case, the clinician may try (1) increasing the defibrillation energy and adjusting and adding pressure to the electrode pads, (2) internal cardioversion, or (3) repeating cardioversion after antiarrhythmic drug (AAD) loading. In the latter case, repeat cardioversion would be futile due to continued early reinitiation of AF. In this setting, it would be more appropriate to initiate an AAD and correct secondary factors contributing to reinitiation of AF before reattempting electrical cardioversion. Often, initiation of an antiarrhythmic agent will result in sinus rhythm in this setting and obviate the need for electrical cardioversion.

In patients with AF without haemodynamic compromise, the clinician should answer the four following questions:

  1. Is the rate adequately controlled?

  2. Should the patient be cardioverted to sinus rhythm?

  3. Should the patient be anticoagulated to lower his/her stroke risk?

  4. Should the patient be started on an anti-arrhythmic agent?

Each of these issues is discussed separately below.

Box 2 Agents used for rate control of atrial fibrillation following cardiac surgery

Intravenous (IV) doses for rate control
  • Diltiazem 0.25 mg/kg IV over 2 min

    • continuous infusion: 5–15 mg/min titrate to heart rate and blood pressure

  • Verapamil 0.075 to 0.15 mg/kg IV over 2 min

    • not recommended as a continuous infusion

  • Metoprolol 2.5 to 5 mg IV bolus over 2 min

    • can give up to three doses as blood pressure allows

    • not recommended as a continuous infusion

  • Esmolol 0.5 mg/kg over 1 min

    • continuous infusion: 0.06 to 0.2 mg/kg/min titrate to heart rate and blood pressure

  • Digoxin 0.25 mg every 2–3 h up to 1.25 mg/24 h period

    • maintenance dose 0.125 to 0.25 mg/day

  • Amiodarone 150 mg IV over 10 min

    • 0.5 to 1.0 mg/min

    • usually 1.0 mg/min IV × 6 h then 0.5 mg/min for 18 h or until patient can take oral treatment

Oral doses for rate control
  • Metoprolol tartrate

    • 50 mg to 450 mg/day orally in 2–3 divided doses

  • Diltiazem

    • 120 to 360 mg/day

    • if converting from iv drip, the oral dose (mg/day) is approximately equal to [rate (mg/hour) × 3 + 3] × 10

  • Verapamil

    • 80–480 mg/day in divided doses

    • start at 40–80 mg twice daily

  • Digoxin

    • 0.5 mg orally × 1 then 0.25 mg orally every 6 h × 2 or 3; maintain at 0.25 mg orally daily; dose adjust in renal failure and in elderly patients

Rate control

Rate control in the immediate postoperative setting may be challenging. Secondary causes of enhanced AV nodal conduction should be treated aggressively. Attention should be given to pain management, patient arousal and fear, anaemia, hypoxia, and intravascular volume status. Unless these factors are addressed in conjunction with pharmacotherapy with AV nodal blocking agents, goals for rate control may be difficult to achieve. Addressing these secondary causes of rapid ventricular rate is crucial to successful rate control. Drug treatments for rate control include non-dihydropyridine calcium channel blockers, β-adrenergic blocking agents (β-blockers, BBs), digoxin, and amiodarone (box 2).

Patients with hypotension who are unable to be cardioverted to sinus rhythm electrically may pose a significant challenge. In this setting, intravenous amiodarone and intravenous digoxin may be the drugs of choice. In a study of critically ill patients with heart rates consistently >120 beats/min, 60 patients were randomised to three groups: (1) intravenous diltiazem bolus followed by continuous infusion; (2) intravenous amiodarone bolus; and (3) intravenous amiodarone bolus followed by continuous infusion. Intravenous amiodarone bolus followed by a continuous infusion was as effective as intravenous diltiazem in achieving a 30% decrease in heart rate (70% vs 75%, p = 0.38). Patients on intravenous diltiazem were more likely to require drug discontinuation due to hypotension compared with intravenous amiodarone (30% vs 5%, p = 0.01).4


