Atrial Fibrillation in the ED

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

The way in which patients with AF present unsurprisingly makes a huge difference to the appropriate treatment and management plan, this may range from nothing at all right through to DC cardioversion. It’s really important to manage this group correctly though and have a sound understanding on the topic as timely appropriate treatment can significantly reduce morbidity and mortality, whereas inappropriate treatment can lead to an increase in the M&M.

In this podcast we’ll run through a few fundamental principles for AF, management, considerations of cardioversion and finish up with a paper that looks at the issue of the patient presenting whom is systemically unwell and has secondary AF, probably the trickiest cohort of all to manage.

Fortunately there are some sound guidelines on the topic of AF.  Firstly the Resuscitation Council’s algorithm on peri arrest arrhythmias which departments in the UK will follow and NICE have their guideline on ‘Management for people presenting acutely with atrial fibrillation.’ There are other guidelines out there such as the European Society of Cardiology guidelines on A Fib, these hold many of the same principles but do vary in some of the smaller print. 

So as we mentioned, tachyarrhythmias tend to present in the recently systemically well patient in who they’ve had a sudden change in state. They may have experienced palpitations, some chest pain, collapse, suddenly developed a feeling of anxiety or that something just isn’t right. With your patient looking unwell you perform a parallel history and examination and you quickly reveal they’ve got an irregularly, irregular pulse and that it’s going along like the clappers. You get an ECG and it confirms that the patient has atrial fibrillation with a fast ventricular rate.


So ALS then asks you to look for signs of compromise, for which they state the following 4 are evidence of such (and they note that compromise is unlikely to develop at a rate of less than 150 beats per minute)

  • Hypotension (systolic < 90mmHg)
  • Chest pain
  • Heart failure
  • Decreased conscious level

If there are any of these signs of compromise then it’s recommended you move onto immediate DC cardioversion, if you leave it then they have a chance to become further compromised and the rhythm to degrade to such a point that the patient ends up in cardiac arrest, bad idea! It’s nice to know that you don’t suddenly need to be clambering around to get the anticoagulation on board before you cardiovert them, NICE states ‘If haemodynamically compromised due to acute onset AF carry out DC cardioversion, do not delay for anticoagulation’.

So this is pretty straight forward, look out for these 4 things and when you find one you’re management plan is sorted, time to synchronise the defib and DC cardiovert them! You also need to think if this is going to happen under general anesthesia or sedation, check out our sedation podcast for a refresher on the topic. Sedation wise choosing the right method and right agent, for the right patient with the right pathology is key.

When picking the energy for your DC synchronised shock irregular rhythms such as AF it’s recommended you use a high energy for your DC synchronised cardioverios 120-150J on a bibphasic defib, as opposed to the lower energy levels that you can get away with on your regular narrow complex tachycardia’s such as SVT or flutter that can normally be cardioverted at lower energy levels 70-120 J on a biphasic defib.

We won’t go further down the management line of compromised AF requiring immediate cardioversion but failure will result in the need for further attempts in DC cardioversion and then administration of an antiarrhythmic and further attempts at DC cardioverion.


For those patients who aren’t compromised you have the option of pharmacological intervention with a number of agents for which ALS suggests amiodarone, or digoxin if signs of acute heart failure (but not acutely compromised with it) and diltiazem or a beta blocker such as iv metoprolol without it.

Arguably this is a harder group of patients to manage, synchronised DC cardioversion gives you an immediate response, most of the time it works and when it hasn’t you get to find this out immedatley . With the pharmological treatment of either rate control or cardioversion the onset of action will be comparitvely delayed and only time & close observations will reveal if the strategy has been successful. You also have the added concern of some of the side effects associated with these medications, predominantly hypotension which can flip the uncompromised patient into a state of compromise fairly easily.


So what about the next group, the patients that present with acute onset AF but whom aren’t partcularly tachcardic, they aren’t compromised but they are aware of their AF and therefore are aware that it’s recently started?

Firstly the ‘being aware’ or having insight into running in AF is really important. Your chance of developing an atrial thrombus increases the longer you have AF, the act of cardioversion puts you at an increased risk of this thrombus being dislodged and firing off with embolic consequences, worse case scenario of a stroke. Therefore the management of uncomplicated AF is centred around the duration of AF, with 48hr being a cut off that NICE recommend as the boundary between risk and benefit of cardioversion, with the patients <48 hours duration holding a low risk for thromboembolism as a result of immediate cardioversion.

