Abstract
AIMS: Atrial fibrillation (AF) is associated with cognitive decline, but the role of electroencephalography (EEG) in assessing cognitive dysfunction in AF patients is underexplored. OBJECTIVE: This study investigated the relationship between resting-state EEG patterns and cognitive impairment in AF patients. METHODS: We recruited 120 participants from the Affiliated Xuancheng Hospital, China (January 2023 to January 2024), categorizing them into healthy controls and AF patients. Resting-state EEG metrics, including power spectral density (PSD), functional connectivity (FC), cross-frequency coupling (CFC), and sample entropy (EnSA), were analyzed alongside the Montreal Cognitive Assessment (MoCA) scores. Mediation analysis explored EEG's role in the AF-cognitive decline relationship. RESULTS: AF patients had significantly lower MoCA scores. PSD analysis showed increased δ and θ and decreased α and β activity. FC was reduced in the α and β bands but increased in localized θ and γ bands. CFC analysis revealed elevated θ-β and θ-γ phase-amplitude coupling (PAC), reduced β-γ PAC, and lower EnSA. EEG metrics were significantly correlated with MoCA scores, with θ-β PAC mediating cognitive decline. CONCLUSION: AF patients exhibit distinctive EEG changes, with θ-β PAC mediating cognitive impairment, suggesting the potential of resting-state EEG for cognitive assessment in AF patients.