Atrial Fibrillation and Dementia

A Report From the AF-SCREEN International Collaboration

Léna Rivard, MD, MSc; Leif Friberg, MD, PhD; David Conen, MD, MPH; Jeffrey S. Healey, MD, MSc; Trygve Berge, MD, PhD; Giuseppe Boriani, MD, PhD; Axel Brandes, MD, DMSc; Hugh Calkins, MD; A. John Camm, MD; Lin Yee Chen, MD; Josep Lluis Clua Espuny, MD, PhD; Ronan Collins, MD; Stuart Connolly, MD; Nikolaos Dagres, MD; Mitchell S.V. Elkind, MD, MSc; Johan Engdahl, MD, PhD; Thalia S. Field, MD, MHSc; Bernard J. Gersh, MB, ChB, DPhil; Taya V. Glotzer, MD; Graeme J. Hankey, MD; Joseph A. Harbison, MD, FD; Karl Georg Haeusler, MD; Mellanie T. Hills, BSc; Linda S.B. Johnson, MD, PhD; Boyoung Joung, MD, PhD; Paul Khairy, MD, PhD; Paulus Kirchhof, MD; Derk Krieger, MD, PhD; Gregory Y.H. Lip, MD; Maja-Lisa Løchen, MD, PhD; Malini Madhavan, MBBS; Georges H. Mairesse, MD; Joan Montaner, MD, PhD; George Ntaios, MD, MSc; Terence J. Quinn, MD; Michiel Rienstra, MD, PhD; Mårten Rosenqvist, MD, PhD; Roopinder K. Sandhu, MD, MPH; Breda Smyth, MD; Renate B. Schnabel, MD, MSc; Stavros Stavrakis, MD, PhD; Sakis Themistoclakis, MD; Isabelle C. Van Gelder, MD, PhD; Ji-Guang Wang, MD, PhD; Ben Freedman, MBBS, PhD


Circulation. 2022;145(5):392-40. 

In This Article

Abstract and Introduction


Growing evidence suggests a consistent association between atrial fibrillation (AF) and cognitive impairment and dementia that is independent of clinical stroke. This report from the AF-SCREEN International Collaboration summarizes the evidence linking AF to cognitive impairment and dementia. It provides guidance on the investigation and management of dementia in patients with AF on the basis of best available evidence. The document also addresses suspected pathophysiologic mechanisms and identifies knowledge gaps for future research. Whereas AF and dementia share numerous risk factors, the association appears to be independent of these variables. Nevertheless, the evidence remains inconclusive regarding a direct causal effect. Several pathophysiologic mechanisms have been proposed, some of which are potentially amenable to early intervention, including cerebral microinfarction, AF-related cerebral hypoperfusion, inflammation, microhemorrhage, brain atrophy, and systemic atherosclerotic vascular disease. The mitigating role of oral anticoagulation in specific subgroups (eg, low stroke risk, short duration or silent AF, after successful AF ablation, or atrial cardiopathy) and the effect of rhythm versus rate control strategies remain unknown. Likewise, screening for AF (in cognitively normal or cognitively impaired patients) and screening for cognitive impairment in patients with AF are debated. The pathophysiology of dementia and therapeutic strategies to reduce cognitive impairment warrant further investigation in individuals with AF. Cognition should be evaluated in future AF studies and integrated with patient-specific outcome priorities and patient preferences. Further large-scale prospective studies and randomized trials are needed to establish whether AF is a risk factor for cognitive impairment, to investigate strategies to prevent dementia, and to determine whether screening for unknown AF followed by targeted therapy might prevent or reduce cognitive impairment and dementia.


Individuals with atrial fibrillation (AF) are at increased risk of cognitive impairment and dementia.[1–10] Whether the link is causal is an unanswered question. The prevalence of both dementia and AF is expected to increase with population aging worldwide. Projections indicate that the number of individuals with AF will increase by 150% in the next 4 decades and that the incidence of dementia will double with every 5.9-year increase in age, reaching >75 million people worldwide by 2030 and >135 million by 2050.[11,12] AF and dementia will therefore exert an increasing health and economic toll.

The AF-SCREEN International Collaboration was founded in 2015 with the purpose of promoting discussion and research about screening for unknown or undertreated AF to reduce stroke and death and to advocate for implementation of country-specific AF screening programs ( The collaboration includes >170 physicians (cardiologists, electrophysiologists, primary care physicians, stroke neurologists, and geriatricians), nurses, allied health professionals, epidemiologists, health economists, and patient group representatives from 37 countries. Between 2020 and 2021, 46 expert members of the AF-SCREEN International Collaboration prepared a document outlining the current knowledge of AF and dementia. In September 2020, 90 members (including the 46 members of the writing committee) discussed the draft document and critical gaps at the Virtual AF-SCREEN International Collaboration Meeting. Key points were determined using a Delphi process and retained if agreement was >85%. Complete details of the online survey results from 115 members are provided in Supplemental Material 1.

Numerous observational studies over the past 10 years, including several meta-analyses, provide growing evidence that AF is associated with cognitive impairment and dementia, even in the absence of clinically overt previous stroke[2–10,13] (Table S1). However, many studies included in the meta-analyses are small, are cross-sectional, have short follow-up, or were conducted in highly selected populations (eg, primarily the White population or hospitalized patients). Other important issues pertain to the tools used to assess both cognition and AF and the variability of risk factors over time. Several observational studies have considered the presence of risk factors at baseline without considering the emergence of or change in risk factors during follow-up in a time-dependent fashion. Therefore, direct comparison between studies is difficult. Furthermore, the relationship between AF and dementia is complex, because they share epidemiologic similarities and several risk factors such as advanced age, arterial hypertension, diabetes, hyperlipidemia, sleep apnea, coronary artery disease, heart failure, chronic kidney disease, obesity, physical inactivity, and excessive alcohol consumption.[14,15] It is therefore pertinent to question whether the association between AF and cognitive impairment/dementia could be explained by shared pathophysiology or whether AF is implicated in the causal pathway (Table 1). The association between AF and cognitive impairment/dementia seems to persist even after adjusting for known risk factors[6,10,16,17] and is stronger in younger patients compared with older patients with a higher burden of shared risk factors, which would not be expected if the association was purely attributable to confounding.[7] Two other factors favoring causality are temporality (ie, AF preceding cognitive decline) and a biological gradient between AF burden (ie, time since diagnosis and proportion of time spent in AF, discussed later) and cognitive impairment.[7,8,16,18] The objective of this report is to review the current understanding of the relationship between AF and cognitive impairment/dementia, treatment, and potential value of early AF detection by screening. Our aim is to identify knowledge gaps to focus research efforts to prevent or delay the onset of cognitive impairment/dementia associated with AF.