Cardiology Medicine

Atrial Fibrillation

Atrial Fibrillation is the most common arrhythmia (1, 2) and it can often feel like an on call shift is dominated by patients with AF when it is not controlled. Patients with AF present to Emergency Departments and acute medical takes frequently and it complicates many surgical patient admissions too. On the face of it, it is simple to manage with well-defined treatment options. However, the actual decision-making can feel unfamiliar and less straightforward when the picture becomes more complicated.

Professor Kahled Albouaini is a Consultant Cardiologist at the Royal Liverpool and Broadgreen University Hospitals, Liverpool. In this episode, he takes us through AF and simplifies many of the concepts and treatment options. As a bonus, we have a second VIDEO episode entitled “The Many Faces Of AF” that takes you through some of the ECG findings in AF and how to approach them.

Listen below:

Or take it with you:


Not iTunes

BONUS!! See below for our bonus vodcast episode: The Many Faces of AF



Atrial Fibrillation is common. It is the most common arrhythmia with a prevalence of 1-2% in the general population and is projected to increase (1, 2). With a significant proportion of AF being asymptomatic (3), the actual incidence may be higher still.

The lifetime risk for developing Atrial Fibrillation is 1 in 4 for men and women aged 40 and over (4)


The patient with Atrial Fibrillation has a higher mortality risk. Given the prevalence of AF and the rate it complicates admissions, it can cause problems for many patients. How does AF cause a problem?

af fast

AF is always fast. The atria contract with chaotic, rapid activity at a rate over 300 bpm, seen on the ECG as no distinct P waves and fibrillation which may be fine or coarse.

The AV node slows down the conductance of the fast atrial activity but because it is so fast, AF is conducted with a rapid ventricular response. Due to the chaotic and irregular nature of the atrial activity, this is replicated in the conducted activity and give the irregularly irregular ventricular rhythm that is synonymous with AF.

Cardiac output is reduced in the presence of AF. Ventricular filling occurs in diastole and is affected in two ways. First, the atria would usually deliver a “kick” by contracting and aiding ventricular filling in diastole. This may contribute 20-50% of ventricular filling and is lost in AF. Secondly, when the ventricles are contracting rapidly, less time is spent in diastole which reduces the time available for filling and therefore the amount of filling.

With prolonged AF, the ventricle remodels. Remodelling encompasses the changes that occur in the ventricle. The chaotic contractions of the ventricle lead to a reduced left ventricular ejection fraction (LVEF). Reduced LVEF causes the left ventricle initially to hypertrophy to compensate but then dilates, further reducing the ejection fraction.

The dilated ventricle results in secondary mitral regurgitation. The mitral valve leaflets are normal, but due to the dilation of the cardiac ring, they are pulled apart from one another and become lax and prone to regurgitation. This regurgitation coupled with the disorganised atrial activity dilates the left atrium.

The reduced diastolic filling and reduced LVEF explain why AF reduces cardiac output by as much as 50%

An echocardiogram that shows atrial dilatation, mitral regurgitation, ventricular dilatation and reduced LVEF suggests the patient has a significant burden from their AF

The reduction of cardiac output has a knock on effect on the kidneys where the renal blood flow is reliant upon adequate cardiac output and over time patients with AF have a deterioration in their renal function.

The dilated left atrium gives rise to stasis of blood within. This stasis or turbulent flow alongside endothelial injury creates a pro-thrombotic state. The left atrial appendage forms a pocket for clot formation. Once a thrombus has formed, the patient is at significantly higher risk of thromboembolic events such as stroke, limb ischaemia and bowel ischaemia. Up to 20% of all strokes are caused by thromboembolic events resulting from AF (11).

virchow's triad

Causes of AF

The most common cause depends upon prevalence of rheumatic heart disease. Valvular heart disease caused by rheumatic heart disease is the most common cause worldwide. Valvular AF refers to AF related to predominantly mitral stenosis or prosthetic heart valves.

In the UK, where rheumatic heart disease prevalence is low, AF is commonly non-valvular and can be caused by:

  • Hypertension
  • Coronary Artery Disease
  • Hyperthyroidism
  • Alcohol
  • Pulmonary embolism
  • Pericarditis
  • Cardiomyopathy

Classification of AF

Besides valvular versus non-valvular, Atrial Fibrillation can be divided according to the duration.

