Apixaban and rivaroxaban cut serious events 69% in patients doctors usually leave off blood thinners

A randomized trial in South Korea found that apixaban or rivaroxaban cut the combined rate of stroke, systemic embolism, major bleeding and cardiovascular death by 69% in people with atrial fibrillation at intermediate stroke risk, a group that often goes without anticoagulants. The SINGLE-AF trial, presented at the European Society of Cardiology congress in Munich on August 28, 2026 and published in the New England Journal of Medicine, enrolled 1,803 patients and followed them for 24 months.

The 69% is a relative figure attached to very small absolute numbers, and the trial’s own investigators and outside cardiologists disagree about what it should change.

Who counts as intermediate risk, and what the trial gave them

Intermediate risk here means a CHA2DS2-VASc score of 1 in men or 2 in women, one point of stroke risk short of the threshold where anticoagulation is routinely advised. Patients were randomized 1:1 at 18 South Korean centers to apixaban 5 mg twice daily, rivaroxaban 20 mg once daily, or no anticoagulation, in an open-label design. According to the European Society of Cardiology’s press release, the study was funded by South Korea’s Ministry of Health and Welfare along with Samjin Pharmaceutical and Hanmi Pharmaceutical.

Mean age was 60, and 24% of participants were women, according to the American College of Cardiology’s summary. The patients were relatively young, and they sit in the zone where the balance between stroke prevention and bleeding risk is closest, which is why the question had stayed open. Nobody had run a randomized comparison of newer DOACs against no treatment in this group; Joung called the result the first such evidence from a randomized trial.

The numbers behind 69%

At 24 months, serious events occurred in 0.5% of patients on a direct oral anticoagulant, or DOAC, and 1.5% of those on nothing, a hazard ratio of 0.31 with a 95% confidence interval of 0.10 to 0.94 and a p-value of 0.028. The interval is wide, and its upper edge sits just under 1.0. That is a statistically significant result that is also a fragile one.

Ischemic stroke drove the difference: 0.1% on DOACs against 1.1% on control, which ACC reports as 3 cases versus 10. Major bleeding was 0.3% against 0.5%, so the anticoagulated group was not bled more in this trial. Professor Boyoung Joung of Yonsei University in Seoul, the principal investigator, said the study now supplies randomized trial evidence that intermediate-risk AF patients benefit from DOAC therapy “without increased major bleeding.”

Put in absolute terms, the DOAC group avoided one composite event for roughly every hundred patients treated for two years, a gap of a single percentage point.

Why cardiologists are not calling it a new standard

Rod Passman of Northwestern University’s Feinberg School of Medicine, who commented for TCTMD, was blunt: “This is not a game changer.” He pointed to the small number of events, to the fact that the population was exclusively South Korean, and to chronic kidney disease, a factor that pushes real-world anticoagulation decisions but was excluded. He calculated a number needed to treat of 200 to prevent one stroke, and added that committing to a blood thinner for life is “not a trivial decision,” pointing to patients who give up skiing or mountain biking to avoid bleeds.

Paulus Kirchhof, writing in an accompanying commentary, acknowledged the limitations while suggesting anticoagulation in patients with atrial fibrillation and a single risk factor “may be encouraged,” though “in patients with low AFib burden, no therapy may be preferred.” The investigators themselves urged caution given event rates lower than expected and a modest sample size.

What current guidelines already say

The premise that doctors usually leave these patients untreated is a simplification. The 2023 ACC/AHA/ACCP/HRS atrial fibrillation guideline moved away from rigid score cutoffs. It recommends decisions based on yearly thromboembolic risk, and for patients at an intermediate annual stroke risk of about 1% to 2% it points clinicians toward AF burden, sex and blood pressure control as tie-breakers in a shared decision.

SINGLE-AF supplies randomized evidence for one side of that conversation: that apixaban or rivaroxaban lowered serious events over 24 months without a visible rise in major bleeding. The counterweight, in the words of the outside commentators, is how few events there were to learn from. The ScienceDaily repost of September 25 adds no new data to the August results. The trial does not say which patients within the intermediate band gain most, so the choice between a small absolute benefit and lifelong medication remains an individual one. What would settle it is a larger, more diverse trial with enough events to narrow that wide confidence interval, and no such study has been announced in the sources reviewed. Until then, the 0.5% against 1.5% split from 1,803 South Korean patients is the best randomized evidence there is.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview