The cause of AF is not fully understood. There is a description of 2 requirements for AF:
Substrate for maintenance
Ways that this may occur include:
The focal activation concept suggests that there is an area of the atria that is the primary source of the rhythm disturbance. This may be due to increased automaticity of this region or the presence of reentry pathways within it. The multiple wavelets concept describes reentry pathways on a larger scale. This can be best visualised in an enlarged atrium, where the pathways are able to reenter and trigger further electrical activity.
The disordered electrical activity means that the atria no longer serve their function of helping to load the ventricle prior to systole. This means that ventricular filling is now dependent on passive filling (although this does normally provide the majority of filling anyway) In otherwise well people, this reduction may cause no problems, but in patients with cardiac function that is already borderline, this may precipitate failure.
The disorganised electrical activity also leads to blood stasis in the atria. This can lead to clot formation which may embolise from the left side to cause a stroke.
There are a number of causes of AF, with different ways of categorising them. One useful way is:
This may be asymptomatic in patients (commonly). Some patients may present with palpitations from sensing the irregular heart beat Other presentations may arise from the reduced cardiac function that arise:
In unfortunate cases, the presentation may be of a stroke due to thromboembolism.
The initial investigation will aim to:
Confirm the nature of the pathology
Identify potential triggers
Key investigations will often include:
U&Es, inc Mg2+
Coagulation - pre anticoagulation
The key features on an ECG are:
Irregularly irregular rhythm
Absent p waves
There is variable conduction through the AV node resulting in an irregular ventricular response. This does however maintain a normal sub-nodal conduction pattern, and so the QRS complex will be narrow unless there are other abnormalities. The rate of AV conduction is dependent on a number of factors e.g. vagal tone, AV node function, drugs. It is common for AF to run fast, with rates of 110-160 bpm. The fibrillatory waves can be quite fine or coarse, with coarse waves having the potential to mimic p waves.
This is primarily dependent on the duration:
Recurrent - more than 2 episodes
Paroxysmal - self terminating, < 7 days duration
Persistent - not self terminating, > 7 days duration
Permanent - Duration of > 1year where rhythm control options are not further considered.
The principles of management are:
Management of trigger
Rate vs rhythm control
Assessment for anticoagulation
Patients who present with compromise secondary to the condition should be managed as with other critically ill patients e.g. A to E approach.
Rate vs Rhythm Control
Rhythm control is not always attempted due to the challenge of maintaining any successful cardioversion, and the lack of strong evidence of a benefit from this approach over rate control. Rhythm control should be considered in patients with:
New onset AF
AF with a reversible cause
Heart failure related to the AF
Others who would seem to benefit on clinical assessment.
Otherwise a rate control strategy is advocated. Rhythm control/cardioversion in patients with AF of >48h duration may be at risk of triggering movement of an atrial thrombus. As such, cardioversion should only be undertaken in AF with a duration of less than this timeframe or in patients who have been suitably anticoagulated or had a thrombus excluded (through TOE).
Rate control options include:
Rate limiting calcium channel blocker
Digoxin monotherapy - patients who are sedentary with non-paroxysmal AF
Rhythm control Cardioversion may take the form of an electrical or pharmacological approach. Pharmacological cardioversion may include:
Flecainide (only if no evidence of structural heart disease)
Electrical cardioversion is recommended in those with AF of over 48h duration. Both TOE guided and standard cardioversion are options. Amiodarone treatment for 4 weeks before and 12 months after cardioversion may help the maintenance of sinus rhythm.
The need for long term prophylaxis against AF should then be considered based on the patient’s clinical status. Options include:
Amiodarone- patients with LV impairment
A ‘pill in the pocket’ strategy may be an option for patients with paroxysmal AF with fairly predictable triggers e.g. caffeine.
Left atrial ablation is an interventional option that may be considered in patients who have failed with medical therapy. A ‘pace and ablate’ strategy is an alternative approach in some patients.