  • Paroxysmal AF: episodes lasting less than 7 days (may terminate spontaneously or with intervention)
  • Persistent AF: lasts more than 7 days
  • Longstanding persistent AF: lasts longer than 12 months
  • Permanent AF: persistent AF where rhythm control is no longer pursued

A further method often used to classify AF is by fast and slow. Remember, AF is always fast and what is meant by fast and slow is the ventricular response.

  • Fast AF: AF with ventricular response >100 bpm
  • Controlled AF: AF with ventricular response 60-100 bpm
  • Slow AF: AF with ventricular response <60 bpm


Atrial flutter must be differentiated from fibrillation. Coarse AF can appear like flutter but is different.

The importance of differentiating the two is due to the treatment options. Flutter is much easier to ablate. One treatment option for atrial fibrillation is flecainide. However, if this is given in isolation to a patient with flutter then the increased conductance through the AV node places the patient at risk of ventricular fibrillation and beta-blocker must be given alongside flecainide. It would be best to leave this aspect of management to the experts.

The inferior leads (II, III, aVF) and V1 are the best leads to inspect for flutter waves. Flutter waves have a regular pattern with a characteristic “saw tooth” appearance. Coarse AF would still appear chaotic and irregular.



Coarse AF:

coarse AF

Presentation of AF

A significant proportion of AF is asymptomatic (3) and may be found by chance during healthcare visits for other reasons.  Those who do present with AF as their primary problem can present with a number of symptoms:

  • Palpitations
  • Chest pain
  • Fatigue
  • Breathlessness
  • Syncope/presyncope
  • Features of left ventricular failure
  • Thromboembolic events such as stroke, limb ischaemia, bowel ischaemia



controlled AF

It sounds silly to say but it is an ECG based diagnosis so in any patient you suspect AF, you need to perform an ECG in order to diagnose it.

As with any condition and patient, your approach to investigations will be tailored depending on how they have presented. Investigations to consider would include:

  • Full blood count (FBC) – anaemia and infection can both cause AF or explain a sudden poor control
  • Urea & Electrolytes (U&Es) – derangement of potassium will need correcting and can contribute to AF, an AKI may indicate severe hypovolaemia and need for IV fluids or may be as a result of the AF itself (you decide which it is)
  • Magnesium – hypomagnesaemia is sometimes considered to contribute and may need correcting
  • Thyroid function tests (TFTs) – hyperthyroidism is an important cause to find
  • Chest X-ray – may show signs of LVF, pneumonia if suspected or cardiomegaly
  • INR – a patient on warfarin would need this checking, but if not anti-coagulated, they soon will be and if starting warfarin then a baseline INR is needed
  • Echocardiogram – essential for managing AF, it does not need to be as an inpatient if the rate is controlled and the patient is well. A transthoracic echo is sufficient in most cases but a transoesophageal may be ordered if it is vital to assess for an atrial thrombus.


Do you send a troponin?

Only if you think the patient has an acute coronary syndrome.

It is very frequent that troponin is sent in patients with arrhythmias, AF included, because it is a heart problem. In the absence of signs and symptoms that suggest the patient has myocardial ischaemia, why are we doing the test? Only perform troponins if clinically indicated by signs and symptoms that suggest myocardial ischaemia or infarction. A tachycardia, if fast enough, will cause a mismatch between oxygen supply and demand to the myocardium and result in myocardial damage and a troponin rise that does not result from a coronary artery blockage. The ECG in fast AF may well show lateral ST depression which is rate related rather than ischaemia related. A positive troponin that was ordered with good but incorrect intentions may distract from the AF management and lead you down the wrong path entirely.


The four important aspects of acute AF management are:

  1. Trigger – sought through your history, examination and investigations (see above)
  2. Rate control – or rhythm control
  3. Anti-coagulation – whether warfarin or alternative but NOT aspirin
  4. Echocardiogram – either IP or OP oncerate is controlled

rate v rhythm

The 2014 NICE guidelines (5) are a great aid for decision making when consider rate versus rhythm and the exact method to use.

Rate versus rhythm

Rate control is the first line strategy for patients with AF unless:

  • whose atrial fibrillation has a reversible cause
  • who have heart failure thought to be primarily caused by atrial fibrillation
  • with new‑onset atrial fibrillation
  • with atrial flutter whose condition is considered suitable for an ablation strategy to restore sinus rhythm
  • for whom a rhythm control strategy would be more suitable based on clinical judgement

When considering the reversible causes of AF, it is easy to become distracted by AF running at  a fast rate and neglect the cause. The cause still needs treating. You may find that with this treatment, the patient returns to sinus rhythm and cardioversion is not needed.