The majority of patients who develop AF will convert to sinus rhythm spontaneously within 24 h. Because of this, if the patient tolerates AF haemodynamically and the rate is well controlled, it may be prudent to allow 24 h for spontaneous cardioversion before pharmacologic or electrical intervention. During this period of time, it is important to correct all modifiable factors contributing to AF such as pain management, haemodynamic optimisation, weaning of intravenous inotropes, correcting electrolytes and metabolic abnormalities, and addressing anaemia or hypoxia. The exact duration of AF is an important consideration whenever cardioversion is considered. In general, patients who have AF for >48 h are at increased risk of stroke and should, therefore, be therapeutically anticoagulated within 48 h of AF onset. If AF has occurred for >48 h or if the exact duration of AF is unknown, a transoesophageal echocardiogram (TOE) to rule out left atrial clot should be performed before attempted cardioversion. If the patient is still in AF after 24 h, a strategy to convert the patient to sinus rhythm is usually initiated between 24–36 h and executed before 48 h to minimise the risk of stroke (fig 2).

Figure 2 (A) Risk factors used in this proposed algorithm are: age >65 years, prior history of atrial fibrillation (AF), chronic obstructive pulmonary disease, structural heart disease, and concomitant valve surgery. Although several other risk factors have been proposed, these risk factors are more consistently described and easy to identify clinically. (B) Opportunity to intervene will depend on when the patient is seen before or after surgery. Antiarrhythmic drugs (AAD) and β-blockers (BB) may be started preoperatively or postoperatively. The efficacy of oral AAD loading begun within 6 days of surgery is unknown. Efficacy of amiodarone and sotalol has been demonstrated when started postoperatively (see box 3).

Appropriate pharmacologic agents for cardioversion are limited in the population of patients following cardiac surgery because of the drug interactions and the concern about proarrhythmia associated with AADs in patients with ischaemic and structural heart disease. Furthermore, evidence of the efficacy and safety of these agents in patients after cardiovascular surgery is insufficient. In most instances, intravenous amiodarone may be the only appropriate agent, although its efficacy in acute cardioversion of AF is modest. For these reasons, the most effective means of achieving sinus rhythm in patients with continued AF is DC cardioversion. This procedure is highly efficacious and has a low rate of complications.3 Limitations to electrical cardioversion include the need for anaesthesia and the risk of skin burns.


Because postoperative AF is associated with a higher risk of stroke,5 stroke prevention strategies in this setting are extrapolated from the guidelines for managing anticoagulation in AF that is not related to cardiac surgery. It is up to the clinician to balance carefully the risk of bleeding following surgery when considering anticoagulation in patients who have not been actively cardioverted. In most cases, the risk of bleeding is minimal, and the priority should be placed on stroke prevention.

In patients who have been cardioverted within 48 h of onset of AF with no further recurrence of the arrhythmia, it is not necessary to treat with anticoagulation. In patients who have been in AF for >48 h or who have been in AF for an indefinite period of time, anticoagulation with heparin initially followed by oral anticoagulation should be initiated. Attempts at cardioversion in this setting should be performed after a TOE has ruled out the presence of a left atrial thrombus. Anticoagulation should be continued for a minimum of 3–4 weeks.

Initiation of antiarrhythmic agents

The decision to initiate AADs for secondary prevention of AF is guided by patient risk factors, number of episodes of AF, and the presence or absence of haemodynamic disturbances during the episode of AF. Patients without risk factors for AF, who develop a single episode of AF that is well tolerated, do not need to be started on an AAD as the majority of patients (60%) will have a single episode of AF. Patients with risk factors for AF, recurrent AF, or with haemodynamic compromise during episodes of AF, should be started on an AAD. Although several AADs have been used, the most commonly used agents after cardiovascular surgery include amiodarone and sotalol. Sotalol requires daily monitoring of the QT interval during the first five doses. During this loading period, it should be discontinued if the corrected QT interval exceeds 480 ms or if there is any evidence of proarrhythmia. Because sotalol is predominantly renally cleared, its use is contraindicated in patients with renal dysfunction. In patients with severe left ventricular dysfunction and congestive heart failure, amiodarone is the drug of choice.