NICE’s guideline on Management for people presenting acutely with atrial fibrillation is really useful on this topic and we’ll just run through the main points;

Onset < 48hrs, either rhythm or rate control 

  • If using a rhythm control strategy consider either pharmacological or electrical cardioversion depending on clinical circumstances and resources
  • Use flecainide or amiodarone for chemical cardioversion (rhythm control)but flecainide only if no evidence of structural or ischaemic heart disease (Do not offer magnesium or a calcium-channel blocker for pharmacological cardioversion

Onset> 48 hours or uncertain duration of acute AF, use rate control

  • Consider for delayed long-term rhythm control
  • Delay cardioversion until they have been maintained on therapeutic anticoagulation for a minimum of 3 weeks


NICE are big fans of the CHA2DS2-VASc risk score when deciding who should be given anticoagulation. It’s a risk stratification tool that helps to quantify someone’s risk in AF for developing a stroke (, they also recommend anticoagulation in a couple of other categories

  • If your patient presents with AF with an onset < 48 hrs that remain in AF – anticoagualte them
  • If you think there’s a high chance they’ll go back into AF – anticoagulate them
  • In any patient who has been in AF at any point use the the CHA2DS2-VASc risk score to determine if they need anticoagulation

The CHA2DS2-VASc score is a way of prediciting the stroke risk (only for patients with non valvular AF, so those in whom AF isn’t deemed secondary to valvular disease)

Screen Shot 2015-08-04 at 08.39.22


NB this advice is a little different to that of the European Society of Cardiology guidelines on the topic.


What the guideline doesn’t cover is what to do when patients present acutely unwell with AF that may have a rapid ventricular response but when this is a result of an underlying pathology e.g. the patient with pneumonia with a tachycardia of 140-160. You often see these sorts of patients managed with some medications to slow them down but is this the best strategy? It’s often an area of debate as to whether this is a physiological response that is assisting the patient, or whether (being a dysrhythmia) it needs to be rate or rhythm controlled.

There’s a lot of literature out there on the topic of primary fast AF but secondary well that’s pretty scant. A recent paper from the Annals of Emergency Medicine may shed a bit of light on the topic.

It was published in May 2015 and was a retrospective cohort study looking at patients presenting to 2 Canadian university EDs with good onsite cardiology services and cover. It looked at all patients in whom their first ECG showed AF (picked up by the electronically read ECG database) and in whom AF was not thought to be the primary presenting problem (i.e. secondary AF). These patients then had a chart review.

Notably patients who had cardiac procedures (e.g. pacemaker insertion, PCI etc) were excluded, as were patients presenting to the ED for anticoag monitoring.

The primary outcome was the safety of rate or rhythm control attempts and they quantified this as the presence or absence of an adverse ED event (i.e. only those occurring in the department). Major adverse outcomes (AE’s) included;

  • Hypotension requiring vasopressors/inotropes
  • Intubation or NIV
  • New bradycardia requiring pharmacological intervention/pacing
  • Confirmed stroke or thromboembolic event
  • Chest compressions or death

Minor adverse outcomes

  • New hypotension requiring at least 500mL crystalloid bolus
  • Supplementary oxygen via bag-valve-mask

The authors believed there would be a difference in the patients in whom rate and rhythm control was implemented when compared to those who weren’t, the authors used propensity scoring adjustments (this effectively tries to reduce bias due to cofounding variables. These factors included things such as age, vital signs, rhythm and PMH.

They screened over 1500 ECGs of AF/flutter, of whom 416 were presenting with an acute underlying medical illness (i.e. secondary AF). Of this 400+ patients a quarter were managed with rate control, 15 were chemically cardioverted, 15 were electrically cardioverted and the remaining three quarters had no rhythm or rate therapy.

Both treated and non-treated groups were similar in demographics and initial observations and less than 10% of patients had a heart rate that exceeded 150 beats per minute.

Main results……..

Patients receiving treatment for their AF; adverse outcomes in 40.7% (95% CI 4.5%-10.9%)

Patients not receiving treatment for their AF; adverse outcomes in 7.1% (95% CI 4.5-10.9%)

It’s also worth knowing that the adverse event rate was similar in those getting rate or rhythm control, so one didn’t seem to be more of a problem over the other.

Of note rate or rhythm control was only successful in 19% and 13% of patients respectively (with rate control defined as heart rate decreased > 20 BPM within 4 hours)

So it’s fair to say that the paper is a long way off a prospective multi centre RCT, it’s not necessarily applicable to our population, our decision making processes and the way in which the data was obtained isn’t ideal. But it does make for an interesting read and it certainly makes you really think about why you’re trying to intervene with a secondary dysrhythmia such as AF and Flutter, are you just treating numbers and not concentrating on the underlying pathology that you should be devoting your attention to, especially with the near 6 fold increase in complications seen in this study.

Well make sure you get in touch with any comments or questions and we’ll be back again soon with another podcast!




Resuscitation Council’s algorithm on peri arrest arrhythmias

NICE 2014; Management for people presenting acutely with atrial fibrillation

European Society of Cardiology guidelines on A Fib


Scheuermeyer, Frank X., et al. “Emergency Department Patients With Atrial Fibrillation or Flutter and an Acute Underlying Medical Illness May Not Benefit From Attempts to Control Rate or Rhythm.” Annals of emergency medicine 65.5 (2015): 511-522.

Add this content to your CPD

Leave a Comment

Your email address will not be published.

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