New-onset atrial fibrillation refers more to the patients who present acutely symptomatic with a defined onset time within 48 hours. These patients can be cardioverted acutely alongside initiation of anti-coagulation and do not need to be therapeutically anti-coagulated prior. Those patients outside of this window and receive anti-coagulation for 4-6 weeks and undergo a planned cardioversion.


There is a group of patients, though uncommon in AF, who will be unstable and need immediate cardioversion which can sometimes be misinterpreted. As per the resuscitation council guidelines above, the adverse features requiring immediate electrical cardioversion are:

  • Shock – more than just a bit of a low BP, this would be hypotension or signs of hypoperfusion. The blood pressure may be slightly low and will be variable but shock is different to this. NICE goes further and describes “life‑threatening haemodynamic instability caused by new‑onset atrial fibrillation”
  • Syncope – a patient who collapses in front of you having switched into fast AF, not the patient who feels dizzy when they stand up.
  • Myocardial ischaemia – a patient who has objective evidence of ongoing myocardial damage as a result of their AF
  • Heart failure – acute pulmonary oedema/LVF, not the slight breathlessness which is a symptom of AF or a bit of pedal oedema


Electrical versus Pharmacological Rhythm Control

If a patient prevents with symptom duration less than 48 hours, either pharmacological or electrical cardioversion can be used. If a patient presents after 48 hours then rate control should be used.

The choice will be that of the treating doctor, the patient themselves and the clinical situation. In the middle of the night on an Acute Medical Unit, the resources may not be present for an organised and safe DCCV (DC Cardioversion) but amiodarone is possible. Pharmacological methods can be used initially with a plan to attempt DCCV during the day when it is possible. The patient should be counselled on either strategy and rate control to allow them to make a choice too.

Both electrical and pharmacological cardioversion methods can be delivered safely and have reasonable success rates. Electrical cardioversion has reported success rates of approximately 90% compared to pharmacological success rates of approximately 70% with both having few reported complications (6, 7). Your hospital or trust may well have a policy that can guide you as to the method you choose and is worth being familiar with.

Electrical cardioversion will require coordination between multiple parties including anaesthetist or appropriately trained person for sedation, appropriate environment (eg CCU, theatres, resus or ICU), available defibrillator capable of synchronisation, adequate nursing staff, equipment to manage complications, defib operator +/- supervisor if inexperienced. Ideally, the patient should undergo a formal consent process prior to the procedure.

Should you opt for pharmacological cardioversion, the agent you choose will depend on hospital policy and the patient themselves. Fleicanide can only be given to patients with no evidence of structural heart disease. Amiodarone can be used in the majority of patients however. Amiodarone comes with some cautions too and how it is given may vary so it is worth checking with nursing staff and cardiology/seniors what they are happy with before jumping in.

Amiodarone is given as a loading dose followed by a long infusion. The loading dose can be 5mg/kg but more commonly we would give 300mg over 30min to 1 hour followed by 900mg over 24 hours. The initial loading dose can be given via a larger (18G/green or bigger) peripheral cannula sited in a large vein (ie the antecubital fossa). The following longer infusion can be given via a midline but ideally a centrally placed line. Pragmatically, a midline is simpler, requires less training to site and manage and has fewer complications. What you are able to use may depend on what is available and your hospital policy.

Amiodarone can cause local infusion site reactions and tissue necrosis if it extravasates (the cannula “tissues” or “blows”). Rapid administration can cause hypotension and circulatory collapse.

Rate Control

The mainstay of rate control in AF are beta-blockers, commonly bisoprolol. For patients who cannot tolerate beta-blockers, rate limiting calcium channel blockers such as diltiazem can be used. Verapamil should be avoided due to negatively inotropic properties.

Bisoprolol can be given as 1.25mg and upwards doses. The dose you use will vary depending of patient factors. 2.5mg is a good starting dose and can be repeated as necessary. Bisoprolol has a peak effect given orally between 2-4 hours. You may see an effect sooner than this as the heart rate begins to reduce. If the response is inadequate ie the rate remains above 100-110bpm, then give another 2.5mg dose up to a maximum of 10mg.