Primary prevention strategies

Several strategies to reduce the incidence of postoperative AF after cardiac surgery have been studied. A discussion of all primary prevention strategies is beyond the scope of this review. Because AADs are easily administered, widely available, and have proven efficacy in the prevention of postoperative AF, this discussion of primary prevention is limited to BBs and AADs.

β-adrenergic blocking agents

A large body of evidence supports the fact that BBs are effective in reducing the risk of postoperative AF. Several agents (atenolol, metoprolol, timolol, propanolol, and carvedilol) have been studied in the postoperative setting. Significant differences in relative efficacy have been reported in these studies. In a meta-analysis, BB significantly reduced the risk of postoperative AF by 64% (odds ratio (OR) 0.36, 95% confidence interval (CI) 0.28 to 0.47; p<0.001); however, when trials where non-study BB which were withdrawn in the placebo arm were excluded, there was a more modest 31% reduction in postoperative AF (OR 0.69, 95% CI 0.54 to 0.87; p = 0.002).5 Therefore, in contemporary practice, BBs can be expected to lower the risk of postoperative AF by approximately 30%, leaving many patients to develop AF despite treatment.

Based on the available literature on BB in the postoperative setting, the following conclusions can be made:

  • BBs significantly, but modestly, lower the incidence of postoperative AF following CABG, valve, and CABG + valve surgery.

  • BB withdrawal is a significant risk factor for the development of AF and should be avoided.

  • BBs should be initiated as early as possible in the postoperative period. Initial intravenous administration of metoprolol or esmolol may provide more effective serum drug concentrations and provide more “agility” to the clinician for titration of the medication around the volatile haemodynamics and risk of bradycardia in the early postoperative period.

The 2006 American College of Cardiology/American Heart Association/European Society of Cardiology (ACC/AHA/ESC) guidelines for the management of AF gave BBs a class I indication for the prevention of postoperative AF.6

Antiarrhythmic agents

Significant clinical issues limit the general use of AADs in patients undergoing cardiac surgery. Patients undergoing cardiac surgery usually have coronary artery disease, structural heart disease, and impaired left ventricular systolic function. Furthermore, drug–drug interactions, alterations in absorption, first pass metabolism, protein binding, volumes of distribution, and impairment of elimination pathways, may unpredictably impact the pharmacokinetics and bioavailability of these agents in the early postoperative period. These factors contribute to the risk of proarrhythmia associated with Vaughn Williams class I (sodium channel blockers) and III (potassium channel blockers) AADs. Among antiarrhythmic agents studied for the prevention of AF following cardiac surgery, amiodarone and sotalol appear to be both safe and efficacious in reducing the risk of postoperative AF.

Amiodarone is a class III antiarrhythmic agent. Due to its class III antiarrhythmic properties, it can prolong the QTc interval. However, proarrhythmia is rare with amiodarone even in patients with significant ischaemic and structural heart disease. Amiodarone is metabolised by the liver and has many drug–drug interactions that are often neglected by physicians prescribing amiodarone. A partial list of drugs frequently used perioperatively that have significant interactions with amiodarone is provided in table 1. Careful attention should be paid to the potential drug–drug interactions associated with amiodarone.

Table 1 Commonly prescribed medications and their interactions with amiodarone (this represents a partial list of common drug–drug interactions)

