Quantitative EEG as a biomarker in mild cognitive impairment with Lewy bodies

Abstract

Objectives: To investigate using quantitative EEG the (1) differences between patients with mild cognitive impairment with Lewy bodies (MCI-LB) and MCI with Alzheimer's disease (MCI-AD) and (2) its utility as a potential biomarker for early differential diagnosis.

Methods: We analyzed eyes-closed, resting-state, high-density EEG data from highly phenotyped participants (39 MCI-LB, 36 MCI-AD, and 31 healthy controls). EEG measures included spectral power in different frequency bands (delta, theta, pre-alpha, alpha, and beta), theta/alpha ratio, dominant frequency, and dominant frequency variability. Receiver operating characteristic (ROC) analyses were performed to assess diagnostic accuracy.

Results: There was a shift in power from beta and alpha frequency bands towards slower frequencies in the pre-alpha and theta range in MCI-LB compared to healthy controls. Additionally, the dominant frequency was slower in MCI-LB compared to controls. We found significantly increased pre-alpha power, decreased beta power, and slower dominant frequency in MCI-LB compared to MCI-AD. EEG abnormalities were more apparent in MCI-LB cases with more diagnostic features. There were no significant differences between MCI-AD and controls. In the ROC analysis to distinguish MCI-LB from MCI-AD, beta power and dominant frequency showed the highest area under the curve values of 0.71 and 0.70, respectively. While specificity was high for some measures (up to 0.97 for alpha power and 0.94 for theta/alpha ratio), sensitivity was generally much lower.

Conclusions: Early EEG slowing is a specific feature of MCI-LB compared to MCI-AD. However, there is an overlap between the two MCI groups which makes it difficult to distinguish between them based on EEG alone.

Keywords: Alzheimer’s disease; Biomarker; Dementia with Lewy bodies; Quantitative electroencephalography.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

Study flowchart

Fig. 2

Mean power spectra for the three diagnostic groups. Shaded areas indicate standard errors. HC, healthy controls; MCI-AD, mild cognitive impairment with Alzheimer’s disease; MCI-LB, probable mild cognitive impairment with Lewy bodies

Fig. 3

Group comparison of quantitative EEG characteristics. In each boxplot, the central line corresponds to the sample median; the upper and lower border of the box represent the 25th and 75th percentile, respectively; and the length of the whiskers is 1.5 times the interquartile range. Corresponding results from statistical comparisons between the groups are presented in Table 2. DF, dominant frequency; DFV, dominant frequency variability; HC, healthy controls; MCI-AD, mild cognitive impairment with Alzheimer’s disease; MCI-LB, probable mild cognitive impairment with Lewy bodies

Fig. 4

Comparison of quantitative EEG characteristics between patients with and without visual hallucinations. a Comparison of alpha power and b theta/alpha ratio between MCI-LB patients with visual hallucinations (VH+, N = 9) and without visual hallucinations (VH−, N = 30). In each boxplot, the central line corresponds to the sample median; the upper and lower border of the box represent the 25th and 75th percentile, respectively; and the length of the whiskers is 1.5 times the interquartile range