Early rhythm control after AF-related stroke was linked to fewer recurrences

Design

Nationwide claims-based cohort with propensity overlap weighting and an 8-week landmark

Population

14,355 adults in Japan with ischaemic stroke and AF (median age 83); 268 received early rhythm control

Primary outcome

Admission for recurrent ischaemic stroke or systemic embolism

Effect

HR 0.56 (0.32–0.97); 5-year incidence 8.0% vs 13.8%; mortality HR 0.72 (0.53–0.97)

A nationwide Japanese claims cohort, published in Neurology on 21 October, followed 14,355 adults admitted with ischaemic stroke and atrial fibrillation who survived event-free to eight weeks. Median age was 83. Only 268 (1.9%) started rhythm control within eight weeks, using antiarrhythmic drugs, ablation or cardioversion.

After weighting, early rhythm control was associated with a lower risk of recurrent stroke or systemic embolism (HR 0.56; 5-year incidence 8.0% vs 13.8%) and lower all-cause mortality (HR 0.72). Bleeding and heart failure admissions did not differ significantly.

Early rhythm control already reduced cardiovascular events in AF populations in trials such as EAST-AFNET 4. This study asks whether the same holds after a stroke. The answer is encouraging but fragile: very few patients were treated, and those chosen for rhythm control were probably fitter in ways claims data cannot capture.

It is a reason to involve cardiology early after AF-related stroke, not yet a reason to start rhythm control in everyone.

• Involve cardiology early after AF-related stroke to discuss whether rhythm control is suitable.
• Anticoagulation remains the cornerstone of secondary prevention; rhythm control would be added, not substituted.
• Fitter patients are the ones most likely to have been selected for rhythm control in this cohort.
• Watch for randomised trial results before changing standard pathways.

Why it matters

It raises whether stroke units should treat the rhythm, not just the clotting risk.

Don't overread it

This is observational with only 268 treated patients; selection of fitter patients may explain much of the benefit.

The statistics, in plain English

The hazard ratio of 0.56 (95% CI 0.32 to 0.97) has an upper limit close to 1, and the 5-year risk difference of −5.9 points has an interval (−11.0 to 0.8) that crosses zero. Both point to a possible benefit that the data cannot pin down.

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.