- Design
- Multicentre, open-label, adjudicator-masked randomised superiority trial (SINGLE-AF)
- Population
- 1,803 patients with AF, CHA2DS2-VASc 1 (men) or 2 (women), South Korea
- Primary outcome
- Stroke, systemic embolism, major bleeding or cardiovascular death at 24 months
- Effect
- 0.5% v 1.5%; difference −1.0 points (−2.0 to −0.1); HR 0.31 (0.10–0.94)
US and European guidelines already suggest considering anticoagulation in atrial fibrillation at intermediate stroke risk, but as a weaker recommendation built on observational data. SINGLE-AF, published in the NEJM on 28 August, is the randomised test.
In South Korea, 1,803 patients with a CHA2DS2-VASc of 1 (men) or 2 (women) — mean age 60 — were randomised open-label to a DOAC or no anticoagulant. At 24 months the composite of stroke, systemic embolism, major bleeding or cardiovascular death occurred in 0.5% on a DOAC and 1.5% without (difference −1.0 percentage point, 95% CI −2.0 to −0.1; HR 0.31, 0.10 to 0.94). Strokes were 3 v 10. Major bleeding appeared similar, there were no cardiovascular deaths, and serious adverse events were 8.9% v 9.3%.
The absolute gain is small — about one event avoided per 100 patients over two years — and rests on 17 events in total, in a single East Asian population. But it points the same way as the guidelines, and in a trial designed to count bleeding against benefit. For a patient in this band, the choice is now better informed; make it with them, weighing their bleeding profile and preferences. Guidelines have not changed on the strength of this one trial.
- For AF with CHA2DS2-VASc of 1 (men) or 2 (women), discuss a DOAC rather than defaulting to no treatment.
- Expect roughly one event avoided per 100 patients over two years.
- Major bleeding was not clearly higher on a DOAC in this trial.
- Record the patient's choice and revisit the score as they age or gain risk factors.
Why it matters
It moves a weaker, observation-based recommendation onto randomised ground for a large group seen in every clinic.
Don't overread it
The result rests on 17 events in a Korean population with low bleeding rates; it has not changed guideline strength.
The statistics, in plain English
With only 17 primary events in total, the hazard ratio of 0.31 has a wide interval (0.10 to 0.94) — the true reduction could be large or modest. The absolute difference of 1 percentage point over two years is the more useful figure for a patient: about 100 people treated for two years to prevent one event. An open-label design is offset partly by masked adjudication of outcomes.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for top clinical updates, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free