DOACs after intracranial bleeding were associated with fewer strokes and deaths without more rebleeding

Design

Nationwide registry cohort with inverse probability weighting and competing-risk analysis

Population

1,502 Finnish adults with first atrial fibrillation diagnosis and previous intracranial bleeding

Primary outcome

Ischaemic stroke, recurrent intracranial bleeding and death from other causes

Effect

DOAC vs none: stroke HR 0.17 (0.04 to 0.71), death HR 0.20 (0.09 to 0.46), intracranial bleed HR 0.70 (0.19 to 2.50)

Using linked Finnish registries, researchers followed people diagnosed with atrial fibrillation at least four months after an intracranial bleed. They compared DOACs, warfarin and no anticoagulation, adjusting for baseline differences.

Compared with no anticoagulation, DOACs were associated with lower stroke and death and no increase in intracranial bleeding. Warfarin was associated with lower stroke and death too, but with more intracranial bleeding, unless time in therapeutic range was good.

Patients chosen for anticoagulation were probably healthier, which could exaggerate the benefit. Still, the large gaps in absolute risk support considering a DOAC rather than defaulting to none. Guidelines have not changed on the strength of this study.

• One-year ischaemic stroke was 0.7% on DOACs, 2.1% on warfarin and 4.1% on no anticoagulant.
• Intracranial rebleeding was not raised on DOACs (HR 0.70) but was on warfarin (HR 1.85).
• Warfarin with time in range of 70% or more performed far better than poorly controlled warfarin.
• 41% had traumatic bleeds only; the balance may differ after spontaneous haemorrhage.
• Ongoing randomised trials of anticoagulation after haemorrhage should guide future practice.

Why it matters

More than half of these patients received no anticoagulant, and they had the highest stroke and death rates.

Don't overread it

This is an observational registry study with limited detail on bleed type, so confounding by indication is likely.

The statistics, in plain English

The DOAC group was small (about 225 people), so the hazard ratios have wide intervals. The intracranial bleeding interval from 0.19 to 2.50 means a modest increase cannot be excluded.

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.