EEG in silent small vessel disease: sLORETA mapping reveals cortical sources of vascular cognitive impairment no dementia in the default mode network

INTRODUCTION: Vascular cognitive impairment, no dementia (vCIND) is a prevalent and potentially preventable disorder. Clinical presof the small vessel subcortical subtype may be insidious and difficult to diagnose in the initial stage. We investigated electroencephalographic sources of subcortical vCIND in comparison to amnesic multidomain mild cognitive impairment (amdMCI) to determine the additional diagnostic value of quantitative electroencephalograhy (EEG) in this setting.

METHODS: Fifty-seven community residing patients with an uneventful central neurological history and first presentation of cognitive decline without dementia were included, 35 patients were diagnosed with vCIND and 22 with amdMCI. A cognitive control group, deliberately recruited from a cerebrovascular impaired cohort, consisted of cognitively healthy participants who experienced a fully recovered first ever transient ischemic attack (TIA) without clinical or magnetic resonance imaging evidence of stroke. From standard EEGs, the differences in standardized low-resolution brain electromagnetic tomography (sLORETA) sources were determined for the discrete frequency ranges 1-4 (delta), 4-8 (theta), 8-10.5 (alpha1), 10.5-13 (alpha2), 13-22 (beta1), and 22-30 (beta2) Hz.

RESULTS: In vCIND, a statistically significant decrease in parietooccipital alpha1 relative power current density compared with TIA and mild cognitive impairment patients was found. There was a significant decrease in frontal and parietooccipital beta1 relative power current density in vCIND compared with TIA patients. A significant increase in (pre) frontal delta relative power current density in vCIND compared with amdMCI was found as well. In amdMCI, delta relative power current density was significantly increased in the core limbic system.

DISCUSSION: Cortical sources of abnormal EEG activity in regions implicated in the default mode network are revealed by sLORETA at an early stage in vascular cognitive impairment. Mapping of parietooccipital alpha1, frontoparietooccipital beta1 and (pre) frontal delta loci in vCIND may reflect early executive and visuospatial dysfunction in this cohort. Standard EEG with sLORETA mapping might be an additional, noninvasive, and cost-effective tool in the diagnostic workup of patients presenting with a cognitive decline.