Atrial Fibrillation and Risk of Stroke

A Nationwide Cohort Study

Christine Benn Christiansen; Thomas A. Gerds; Jonas Bjerring Olesen; Søren Lund Kristensen; Morten Lamberts; Gregory Y.H. Lip; Gunnar H. Gislason; Lars Køber; Christian Torp-Pedersen


Europace. 2016;18(11):1689-1697. 

In This Article

Abstract and Introduction


Aim Although the relation between stroke risk factors and stroke in patients with atrial fibrillation (AF) has been extensively examined, only few studies have explored the association of AF and the risk of ischaemic stroke/systemic thromboembolism/transient ischaemic attack (stroke/TE/TIA) in the presence of concomitant stroke risk factors.

Methods and results From nationwide registries, all persons who turned 50, 60, 70, or 80 from 1997 to 2011 were identified. Persons receiving warfarin were excluded. The absolute risk of stroke/TE/TIA was reported for a 5-year period, as was the absolute risk ratios for AF vs. no AF according to prior stroke and the number of additional risk factors. The study cohort comprised of 3 076 355 persons without AF and 48 189 with AF. For men aged 50 years, with no risk factors, the 5-year risk of stroke was 1.1% (95% confidence interval 1.1–1.1); with AF alone 2.5% (1.8–3.2); with one risk factor and no prior stroke or AF 2.5% (2.3–2.7); and with one factor, no prior stroke and AF 2.9% (1.4–4.3). In men aged 50 years with prior stroke as the only risk factor, 5-year risk was 10.2% (9.1–11.3). In men aged 70 years, the corresponding risks were 4.8% (4.7–4.9), 6.8% (5.7–7.9), 6.6% (6.3–6.8), 8.7 (7.4–9.9), and 19.1% (18.1–20.1), respectively. In women aged 50 years, the risk was of 0.7% (0.7–0.7), 2.1% (0.9–3.2), 1.6% (1.4–1.8), 4.1% (0.6–7.6), and 7.2% (6.3–8.2), respectively, and in women aged 70 years 3.4% (3.3–3.5), 8.2% (7.0–9.5), 4.6% (4.4–4.8), 9.1% (7.5–10.6), and 15.4% (14.5–16.4), respectively.

Conclusions Stroke/TE/TIA risk was particularly increased when prior stroke/TE/TIA was present. Atrial fibrillation is associated with an increase in risk of stroke/TE/TIA in the absence of other risk factors but only a moderate increase in risk when other risk factors are present.


Atrial fibrillation (AF) is a common and well-established risk factor for ischaemic stroke, systemic thromboembolism (TE), and transient ischaemic attack (TIA).[1] Other aetiologies for ischaemic strokes are large-artery atherosclerosis, small-vessel occlusion, stroke of other determined aetiology, and stroke of undetermined aetiology as defined by the TOAST (Trial of Org 10 172 in Acute Stroke Treatment) criteria.[2] It has been estimated that cardioembolism is the source of stroke in 16–30% of cases of ischaemic strokes.[3,4]

Atrial fibrillation detection is important in stroke prevention, because anticoagulation is known to decrease stroke risk in the presence of AF.[5] It has also been demonstrated that platelet inhibition is not a replacement for proper anticoagulation in the presence of AF.[6] In persons without AF, antithrombotic stroke prevention, if any, consists of antiplatelet therapy.[7] About 80% of ischaemic strokes occur in persons without AF, and it is therefore important to develop a path to examine the optimal prevention of stroke when there is no obvious AF.

Anticoagulation in AF is guided by risk scores including the CHA2DS2-VASc, which combines heart failure, hypertension, two age strata, diabetes, prior stroke, vascular disease, and gender. The presence of at least one of these risk factors merits consideration of oral anticoagulation therapy.[8] One key question to examine is whether it is the components of the risk score that determines risk of stroke or whether these components are only important in the setting of AF. If AF appears to be a critical factor for the risk of stroke, anticoagulation should focus on patients with AF. If the risk of stroke depends on risk factors independently of AF, anticoagulation in patients at high risk of stroke could be appropriate in spite of no apparent AF.

To investigate how much concomitant AF changes the risk of ischaemic stroke/systemic thromboembolism/transient ischaemic attack (stroke/TE/TIA), we performed a nationwide study based on Danish registries of the 5-year absolute risk of stroke. We report the absolute risks of stroke/TE/TIA within the 5-year period following one of the landmark birthdays of 50, 60, 70, and 80 in subgroups defined by the established risk factors that are included in CHA2DS2-VASc.