SINGLE-AF: a DOAC cut the net composite in intermediate-risk atrial fibrillation

Design

Multicentre, open-label, adjudicator-masked randomised superiority trial

Population

1,803 Korean adults with AF, CHA2DS2-VASc 1 (men) or 2 (women)

Primary outcome

Stroke, systemic embolism, major bleeding or cardiovascular death at 24 months

Effect

0.5% vs 1.5%; difference −1.0 point (95% CI −2.0 to −0.1); HR 0.31 (0.10–0.94)

SINGLE-AF, published in NEJM on 28 August 2026, randomised 1,803 patients in South Korea with atrial fibrillation and an intermediate stroke risk, a CHA2DS2-VASc score of 1 in men or 2 in women, to a DOAC or no anticoagulation. Mean age was 60 and fewer than a quarter were women. The trial was open-label with masked outcome adjudication.

At 24 months the primary composite of stroke, systemic embolism, major bleeding or cardiovascular death occurred in 0.5% of the DOAC group and 1.5% without anticoagulation (HR 0.31). Stroke occurred in 0.3% and 1.1%. Major bleeding appeared similar and there were no cardiovascular deaths. Serious adverse events were about equal.

US and European guidelines already give anticoagulation a class IIa recommendation here, based largely on observational data; this is the first randomised trial behind it. Because major bleeding sits inside the composite, the result reads as net benefit. The caveats are real: only 17 events in total, a young East Asian population, and an absolute difference of one percentage point over two years. The decision remains a shared one, but the default in a patient with one risk factor beyond sex can now lean towards a DOAC.

• Consider offering a DOAC to men with a CHA2DS2-VASc of 1 and women with 2, and discuss the absolute benefit.
• Explain the size of the gain: about one fewer serious event per 100 patients over two years.
• Assess bleeding risk and correct modifiable factors before starting, as for higher-risk patients.
• Use the correct DOAC dose for age, weight and renal function; underdosing is common.

Why it matters

The class IIa grey zone now has randomised evidence behind anticoagulation rather than only cohort data.

Don't overread it

With 17 events in an open-label trial in younger Korean patients, the size of the benefit is uncertain.

The statistics, in plain English

The hazard ratio of 0.31 sounds large, but it rests on 4 events against 13. Its confidence interval (0.10 to 0.94) only just excludes no effect, so the true benefit could be much smaller. The absolute difference, one percentage point over two years, is the number to give a patient.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.