Electroencephalographic Upper/Low Alpha Frequency Power Ratio Relates to Cortex Thinning in Mild Cognitive Impairment

<b><i>Objective:</i></b> Temporoparietal cortex thinning is associated with mild cognitive impairment (MCI) due to Alzheimer disease (AD). The increase in EEG upper/low α frequency power ratio has been associated with AD converter MCI subjects. We investigated the association of the EEG upper/low α frequency power ratio with patterns of cortical thickness in MCI. <b><i>Methods:</i></b> 74 adult subjects with MCI underwent clinical and neuropsychological evaluation, electroencephalography (EEG) recording and high-resolution 3-dimensional magnetic resonance imaging (MRI). The EEG upper/low α frequency power ratio as well as cortical thickness were computed for each subject. Three MCI groups were detected according to increasing tertile values of EEG upper/low α frequency power ratios, and the difference of cortical thickness among the groups was estimated. <b><i>Results:</i></b> The EEG high upper/low α frequency power ratio group had a total cortical grey matter volume reduction of 471 mm², greater than that of the EEG low upper/low α frequency power ratio group (p < 0.001). The EEG high upper/low α frequency power ratio group showed a similar but less marked pattern (160 mm²) of cortical thinning when compared to the EEG middle upper/low α frequency power ratio group (p < 0.001). Moreover, the EEG high upper/low α frequency power ratio group had wider cortical thinning than other groups, mapped to the supramarginal gyrus and precuneus bilaterally. No significant regional cortical thickness differences were found between middle and low EEG upper/low α frequency power ratio groups. <b><i>Conclusion:</i></b> A high EEG upper/low α frequency power ratio was associated with temporoparietal cortical thinning in MCI subjects. The combination of upper/low α frequency power ratio and cortical thickness measurement could be useful for identifying individuals at risk for progression to AD dementia and may be of value in the clinical context.