Standard-dose DOACs, not dose reduction, serve Asian patients with AF

Design

Prespecified patient-level meta-analysis of four randomised DOAC trials (COMBINE AF)

Population

10,212 Asian and 61,471 non-Asian patients with atrial fibrillation

Primary outcome

Stroke/systemic embolism, major bleeding, and net clinical outcome vs warfarin

Effect

Standard-dose DOAC stroke/SEE HR 0.65 (Asian) vs 0.86 (non-Asian); low-dose vs standard HR 1.57 (1.15–2.13) in Asians

COMBINE AF pooled patient-level data from the four pivotal DOAC-versus-warfarin trials in atrial fibrillation, comparing 10,212 patients of Asian ancestry with 61,471 non-Asian patients. Asian patients were on average younger and lighter, with worse renal function and more prior stroke, and in the warfarin arm spent less time in therapeutic range (57.7% vs 66.2%).

Against warfarin, standard-dose DOACs reduced events more in Asian than non-Asian patients: stroke or systemic embolism (hazard ratio 0.65 vs 0.86), major bleeding (0.62 vs 0.91) and the net clinical outcome (0.76 vs 0.94), with interaction P<0.02 for each. The gastrointestinal bleeding excess seen with standard-dose DOACs in non-Asian patients (hazard ratio 1.41) was absent in Asian patients (0.92). Crucially, lower-dose DOACs raised stroke or embolism in Asian patients (hazard ratio 1.57, 95% CI 1.15–2.13) compared with standard dose.

The message cuts against a common reflex. In lighter, older Indian patients the temptation is to reduce the DOAC dose; here that choice raised strokes. Use the standard dose unless the drug's own label criteria for reduction are met.

• Prefer a standard-dose DOAC over warfarin in Asian patients with atrial fibrillation.
• Reduce the dose only when the specific drug's label criteria are met, not for low weight or age alone.
• Lower-dose DOACs raised stroke or embolism by about 57% versus standard dose in Asian patients.
• Standard-dose DOACs did not increase gastrointestinal bleeding in Asian patients, unlike in non-Asian patients.
• The benefit held across the range of body weight and renal function studied.

Why it matters

It challenges reflexive dose reduction in light or older Asian patients, which this analysis links to more strokes.

Don't overread it

The Asian versus non-Asian comparison is observational within the pooled trials, not a randomised contrast — an interaction signal, not proof of a mechanism.

The statistics, in plain English

Interaction P-values below 0.02 mean the ancestry difference is unlikely to be chance; a hazard ratio of 1.57 for lower-dose DOACs sits above 1.0 with its interval (1.15–2.13), so the harm from under-dosing is statistically real, not borderline.

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.