Studies evaluating the use of amiodarone for postoperative AF vary in dosing strategies. However, regardless of the dosing strategy, amiodarone has been demonstrated to decrease the risk of postoperative AF. Daoud et al randomised 124 patients undergoing CABG with or without valve surgery to oral amiodarone (200 mg orally three times daily for 7 days followed by 200 mg/day until discharge) versus placebo. A significant reduction in postoperative AF was observed in the amiodarone treated group (25% vs 53%, p = 0.003). Patients in the amiodarone treated group had significantly mean (SD) shorter hospital stays (6.5 (2.6) vs 7.9 (4.3) days, p = 0.04) and lower total hospital cost compared to the placebo group.7 In a large (601 patients) randomised study comparing oral amiodarone (10 mg/kg/day started 6 days preoperatively and continued for 6 days following surgery) to placebo, Mitchell et al reported a significant reduction in atrial tachyarrhythmias with amiodarone (16.1% vs 29.5%, hazard ratio (HR) 0.52, 95% CI 0.34 to 0.69; p<0.001). Postoperative BBs were used in half of patients in both the treatment and control group. The risk of sustained ventricular arrhythmias was also lower in the amiodarone group versus the placebo group (0.3% vs 2.6%, p = 0.04).8 In a meta-analysis of 18 trials including 3295 patients, amiodarone reduced AF from 33.2% in the control group compared to 19.8% in the study group (OR 0.48, 95% CI 0.40 to 0.57). Amiodarone was associated with more bradycardia (OR 1.66, 95% CI 1.73 to 2.47) and less ventricular tachycardia or ventricular fibrillation (OR 0.45, 95% CI 0.29 to 0.69) in this study.5 In the AFIST II study, White et al randomised 160 patients undergoing cardiac surgery to amiodarone versus placebo and then to atrial septal pacing versus no pacing using a 2×2 factorial design. All therapies were initiated within 6 h after surgery. Amiodarone was given by intravenous infusion for the first 24 h (1050 mg total) followed by oral therapy for 4 postoperative days (4800 mg total). Postoperative BB use was very high in both the amiodarone and the placebo arm of this study (80.5% and 83.1%, p = NS). Amiodarone was found to decrease the risk of AF by 42.7% compared with placebo.9 This finding is significant in that amiodarone provided further reduction in postoperative AF above and beyond the impact of BBs. There was no difference in AF between the pacing and no pacing groups.

In general, studies of amiodarone prophylaxis show that amiodarone may reduce the incidence of postoperative AF by approximately 50%. This appears to be true whether amiodarone is given orally for 1 week before surgery or intravenously starting immediately postoperatively. Efficacy seems to be related to the total loading dose due to its large volume of distribution. Amiodarone may provide further reduction in the risk of AF when added to a BB,9 although this has not been consistently demonstrated in clinical trials.

Sotalol is a Vaughn-Williams class III antiarrhythmic agent with potassium channel and β-adrenergic blocking properties. Many randomised studies have evaluated the use of sotalol for the prevention of AF after cardiac surgery. In many of these studies it is not entirely clear how many patients in the placebo group were on BBs, and in other studies patients had their preoperative BBs withheld following surgery.10 11 These issues must be considered when interpreting the results of sotalol on AF in these trials. In a meta-analysis of 14 trials evaluating sotalol for the prevention of postoperative AF, sotalol reduced AF significantly compared to a placebo control (33.7% vs 16.9%, OR 0.37, 95% CI 0.29 to 0.48). In the studies that compared sotalol to a BB, AF was reduced with sotalol compared to BB (25.7% vs 13.7%, OR 0.42, 95% CI 0.26 to 0.65).5

Sotalol should not be used in patients with severe left ventricular dysfunction. Because it is renally cleared, its use is contraindicated in significant renal insufficiency. Sotalol has the potential to prolong the QT interval and result in torsade de pointes. For this reason the QT interval must be monitored closely during the first five doses of sotalol. In the meta-analysis mentioned above, more patients were withdrawn from treatment in the sotalol group due to hypotension and bradycardia (6.0 vs 1.9%, p = 0.004).5

The 2006 ACC/AHA/ESC guidelines for the management of AF gives sotalol a class IIb indication for the primary prevention of postoperative AF.6