Diltiazem is given at a dose of 120mg initially and repeated to a maximum of 360mg (but it varies depending on the brand used). Diltiazem can be combined with bisoprolol if monotherapy is inadequate at achieving rate control.


Digoxin is an often favoured drug in treating AF. however, it has fallen out of favour in more recent years due to the superiority of beta blockers. As mentioned in the episode, it is recommended for sedentary patients with non-paroxysmal AF. this is because when a patient taking digoxin exerts themselves, the rate can escape the control of digoxin. Digoxin as monotherapy is best reserved for sedentary patients (5).

Digoxin can be given as a loading regime either orally or intravenous. Typical loading is 500mcg followed by a further 500mcg 6 hours later. Caution should be used in patients with poor renal function or frailty and 250mcg can be used instean. Continued dosing would then be determined by a treatment nomogram (8) or dose calculator accessible here

Combination therapy can be given but triple therapy of beta-blocker, diltiazem and digoxin would predispose the patient to severe bradycardia and complete heart block and should be avoided.


Anticoagulation is a vital aspect of AF management. AF is found in up to a third of all ischaemic strokes (9) and confers a worse prognosis with increased risk of death and more severe neurological dysfunction (10).

The CHA2DS2VASc tool is used to assess stroke risk in the presence of AF and comprises:

  1. Congestive Cardiac Failure – 1 point
  2. Hypertension – 1 point
  3. Age >75 – 2 points
  4. Diabetes – 1 point
  5. Stroke/TIA/Thromboembolic event – 2 points
  6. Vascular disease – 1 point
  7. Age 65-74 – 1 point
  8. Sex category female – 1 point

Maximum score = 9 points

Lip et al theorised the approximate stroke rates for each nodal point of the CHA2DS2VASc tool with curious results (11). They demonstrated the possibility that a total score from 0-4 yields an equivocal stroke rate in percentage terms at 1 year (ie 0 = 0%, 1 = 1%, 2 = 2% etc). However, these were estimates risks for a cohort already taking anti-coagulation based upon warfarin conferring a 64% risk reduction for stroke in AF and a low total number of events. That said, CHA2DS2VASc performed the best in identifying a greater number of patients as high risk who developed a thromboembolic event.


  • Low risk: Score 0
  • Intermediate risk: Score 1
  • High risk: Scor >1

According to NICE guidelines for anticoagulation in stroke (5), anticoagulation should not be offered to those under the age of 65 with a score of 0 in men or 1 in women where the only risk factor is being female.

For men with a score of 1, anticoagulation should be considered.

Whether you do or do not provide anticoagulation for these patients will reflect the preference of the cardiology team, local policy and patient choice.

Any patient with a score of 2 or greater should be offered anticoagulation.

The exact anticoagulant agent that is used will again depend upon patient choice and local policy. The choice is the use of D/NOACs (eg rivaroxiban, apixaban, dabigatran) or a vitamin K antagonist (ie warfarin). The trend is for increasing use of the newer agents within their licensing agreements.

An anticoagulant is prescribed taking into account bleeding risk which is assessed by the HAS-BLED risk stratification score (12).

  • Hypertension (systolic >160mmHg) = 1 point
  • Anormal renal or liver function = 1 point each (max 2)
  • Stroke = 1 point
  • Bleeding = 1 point
  • Labile INRs = 1 point
  • Elderly (>65 years) = 1 point
  • Drugs or alcohol = 1 point each (max 2)

Maximum score = 9 points

Low risk = 0 points

Intermediate risk = 1-2 points

High risk = 3 or more points

The reason for calculating both scores would be to categorise the risk of bleeding against the risk of stroke. This would be a discussion to have with your seniors if you feel the bleeding risk is too high compared to the stroke risk.

A common scenario faced is the risk warfarin or anti-coagulation poses to patients who fall. Previously, a study has sought to quantify the risks of falls and estimated a patient would need to fall 300 times a year before the risk of bleeding outwieghed the risk of stroke (13). However, with increasing numbers of patients on anti-coagulation  and an aging population who will no doubt become increasingly at risk of falls, this number may change. Hwoever, perhaps it highlights the knee jerk reaction of witholding anti-coagulation in the presence of falls needs to be thought through fully and will need discussion with patients and their relatives.

echo af

Firstly, AF needs to be rate controlled before the echo can be performed. A tachycardic patient makes full assessment difficult.

