If a primary care physician receives a continuous ECG monitor report indicating the presence of short runs of an irregular, non-P wave associated, atrial rhythm lasting <30 s what should they do? What should the reporting cardiologist call this rhythm? What does it mean for the patient now and into the future?

As someone who reports ECG monitors and gives advice to primary care physicians these questions come up a lot. I usually go with the description of the short runs of an irregular ......but just this weekend I came across a new term whilst listening to the This Week in Cardiology podcast - micro atrial fibrillation (micro-AF).

This podcast is hosted by the cardiologist Dr John Mandrola and I would say this should be required weekly listening for any practicing cardiologist. It is one of the most well thought through and interesting cardiology podcasts around. John is a "conservative cardiologist" and takes a careful evidence based approach. He's practicing doctor, seeing patients and grapples with the common and difficult clinical scenarios we all come across on a daily basis in our practice. 

Anyway, back to micro-AF, my new learning of the week. Of course this is not a new phenomenon but something which we all recognize when we are reporting continuous ECG monitors. Diagnosis of ​AF should be easy. You get an ECG and there 's an irregular rhythm, no P waves and well that is AF but what about when your looking at continuous ECG monitors and there are short runs of irregular rhythm without P waves  - how long  does this irregular rhythm need to last for it to be called AF. Fortunately there is a definition that it should last more than 30 s for us to consider it AF. Of course this is arbitrary definition - who said 30 s was the right length of time and not 15 or 60 s. Like most of these definitions they emerge from guidelines and a good measure of eminence based medicine.

So what about when these irregular, non P-wave containing rhythms which last less than 30 s? What do we call them? I've seen people write things like short runs of atrial ectopics, although an ectopic should be associated with a P wave. I've see short runs of an irregular atrial associated rhythm or  short runs of pulmonary vein rhythm. Alternatively we could call them atrial junk (as per John Mandrola) or we perhaps just micro-AF.

At face value micro-AF seems like quite an attractive term - after all everyone knows AF and I suppose these are little AF episodes. The term could be useful if they predict AF in the future or are AF but just not 30 s worth of AF then that might be reasonable. What matters is the significance and more importantly what we advise colleagues in primary care and outside of cardiology to do with these patients. I'm concerned if we label them as micro-AF then primary care physicians will be immediately reach for anti-coagulation. If we call them "atrial junk" or other names then we will need to qualify that with some advice, otherwise we are likely to generate significant confusion.  So what do these short runs of irregular, non-P wave containing atrial rhythm actually mean.

The Swedish group who have coined the term micro-AF have published several studies in this area. They define micro-AF as an irregular tachycardia of sudden onset with episodes of ≥5 consecutive supra-ventricular beats without P waves lasting less than 30 s. In STROKESTOP2 they reported on 3,763 of older people (75 years) screened with intermittent ECG recordings for 30 s 4 times day for 2 weeks. 121 (6%) had micro-AF and of these who then underwent a 2-week period of continuous ECG monitoring AF was detected in 26 (13%). In a control group of 250 people without micro-AF then the continuous monitoring detected 7 cases of AF (3%) So we can conclude that having this micro-AF is associated with a four-fold increase in the risk of detection of AF.

In another of their studies called STROKESTOP, 50% of participants diagnosed with micro-AF (this time defined as: abrupt onset, >4 supra-ventricular beats, irregular, no P waves) when followed for 2.4 years ended up with AF detected on monitoring compared with just 10% of patients who did not have micro-AF at baseline.

So it certainly looks like having these short runs on the continuous ECG is associated with a higher risk of having longer runs (>30 s) in the future. So what to do if we find AF on continuous monitoring. Is there a case for anti-coagulation of these short episodes of AF? Is it time to anti-coagulate micro-AF. Does any of this reduce the risk of stroke or is it all just atrial junk!

We have data on this because the LOOP study reported that in individuals with stroke risk factors an implantable loop recorder (ILR) screening whilst resulting in a 3-times increase in AF detection and anti-coagulation there was no significant reduction in the risk of stroke or systemic arterial embolism. These tell us that screening for AF and then treating it with anti-coagulation is not a worthwhile use of resources and does not benefit the patient.

So for now I'm going to avoid the term micro-AF for fear it might cause people to confuse these with AF. Are they atrial junk - well I suppose they could be but I'd feel uncomfortable writing that on a report so for the moment I think the best way is to give a narrative verdict and reassure. 

Direct acting oral anticoagulants (DOAC) have revolutionised the management of atrial fibrillation (AF).  There are four medicines licenced to reduce the risk of stroke and thromboembolism in patients with non-valvular AF – Apixaban, Dabigatran, Edoxaban and Rivaroxaban.

Each DOAC has been compared to Warfarin in large randomised clinical trials and all have shown either non-inferiority or in some cases slight superiority to Warfarin. It is difficult to compare the drugs head-to-head because the trials all had slightly different design however recently there has been some uncertainty about the Rivaroxaban ROCKET-AF trial. Rivaroxaban has become one of the most popular DOACs because of its once daily dose regimen and is very well-tolerated. To date it has been my DOAC of choice.

The ROCKET-AF trial was criticised after publication for potentially being biased in favour of Rivaroxaban. This was because the time in the therapeutic range for the Warfarin patients was just 55%. In other words, the INR of patients on Warfarin was between 2 and 3 just over half the time.  The optimum time in the therapeutic range for a patient on warfarin should be >70%.

More recently the trial has come under fire because the INR monitoring was performed using a point-of-care device that has been subsequently recalled by the FDA. The device produced inaccurately low INR results in some patients. This had the potential to produce falsely low INR readings. This meant that the dose of Warfarin prescribed would have been increased thus exposing patients to potentially higher bleeding tendency.

The faulty INR machine used in the trial led to potential overdosing of warfarin which could have led to higher bleeding in the warfarin group. However, in the trial the annual risk of major bleeding was 3.6% with Rivaroxaban versus 3.4% with Warfarin. If the patients on Warfarin had a higher bleeding tendency because nearly 40% were being over-treated this could have increased the major bleeding rate in the Warfarin group and difference in major bleeding caused by Rivaroxaban

The trial authors have now responded to this by publishing a Research letter in the New England Journal of Medicine. They reporting that 37% of patients treated with Warfarin had a condition that could have been affected by the faulty INR device. They performed an analysis of these patients and reported that the main results of the trial were unaffected.  
Sceptics have responded in an article in the British Medical Journal calling for release of the patient-level data as they don’t trust the investigators analysis. So far the company have not indicated that they will release the data.
Should we continue to prescribe Rivaroxaban for new patients and should patients currently on the drug be confident to continue taking it?

In the post-marketing Xantus study with Rivaroxaban the annual rate of major bleeding was 2.1%, stroke 0.7%, and death 1.9% which were either equivalent to or less than those reported in ROCKET-AF. So in real life clinical practice the drug appears to perform in a manner similar to that achieved in the ROCKET-AF trial. Although that is not surprising since it was the Warfarin comparator group where the problem lies and therefore interpreting difference between Warfarin and Rivaroxaban.

Up to this point the main reasons to choose Rivaroxaban are the simple once-daily dosing and good tolerability. Recently Edoxaban has been launched in the UK and this is also once daily. In the clinical trial of Edoxaban versus Warfarin time in the therapeutic range was 68.4%, which is good and a different point of care INR monitoring machine was used. The average CHADS2 score in the trial was 2.8 and Edoxaban was non-inferior to Warfarin for stroke prevention with a trend towards a lower annualised risk of major bleeding of 2.75% versus 3.43% with Warfarin. Certainly Edoxaban looks an attractive candidate to become my DOAC of choice going forward.