P3704Silent ischemic brain lesion detected in patients with subclinical paroxysmal atrial fibrillation

Abstract

Abstract Background Long-lasting atrial fibrillation (AF) has been reported to be associated with an increased risk of dementia, independent of clinical stroke. However, the mechanisms or association in patients with subclinical paroxysmal AF (S-PAF) remain unclear. We evaluated whether S-PAF is associated with silent ischemic brain lesion (S-IBL), one of causes of dementia. Methods We studied 46 patients (35 male, 68±15 yrs) without a history of stroke/transient ischemic attack and AF, who implanted insertable cardiac monitoring (ICM) for unexplained syncope (n=33) or embolic stroke of undetermined source (ESUS) (n=13). All patients underwent cerebral magnetic resource imaging (c-MRI), and S-IBL was defined as infarction, lacuna and microbleeds. The lesions in an acute stage were excluded in patients with ESUS. Results AF was detected in 15/46 patients (11 with unexplained syncope and 4 with ESUS) during follow-up of 7.0±6.6 months, and S-IBL was observed in 18/46 patients (9 infarction, 8 lacuna, or 8 microbleeds). Univariate analysis revealed that higher prevalence of AF (61% vs. 14%, p=0.0015), elder age (73±10yrs vs. 65±16yrs, p=0.0445), dyslipidemia (67% vs. 25%, p=0.007), structural heart disease (44% vs. 14%, p=0.0383), and larger left atrium diameter (41±6 mm vs. 37±5 mm, p=0.0267) were related to S-IBL. On multivariate analysis, prevalence of AF was independently associated with S-IBL (p=0.0070, OR 13.4, 95% CI 1.945–155.813). When receiver-operating-characteristics (ROC) curve analysis and prevalence of AF were used to detect S-IBL, the area under the ROC curve was 0.7341 (sensitivity: 61.1%, specificity: 85.7%). Conclusion Subclinical paroxysmal AF is associated with silent ischemic brain lesion and might cause to dementia.