Other pharmacologic treatments

Given the evidence demonstrating the role of inflammation and oxidative stress in AF after cardiac surgery, a number of anti-inflammatory and anti-oxidant treatments have demonstrated promise. These agents include HMG-CoA reductase inhibitors (statins), omega-3 fatty acids, non-steroidal anti-inflammatory agents, and steroids. In a study of 200 patients undergoing cardiothoracic surgery, patients were randomised to atorvastatin (40 mg/day) versus placebo starting 7 days preoperatively. Atorvastatin was associated with a significant reduction in the incidence of AF versus placebo (35% vs 57%). Multivariate analysis demonstrated a 61% reduction in the risk of AF (OR 0.39, 95% CI 0.18 to 0.85; p = 0.01). Patients who were randomised to atorvastatin and who were also on a BB had a 90% relative reduction in AF (OR 0.1, 95% CI 0.2 to 0.25; p<0.0001). Importantly, 60% and 72% of patients in the placebo and atorvastatin arms, respectively, were on BBs (p = 0.08).12 In a randomised study evaluating the efficacy of N-3 fatty acids for the prevention of AF, 160 patients were randomised to polyunsaturated fatty acids (PUFA, 2 g/day) and to a control group. Treatment was initiated at least 5 days preoperatively and continued until hospital discharge. Patients randomised to PUFA had a significantly lower incidence of postoperative AF compared to the control group (15.2% vs 33.3%, p = 0.013) and a mean (SD) shorter hospital stay (7.3 (2.1) vs 8.2 (2.6) days, p = 0.017).13 Finally, in a study evaluating the role of corticosteroids in prevention of AF following CABG surgery, 88 patients were randomised to methylprednisolone followed by dexamethasone for 24 h postoperatively versus placebo. All patients were treated with a BB postoperatively. Patients randomised to steroids had a significantly lower incidence of postoperative AF (21% vs 51%, p = 0.03).14 Although these pharmacologic agents show significant promise, more studies need to be performed to evaluate their role in AF prevention before they can be routinely be recommended.

Box 3 Antiarrhythmic agents for primary prevention of postoperative atrial fibrillation—how much and how long?

  • If seen 1 week before elective surgery:

    • 400 mg orally twice daily or 200 mg orally three times daily

    • 200 mg orally daily postoperatively until discharge (optional: continue 1 month postoperatively)

  • If initiated postoperatively:

    • 150 mg intravenous bolus over 2 min then 1 mg/min × 6 h then 0.5 mg/min × 18 h. Convert to 400 mg orally daily until discharge (optional: continue 200 mg orally daily × 1 month upon discharge)

  • 80 mg orally twice daily (<70 kg)

  • 120 mg orally twice daily (>70 kg)

  • Use caution in renal failure

  • Continue until discharge (optional: continue for 1 month postoperatively)


Several strategies for the prevention and management of AF have been published. A simplified evidence based approach to the prevention and management of AF is provided in figs 2 and 3. All patients who undergo cardiac surgery should be continued on or started on a BB. Postoperatively, BB should be reinitiated immediately after inotropic agents are successfully weaned off. BB should ideally be initiated intravenously with careful attention to prevent hypotension. Intravenous BB can be converted to oral agents when the patient is able to take oral medications. Patients at high risk of AF are started on an AAD. Amiodarone is the drug of choice but sotalol may be used as an alternative agent (box 3). Patients who develop postoperative AF should be rate controlled and attempts at cardioversion should be performed for prolonged and sustained episodes. Stroke and bleeding risk must be addressed at all times during the management of AF after cardiac surgery. If the duration of AF is unknown or if AF has occurred >48 h, patients are at a higher risk of stroke following cardioversion, and therefore patients should be started on intravenous heparin and therapeutically anticoagulated. Attempts at cardioversion should be performed only after a TOE has ruled out a left atrial thrombus. Patients who have been in AF <48 h and successfully maintained in sinus rhythm do not need to be anticoagulated. Patients who develop recurrent AF or haemodynamically unstable AF should be started on an AAD (amiodarone is the drug of choice in this setting). Duration of AAD treatment following cardiac surgery varies in the literature. Given that it is rare for AF to occur beyond 7 days postoperatively,2 11 one strategy would be to continue AAD for a minimum of 1 week postoperatively. Patients with a prior history of AF or who have poorly tolerated recurrent AF may require longer treatment on a case by case basis. Among patients who develop recurrent AF, it is prudent to start anticoagulation. Duration of anticoagulation should be for a minimum of 3–4 weeks. For this reason, patients should be started on warfarin as soon as any concerns of bleeding or anticipated invasive intervention have been alleviated.

