Patients can have AF without experiencing any symptoms whatsoever. This is termed silent AF. The problem here is that patients may be experiencing a heart rhythm disturbance which is associated with an increased risk of stroke without knowing anything about it. It is always difficult to be accurate about the prevalence of silent AF but it is estimated that 10-40% of all AF patients are asymptomatic. Unfortunately sometimes the first time the patient or their doctor becomes aware of this condition is after the stroke as already happened.
Therefore if you can detect the AF earlier, it would allow intervention that may prevent a potentially life-threatening stroke from happening.
In addition, we know that AF begets AF so if you could pick up short lived episodes of silent AF, early intervention may reduce the AF from becoming more frequent, persistent or even permanent. More importantly detection of silent AF can help patients become more engaged in lifestyle modification which in turn will make them healthier people.
The first thing to think about is who are the people who are most at risk of developing silent AF. The following factors confer increased risk.
- Age (esp once you get above the age of 65 years)
- Elevated body mass index/obesity
- Metabolic syndrome
- Cardiac disease (previous heart attacks and heart failure)
- Cigarette smoking
The risks are even higher in patients who have underlying chronic kidney disease and any of the above risk factors. Similarly anyone who has had a stroke in the past is at a substantially higher risk. Patients who have had AF in the past but the AF has been ‘taken away’ by an ablation, cardioversion or medications are also at a higher risk of having silent AF and at a higher risk of strokes if they have any of the above conditions.
The really unfortunate thing is that most of the factors that i have mentioned such as age, diabetes, cardiac disease and increasing age will not only increase the risk of AF but also greatly increase the risk of strokes. So the people in whom you are most likely to find it are also the people who are most at risk of suffering strokes.
Why is it important to pick up silent AF? It would appear that the risks of silent AF are the same as of visible AF. As things stand, patients who don’t have some evidence of AF are not prescribed anticoagulants. However if AF is found then prescription of anticoagulants will reduce the risk by 60% and this could make a substantial impact on reducing risk in the future.
How do you look for something if you don’t even know of its existence?
- Routine self monitoring of the pulse by patients who are at highest risk or in those above the age of 65: AF is characterised by an irregularly irregular pulse. The pulse will feel completely irregular on palpation. Whilst an irregular pulse can certainly increase suspicion of AF, an ECG is needed to confirm the diagnosis. This is because the pulse may feel irregular because of intermittent extra beats which do not have the same risks of stroke as AF. An ECG will confirm the diagnosis.
- Nevertheless, routine self-palpation of the pulse by patients who are at high risk of AF can be an effective initial strategy. In one study from Belgium, the researchers taught patients of above 75 years of age how to check their own pulse and found that after 3 years of following them up, 10 patients had diagnosed themselves with an irregular pulse and were subsequently confirmed to have AF. This was a pick-up of 4.9%.
- 12 lead ECG: A confirmatory diagnosis of AF can only be made via an ECG. An ECG is best reserved for patients who are found to have an irregular pulse. As AF can be paroxysmal i.e it can come and go, the chances of picking up a paroxysm of AF during a 12 lead ECG just by chance is very slim. Many patients with paroxysmal AF will have a normal ECG but still be exposed to the risks of stroke. So anyone who has the risk factors should not feel exonerated just because they have a normal 12 lead ECG. It is always important to look further.
- There are now mobile phone apps which can help detect AF by recording an ECG on the smartphone. One of these is an excellent little app called KARDIA. The advantage of these is that if the patient feels an irregular pulse, they can record an ECG there and then rather than wait to go and see their doctor for an ECG by which time the AF episode may have spontaneously subsided. Apps like Kardia are understandably superior to having a routine 12-lead ECG. Here is a link to the KARDIA app – https://drsanjayguptacardiologist.com/recommended-products/
- 24 hour ECG monitoring: 24 hour monitoring is better than nothing but again the pick-up for silent AF especially in paroxysmal AF remains low. The pick-up can be doubled if the duration of monitoring is increased to 72 hours. In studies in patients who have already suffered a stroke but are not known to have AF, a 24 hour monitor picked up silent AF in 2.6%. When the duration of monitoring was increased to 72 hours, silent AF was picked up in 4.3% of patients. Currently patients who have had a stroke will at most undergo 24 hour monitoring which is clearly a sub-optimal duration for effective pick-up.
- 30 day monitoring: A much more innovative way to look for silent AF would be a 30 day monitor or patch. These really should be the standard of care but many healthcare systems still use the 24 hour holters which to my mind are a waste of time. The main limitation with these are that they are more expensive. There have been a few small studies to look for silent AF in patients who have already suffered a stroke and they found the detection of silent AF to be between 14-20%.
Implantable loop recorders: Perhaps the gold standard method is an implantable loop monitor (otherwise called a REVEAL device). This is a tiny device which is placed in a tiny pocket under the skin at the top of the chest. It has a battery of up to 2 years and is exceptionally good at picking up heart rhythm changes. There was a very interesting study called the CRYSTAL-AF trial. In this study, patients who had suffered a cryptogenic stroke (meaning an unexplained stroke) were given an implantable loop recorder and the AF detection rate was 8.9% at 6 months, 12.4% at 1 year and 30% at 3 years.
So in summary, here are the main points:
- Silent AF is common.
- Never confuse how you feel with what your future risk is. Just because it doesn’t cause symptoms, doesn’t mean it is not important or dangerous.
- If you fall in a high risk group, it is important to be as vigilant as possible by regularly feeling your own pulse and seeking medical attention if it feels irregular.
- A 24 hour tape (holter) which is often the only investigation that may be offered by your doctor is more often than not, useless. It is better to negotiate for at least a 30 day monitor and certainly if you have ever had an unexplained stroke then a REVEAL device.
I would love to hear your thoughts and would be grateful if you could consider sharing this with anyone who you feel may benefit. Many thanks for all that you do for me.
Keywords: AF; Afib; Atrial fibrillation; Silent AF; Cryptogenic stroke; stroke; implantable loop recorder; REVEAL; Holter
Dan Dobreanu, Jesper Hastrup Svendsen, Thorsten Lewalter, Antonio Hernández-Madrid, Gregory Y.H. Lip, Carina Blomström-Lundqvist, conducted by the Scientific Initiatives Committee, European Heart Rhythm Association, Current practice for diagnosis and management of silent atrial fibrillation: results of the European Heart Rhythm Association survey, EP Europace, Volume 15, Issue 8, August 2013, Pages 1223–1225
Jaakkola, J., Virtanen, R., Vasankari, T. et al. Self-detection of atrial fibrillation in an aged population: three-year follow-up of the LietoAF intervention study. BMC Geriatr 17, 218 (2017).
Hariri E, Hachem A, Sarkis G, Nasr S. Optimal Duration of Monitoring for Atrial Fibrillation in Cryptogenic Stroke: A Nonsystematic Review. Biomed Res Int. 2016
Choe WC, Passman RS, Brachmann J, et al. A Comparison of Atrial Fibrillation Monitoring Strategies After Cryptogenic Stroke (from the Cryptogenic Stroke and Underlying AF Trial). Am J Cardiol. 2015;116(6):889‐893.