Purpose: We sought to compare 1-year outcomes in atrial fibrillation (AF) patients with versus without a previous stroke or transient ischaemic attack (TIA) using data from the prospective international GARFIELD Registry.

Methods: The population comprised consecutively recruited adult patients with newly diagnosed (≤6 weeks) non-valvular AF and ≥1 additional investigator-defined stroke risk factor(s). The effect of previous stroke/TIA was investigated using a Cox proportional hazards model, adjusting for age, sex, heart failure/left ventricular dysfunction, hypertension, diabetes, vascular disease and antithrombotic treatments.

Results: Of 10,608 AF patients, 14% had a history of stroke/TIA. These patients had a higher cardiovascular risk profile vs patients without a history of stroke/TIA: mean±SD CHADS2 3.7±0.9 vs 1.6±0.9; CHA2DS2-VASc 5.2±1.4 vs 2.9±1.4; HAS-BLED 1.7±0.9 vs 0.9±0.7; all p<0.001. Regarding oral antithrombotic treatments initiated at AF diagnosis, patients with a previous stroke/TIA had a higher rate of vitamin K antagonist use (63% vs 55%), a lower rate of antiplatelets alone use (22% vs 26%) and fewer received no antithrombotic treatment (9% vs 15%) (overall p<0.001 for treatments). After adjustment, previous stroke/TIA was associated with a higher risk of stroke, death, and intracranial haemorrhage, but not overall major bleeds, at 1 year (table).

Adjusted HRs for events at 1 year*

EventAdjusted HR (95% CI)P value
Stroke/systemic embolism3.13 (2.13–4.61)<0.001
Death (all cause)1.39 (1.01–1.93)0.047
Major bleed0.83 (0.39–1.77)0.63
Intracranial haemorrhage3.08 (1.09–8.65)0.033
EventAdjusted HR (95% CI)P value
Stroke/systemic embolism3.13 (2.13–4.61)<0.001
Death (all cause)1.39 (1.01–1.93)0.047
Major bleed0.83 (0.39–1.77)0.63
Intracranial haemorrhage3.08 (1.09–8.65)0.033

*For patients with vs without previous stroke/TIA. CI, confidence interval; HR, hazard ratio.

Conclusion: In everyday clinical practice, AF patients with a history of stroke/TIA are at greater risk of fatal and non-fatal events within the first year of diagnosis than those without a history of stroke/TIA.

This content is only available as a PDF.
You do not currently have access to this article.