The risks and benefits of anticoagulation need considering for patients with AF to mitigate the risk of stroke. The CHA2DS2-VASc score is used to calculate the risk of stroke (https://www.mdcalc.com/cha2ds2-vasc-score-atrial-fibrillation-stroke-risk.) A score of 0 denotes low risk and may not require anticoagulation A score of 2 is deemed moderate to high risk and anticoagulation is often recommended. A score of 1 requires further consideration of contributory factors.
New onset AF is a common result of critical illness. It can have a significant impact on patients in terms of CVS stability and increase their risk.
The incidence has been reported as between 5 and 46%. It appears to be associated with a higher mortality - 45% vs 22% in one study. 44% vs 16% in another. This is likely to be representative of the severity of illness rather than purely causative.
It is also associated with increased morbidity:
Increased inpatient stroke risk (2.6 vs 0.6%)
Progression to persistent AF
Severity of illness
Thoracic blunt trauma
It is important to clarify the difference between patients with cardiovascular stability, and those in whom the AF precipitate CVS instability. In general, patients with CVS instability should be managed by an appropriate ALS approach. This will usually involve DC cardioversion. It is important to bear in mind that patients with accessory pathways e.g. WPW syndrome need treating differently.
The management of more stable patients can be broken down to some key steps, similar to AF in non-critically ill patients, although the components do differ.
Rate control vs rhythm control
There are a number of disturbances in critical illness that may be contributing to the AF and which may be amenable to intervention and these should be treated promptly e.g. electrolyte disturbance, fluid disturbance. Adrenergic activity can be a contributing factor and so should aim to be ameliorated where possible e.g. appropriate analgesia.
Magnesium has been demonstrated to be effective in the management of AF, even outside the context of hypomagnesemia. It's benign profile makes it a very useful first line agent.
Rate vs Rhythm Control
A similar challenge exists here for critically ill patients. In general, rate control is advocated, due to the challenges of successful cardioversion acutely and the fact that this will probably occur with the resolution of the critical illness. There is no evidence of any benefit from a rhythm control approach.
A key differentiation is between cardiovascularly stable and unstable patients. As above, unstable patients should be treated with rapid synchronised cardioversion. In resistant cases, the addition of a pharmacological agent may help.
Beta blockers are a useful first line agent. The IV preparations mean that the dose can be titrated to response to minimise the risk of excessive effects. E.g. metoprolol 2.5-5mg Iv over 2-3 mins Esmolol 50-200mcg/kg/min IV
Rate controlling calcium channel blockers are another effective first line agent, and can be useful in those patients with CIs to beta blockers. Some evidence suggests a better response profile to diltiazem. The down side is the negative inotropic effect they have, meaning they should be avoided in those with poor LV function. E.g. verapamil 0.0375-0.15mg/kg IV over 2 mins
Amiodarone has a slower onset of action but is effective and safe in those with cardiac disease (although there are effects of longer term toxicity). It also results in less negative inotropy than the first 2 classes so may be more useful in those with some CVS dysfunction. It can be a useful in resistant AF, although may result in cardioversion, thus requiring caution about thrombus.
Digoxin is not recommended as a first line agent because of its slow onset. The mechanism of action is also of reduced efficacy in the context of the critical illness causes of AF.
The above described risk assessment tools for stroke and bleeding are not validated in critically ill patients, as they focus on the risk at 1 year. There is still a clear thromboembolism risk, but also a different bleeding risk. The evidence is limited with no clear evidence of benefit of anticoagulation and a small risk of minor bleeding. There is some evidence of an increased risk of ischaemic stroke in this group though. An approach similar to non critically ill patients is therefore advocated by some:
AF of < 48h duration can be cardioverted without anticoagulation.
AF of > 48 should have 3 weeks anticoagulation or TOE exclusion of left atrial thrombus before anticoagulation. Anticoagulation should continue for 4 weeks afterwards.
Unfractionated heparin may be the preferred option in patients with potential need for reversal e.g. surgery