How important are short AF episodes?

A study presented at the recent Heart Rhythm Society meeting in Chicago has added more uncertainty about the significance of short-duration AF episodes.

Before I tell you about the study, I need to clarify what I mean by short-duration AF episodes, sometimes called subclinical AF (SCAF).

SCAF is AF on a monitor that is often not felt by the patient. Doctors are seeing more of this because patients are increasingly being monitored–with pacemakers, ICDs, long-term event recorders and implantable cardiac monitors. These devices can pick up minute-long or hour-long AF episodes.

In the past, AF could only be picked up when there were symptoms that prompted an office visit or the AF lasted long enough to be recorded on a standard ECG. We call this clinical AF. And, importantly, patients with this type of AF were enrolled in clinical trials that showed benefits from taking anticoagulants (warfarin and NOACs).

In 2017, there is no clinical trial evidence that treating short-duration AF (SCAF) with anticoagulants confers benefits. These studies are ongoing.

Now to the new study. It’s called the REVEAL-AF study. This Medtronic-funded study was simple. They took about 400 older patients who had no known AF but risk-factors for getting AF and stroke (high-blood pressure, diabetes, obesity, age > 65, etc) and put Medtronic-branded internal cardiac monitors in them. (The tiny and expensive monitors can be injected under the skin in minutes under local anesthetic.)

Get this:

In patients with no known history of AF, REVEAL AF researchers discovered ≥ 6-minute episodes of AF in 30% of these patients over 18 months. Over 30 months, 40% of this group had short-duration AF. Remember, these were people without known AF.

This revelation is disruptive to the status quo because that is nearly exactly the incidence of short-duration AF that we find in patients who have just suffered a stroke of unknown causes.

Three-years ago, the CRYSTAL AF study launched the widespread use of expensive internal monitors by finding short-duration AF in about 30% of older patients who had just had a stroke of unknown cause. The thinking in 2014 went–we have to look for AF in these patients who just had a stroke because if we find AF, then we can prevent another stroke by giving anticoagulant drugs.

Well, hold on.

There are problems with this thinking. First is the REVEAL AF study that I just told you about. REVEAL AF says the average old person with risk factors (millions of these people) will have short-duration AF. That’s the same incidence of those with stroke.

The second problem with the search for short-duration AF comes from a recent analysis of the ASSERT study. In the original ASSERT study, published in 2012, researchers found short-duration AF recorded in cardiac devices (pacers and ICDs) increased the odds of a stroke by 2.5-fold.

But the new look at the ASSERT study found that only AF episodes lasting ≥ 24-hours associated with stroke events. Other studies confirm these findings–namely that only longer-duration AF increases the risk of stroke.

As I concluded in my column posted on theHeart.org | Medscape Cardiology,

“The more we learn about stroke prediction, the less valuable short-duration AF looks as a surrogate marker. Perhaps we should be looking deeper—at the reasons why the atria fibrillate or other systemic/genetic factors.”

Finally, I want to copy part of a comment on my article from Dr. Paul Dorian, a thoughtful Canadian cardiologist who “warned about relentless seeking of information when we do not know how to deal with the data the screening tests provide, especially since there is likely harm from a diagnosis of SCAF ( anxiety, possible therapy adverse effects, insurability, a label of illness, cost, etc.)”

The title of my column is REVEAL AF Dampens Excitement for the Search for Short-Duration AF

JMM

The column over on theHeart.org | Medscape Cardiology has 11 references.

1 comment

  1. Great post. Anticoagulation publications are often controversial. Problem is likely that the existence of subgroups are neglected. Easily could happen that the controversial studies are based on samples with different subgroup structure. Atrial fibrillation itself is rather a symptom of various comorbidities, where the underlying disease may have greater impact on the stroke risk, than the atrial fibrillation.

    The chad2vasc also have a great setback. It is based solely on discharge diagnosis of natinal registries and not incorporates any laboratory test data. Therefore a predictive value is low.

    Dr. György Báthori

Comments are closed.