The reasons for performing the echo are three-fold. Firstly, by assessing for mitral regurgitation and left atrial dilatation, it can help demonstrate the burden of AF on the heart when chronic. A patient who has presented for the first time with AF may have had AF that persisted for a long period that had not become apparent. Mitral regurgitation and a dilated left atrium suggest a greater burden of disease. Mitral stenosis however, would be suggestive of rheumatic heart disease as the trigger for AF.

Secondly, the left ventricular ejection fraction can be estimated. This again shows the burden of disease but reflected in the function of the heart. If a patient has presented with acute heart failure, then you may need an echo to demonstrate the acute deterioration is function and a second echo once the patient has once more been optimised on treatment.

Third, and most improtant in patients where rhythm control is the aim, it can demonstrate the presence of a thrombus in the left atrium. This is important because the absence of a thrombus may allow earlier cardioversion but the presence would require longer anticoagulation. However, stroke prevention with anti-coagulation should not be delayed awaiting the results of an echo if the need and strategy has already been agreed (5).

A trans-thoracic echo (TTE) is usually sufficient, though there are select groups of patients where an trans-oesophageal echo (TOE) would be needed. As a guide, NICE recommends (5):

  • when TTE demonstrates an abnormality (such as valvular heart disease) that warrants further specific assessment
  • in whom TTE is technically difficult and/or of questionable quality and where there is a need to exclude cardiac abnormalities
  • for whom TOE‑guided cardioversion is being considered

take home points

AF is common and can affect patients across specialties so you need to be comfortable with being involved in the management of AF.

Consider in each patient the four pillars of AF:

  1. Trigger – assess for, investigate and treat any underlying causes
  2. Rate (or rhythm) control – initiate beta-blockers and increase as needed
  3. Anti-coagulation – assess stroke risk and bleeding risk
  4. Echocardiogram – assesses the burden of AF, LV function and the presence of a thrombus

Thank you to Professor Albouaini for his help in producing this resource. You can check out his cardiology resources on facebook by searching for Cardiology Cases UK or take a look at his site that offers basic and advanced courses in cardiology:


Speak soon



GET IN TOUCH! Either through the comments section or email if there are any comments or corrections or if you want to be involved with future topics.

RATE US on iTunes to help others find the resource more easily!

SHARE US with others you think would benefit or even your training leads to help us distribute the resource further! Twitter: @FOAMdation


  1. Savelieva, I. and Camm, A. J, Clinical trends in atrial fibrillation at the turn of the millenium. Journal of Internal Medicine, 2001, 250: 369-372
  2. Israel et al, Long-term risk of recurrent atrial fibrillation as documented by an implantable monitoring device: implications for optimal patient care. J Am Coll Cardiol. 2004 Jan 7;43(1):47-52
  3. Lloyd-Jones DM et al, Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation. 2004 Aug 31;110(9):1042-6. Epub 2004 Aug 16.
  4. NICE Clinical Guideline [CG180]: Atrial fibrillation: management. Published June 2014, last updated August 2014. Accessed 07/09/18
  5. Crijns, Harry J.G.M. et al, Contemporary real life cardioversion of atrial fibrillation: Results from the multinational RHYTHM-AF study International Journal of Cardiology, 2014; 172 (3): 588 – 594
  6. Ron Pisters et al, for the Euro Heart Survey Investigators; Clinical correlates of immediate success and outcome at 1-year follow-up of real-world cardioversion of atrial fibrillation: the Euro Heart Survey. EP Europace, 2012;14 (5): 666–674
  7. Friberg L et al, High prevalence of atrial fibrillation among patients with ischemic stroke. Stroke. 2014 Sep;45(9):2599-605. Epub 2014 Jul 17.
  8. Steger C et al, Stroke patients with atrial fibrillation have a worse prognosis than patients without: data from the Austrian Stroke registry. Eur Heart J. 2004 Oct;25(19):1734-40.
  9. Lip GY et al, Identifying Patients at High Risk for Stroke Despite Anticoagulation: A Comparison of Contemporary Stroke Risk Stratification Schemes in an Anticoagulated Atrial Fibrillation Cohort. Stroke. 2010;41:2731–2738

  10. Plisters R et al, A Novel User-Friendly Score (HAS-BLED) To Assess 1-Year Risk of Major Bleeding in Patients With Atrial Fibrillation The Euro Heart Survey. CHEST, 2010;138(5):1093-1100

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.