REVEAL AF Dampens Excitement for the Search for Short-Duration AF

John Mandrola, MD


May 17, 2017

When an industry-funded study presented as a late-breaking clinical trial at a major medical meeting delivers positive results, it usually increases sales of the company's device or drug. Sometimes such a trial advances medical knowledge.

At the Heart Rhythm Society (HRS) 2017 Scientific Sessions, the single-arm, open-label, Medtronic-funded REVEAL AF study of its implantable cardiac monitor (ICM) succeeded in adding knowledge, but its findings may slow the rush to search for subclinical episodes of atrial fibrillation (AF).

Enthusiasm for use of implantable monitors surged after publication of the CRYSTAL AF study in 2014.[1] CRYSTAL AF showed that placing an ICM in patients after cryptogenic stroke improved the detection of 30-second episodes of atrial fibrillation (AF) more than sixfold over conventional follow-up. But finding a half-minute episode of AF is not the same as preventing a stroke.

Even so, the talking point of CRYSTAL AF was that after 3 years, 30% of patients in the ICM group had detectable AF.

What about detecting subclinical AF in high-risk people without symptoms? Prevention is better than treatment, right? Screening for AF has become a hot topic.[2,3]

The REVEAL AF investigators, therefore, set out to determine the incidence of undiagnosed AF in a high-risk population. The thinking here is simple: AF increases the risk of stroke. Stroke may be the first manifestation of AF in people who have no symptoms. Thus, it would be good to detect this hidden AF.

The REVEAL AF methods were straightforward: Insert the ICM in 394 people at high risk for stroke but without known AF and then measure the incidence of AF over time. The mean CHA2DS2VASc score of enrolled patients was 4.4±1.3.

The detection rate of ≥6-minute episodes of AF at 18 months was 29.3%. At 30 months, the detection rate was 40%. The authors also reported that among these patients with ≥6 minutes of AF detected, 56% were prescribed an oral anticoagulant and 15% were prescribed rhythm-control medication. Clinical decisions in REVEAL AF were left up to the treating clinicians.


REVEAL AF found that the incidence of short-duration AF in high-risk patients without a history of stroke was essentially the same as it was in patients with cryptogenic stroke. This observation reduces the value of short-duration AF as a surrogate marker of stroke.

Here is why: When patients present with stroke and the cause is unclear, the prevailing theme now is to look for AF. In the US, this often means use of the well-reimbursed, well-marketed ICM. But if the average high-risk older person has the same amount of short-duration AF as a person who just had a stroke, how does this help decide on therapy?

Medtronic might be tempted to use REVEAL AF results to market greater use of monitoring to detect subclinical AF. This would be a mistake. Why gather more information when there is no evidence that using anticoagulation in patients with short episodes of AF improves outcomes? The key word being short.

In the ASSERT study, older adults (≥65 years) with either a pacemaker or defibrillator and asymptomatic atrial tachyarrthymia episodes ≥6 minutes had a 2.5-fold increased risk of ischemic stroke compared with device patients without subclinical atrial tachyarrhythmia.[4] But a more recent exploratory analysis of ASSERT, published in the European Heart Journal in March 2017, reported that only AF episodes lasting longer than 24 hours were associated with an increased risk of stroke (HR 3.24, 95% CI 1.51–6.95; P=0.003).[5]

Results from a pooled analysis of multiple prospective studies further weakens the value of finding AF episodes of 30-second or 6-minute duration. These authors reported that a daily maximum of 6-hour AF burden carries a 17% increase in risk of stroke or transient-ischemic-attack (TIA) events and a 12-hour AF burden carries a 37% increase in risk.[6]

Given that only longer-lasting AF associates with stroke, a more effective way to detect AF may be to simply feel a pulse or use a blood-pressure cuff designed to detect an irregular rhythm or a mobile smartphone app.

Before embracing a strategy of implanting metal devices in people or turning asymptomatic people into patients, we should have an outcomes trial proving that doing so is beneficial to patients, not just to hospitals, doctors, and device companies.

The other problem with promoting the diagnosis of conditions of uncertain significance is that therapeutic optimism typically pushes doctors to treat. In REVEAL AF, more than half the patients were prescribed anticoagulants. This may turn out to provide a net benefit, but without evidence we don't know. Recall that trials reporting anticoagulant benefits were done in people with either symptomatic AF or AF long enough to be recorded on multiple ECGs.[7,8,9,10]

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.[11]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.