Skip to main content

Well, some of you are saying, this is the secret of life. When you’ve been practicing for years, Atrial Fibrillation(AF) is one of the most irritating arrhythmias.

Today’s blog is my attempt at a flow diagram that simplifies an approach to AF. It is based on the European Society of Cardiology Guidelines.

No more looking up stuff, no more duels with the cardiology registrar. This is the real deal.

There are only 3 drugs you need to know: Digoxin, Metoprolol and Amiodarone.

Below is my flow diagram scribble. Now it is not always as straightforward as this and there are nuances you need to beware of, so BEWARE.

The question I ask is, what are we trying to achieve? RHYTHM or RATE control?

RHYTHM CONTROL: This is usually for those patients with less than 48 hours of atrial fibrillation(paroxysmal AF). It is also for some of the ‘persistent’ AF, for symptomatic patients on rate control, for the young symptomatic and for those with known triggers.

RATE CONTROL: This is usually for the permanent AF, for the elderly and those that are symptomatic. The aim is a heart rate of less than 80 beats per minute.


When considering rhythm control there is only one question to ask: “Was the onset of symptoms less than 48 hours ago?” The patient must be able to give a definite onset time. This 48 hour time cut-off is important as it marks a point past which spontaneous cardioversion rates drop.

Symptoms for <48 hours

Is there haemodynamic instability? By this I mean serious instability, not a little chest tightness. If so, then ELECTRICAL CARDIOVERSION is needed. The recommendation is for a biphasic defibrillator, with antero-posterior pad placement. Given that AF is one of the most resistent arrhythmias, 200J is recommended. Do you give a shot of ‘clexane’? I do. It works very rapidly and has almost immediate effects.

Do not cardiovert Digoxin Toxicity.

If there is no haemodynamic instability, then electricity can still be used, however medical management is also possible. The key question to ask here is – ”Is the structural heart disease?”

In most cases we don’t know as the patient doesn’t come in with an ECHO and an angiogram report. If we are sure there is no structural disease then use flecainide, however if you do not have this information, as occurs in most cases, then AMIODARONE is the drug of choice, irrespective or whether the underlying abnormality is left ventricular hypertrophy, coronary artery disease or congestive cardiac failure.

Symptoms > 48 hours

These patients can be immediately anticoagulated for 3 weeks and then get a trans oesophageal echo(TOE), cardioverted and then continue anticoagulation, for a minimum of a further 4 weeks.

If a TOE is readily available and there is not left atrial thrombus, then the patient can be cardioverted immediately. If a left atrial thrombus is present then anticoagulation should proceed.


When rate control is used, the degree of activity in the patients lifestyle is important.

Non-active, significantly bed, or chair restricted patients, can be treated with DIGOXIN. This slows rate effectively in those without an active lifestyle, as the sympathetic response from activity which will normally increase the heart rate, is not present.

In those patients with an active lifestyle, there are three medications to choose from: calcium channel blockers, beta blockers or digoxin. Beta blockers can be used in just about all cases. To ensure that there are no side effects to those patients with airways disease a B1 selective blocker such as METOPROLOL can be used.

So there it is.

Symptoms for less than 48 hours, can be reverted with electricity or Amiodarone

For those with symptoms greater than 48 hours, cardioversion can occur if a left atrial thrombus is excluded on TOE. If there is a thrombus or it cannot be excluded, then 3 weeks of anticoagulation before another attempt.

For those where rate control is important, consider Digoxin or Metoprolol.

Now I know there are other drugs, sotalol, verapamil, magnesium and by all means use these if you are so inclined. This is a simplified, adverse-event minimising approach that I have.

Let me know if you have a better/simpler approach.


Leave a Reply