Dr Atul Verma:

I'm Atul Verma. I'm the Director of Cardiology at the McGill University Health Centre in Montreal, Canada.

What clinical questions motivated the investigation of the OCEAN trial?

After someone has a successful catheter ablation procedure for atrial fibrillation, one of the first questions I'm always asked during follow-up is, oh, does this mean I can stop my blood thinner? And so many patients kept asking me this question over and over again that I finally decided, you know what, we should do a trial to answer this question.

What was the study design and patient population?

This was a randomised study, occurring at 56 sites in six countries—so really a global effort. And what we did was we included patients who were at least one year after a successful catheter ablation procedure. And we determined success by at least one holter during the first six months, another holter after six months, and then a 48-hour holter prior to enrollment.

We randomised patients either to rivaroxaban for ongoing anticoagulation or aspirin in the other group. And we followed all of these patients for three years. And every patient had a baseline MRI of their brain and another one at three years as well.

What were the main outcomes of the OCEAN trial?

Our primary outcome for the trial was a composite of stroke, systemic embolism or covert embolic strokes seen on MRIs—so these are large strokes that are more than 15 millimeters. And what we reported today in our late-breaking session is that the event rate was incredibly low in these post-ablation patients.

There were only five of the composite endpoint events in over 600 patients on rivaroxaban and only nine in over 600 patients on aspirin. So that meant that the annualised event rate was only 0.3% in the rivaroxaban arm and only 0.66% in the aspirin arm. That is really, really low and much lower than what we were expecting.

Based on these findings, how should clinicians approach long-term anticoagulation decisions for AF patients post-ablation?

So, importantly, all of the patients in OCEAN had a mean CHA2DS2–VASc score of 2.2, and about 30% of the patients had a CHA2DS2–VASc score of 3 or more.

So I think that for patients who've had a successful ablation, you can't detect on routine holters that they have any further atrial fibrillation. If they have a CHA2DS2–VASc score of 1, 2 or I would even argue 3, and they haven't had a recent stroke, I think these are patients in whom you can talk to them about withdrawing oral anticoagulation.

What further research is needed in this area?

In our trial, as I mentioned, most of the patients either had a CHA2DS2–VASc score of 1, 2 or 3. Fewer than 10% of the patients had a CHA2DS2–VASc score of 4 or more. So in those higher-risk patients, we don't know whether we can safely withdraw oral anticoagulation or whether we need to continue it indefinitely, and that will need to be answered by another trial.

We also don't know what role left atrial appendage occlusion is going to play. I think for these lower-risk patients, according to our study, it has no role to play at all. But in higher-risk patients, perhaps, we just don't know.