Figure 3 (A) Haemodynamically unstable is defined as hypotension, ischaemia, congestive heart failure, and mental status change attributed to low cardiac output. (B) Patients who have been in atrial fibrillation (AF) >48 h are at risk of stroke and should be anticoagulated. In this setting or if the duration of AF is unknown, a transoesophageal echocardiogram (TOE) should be performed to rule out left atrial (LA) thrombus before proceeding with elective cardioversion. Anticoagulation with heparin is usually initiated between 36–48 h. The patient should be therapeutically anticoagulated (international normalised ratio (INR) for protime 2–3 or prothrombin time (PTT) 60–80 s) <48 h of onset of AF. (C) Intravenous (IV) amiodarone may be used as a rate control agent in certain patients who are hypotensive, cannot be rate controlled by conventional means, and cannot be electrically cardioverted.4 (D) The majority of patients who have a single episode of postoperative AF will spontaneously convert within 24 h. If the patient is tolerating the rhythm and is adequately rate controlled it is prudent to observe for 24 h before intervention. (E) Direct current synchronised cardioversion (DCCV) is usually performed in the time period from 24 h to no longer than 48 h after the onset of AF to avoid the need for anticoagulation. AAD, antiarrhythmic drugs; BB, β-blockers; ERAF, early reinitiation of atrial fibrillation; r/o, rule out.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal’s username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.


Postoperative AF may affect one to two in every three patients undergoing cardiac surgery. Despite aggressive implementation of preventative strategies, a significant number of patients will develop AF postoperatively. Patients who develop postoperative AF suffer from haemodynamic instability, an increased risk of stroke, congestive heart failure, and prolonged hospitalisations. A number of new pharmacologic agents are being studied with promising results from randomised trials; however, treatment with single agents leave many patients still at risk. Further studies are required to establish whether the incidence of AF can be reduced to <10% when a combination of two or more of these agents is used.


  1. 1.
  2. ▸ A large, multicentre, international study demonstrating clinical factors associated with increased risk of postoperative AF and its consequences.

  3. 2.
  4. ▸ This study demonstrated that, along with age, a signal average P wave duration of >155 ms was associated with a higher incidence of postoperative AF.

  5. 3.
  6. ▸ A review of the data on the comparative efficacy of defibrillation using monophasic versus biphasic shock waveforms.

  7. 4.
  8. ▸ Intravenous amiodarone is an effective rate controlling agent among critically ill patients compared to intravenous diltiazem.

  9. 5.
  10. 6.
  11. ▸ The 2006 guidelines for the management of AF which include evidence of agents efficacious in primary and secondary prevention of postoperative AF.

  12. 7.
  13. ▸ A randomised study demonstrating that oral amiodarone (600 mg/day) started a minimum of 7 days preoperatively and continued at 200 mg per day postoperatively until discharge significantly reduced the incidence of postoperative AF after cardiac surgery. 57% of patients in this study had concomitant valve surgery.

  14. 8.
  15. ▸ A randomised study demonstrating a significant reduction in postoperative AF with 13 days of amiodarone initiated 6 days preoperatively. A significant reduction in AF was demonstrated in several subsets of patients including young patients, elderly patients (>65 years), and patients on BBs. All patients had valve surgery with or without coronary artery bypass grafting.

  16. 9.
  17. ▸ This study demonstrated that intravenous amiodarone initiated within 6 h after cardiac surgery and continued for 4 days postoperatively (total dose 4800 mg intravenously) significantly decreased the incidence of postoperative AF. No significant difference in AF was noted with bi-atrial pacing.

  18. 10.
  19. 11.
  20. A randomised study demonstrating that sotalol 80 mg orally twice daily started 2 h before elective cardiac surgery significantly (and continued for 3 months postoperatively) decreased the incidence of postoperative AF among patients undergoing CABG surgery (n = 220) and aortic valve replacement ± CABG (n = 35).

  21. 12.
  22. 13.
  23. 14.


  • Competing interests: In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The author has no competing interests.