University of Twente Student Theses


Correlation of qEEG and cognitive impairment in cardiac arrest survivors: first steps towards a prediction model

Kruijf, N.L.M. de (2022) Correlation of qEEG and cognitive impairment in cardiac arrest survivors: first steps towards a prediction model.

[img] PDF
Abstract:Background: Half of the cardiac arrest survivors experience long-term cognitive impairment. Cognitive rehabilitation might have beneficial effects on cardiac arrest survivors. Therefore, early identification of patients at risk for developing long-term cognitive impairment is important. To date, early predictors for long-term cognitive functioning are not available yet. Here we study the potential role of early EEG. Objectives: Primary objective is to study the association between quantitative EEG (qEEG) measures and cognitive functioning in cardiac arrest survivors. Secondary objective is to examine the predictive value of early qEEG measures in addition to demographic and clinical parameters for predicting cognitive impairment during hospital admission. Methods: We performed a prospective longitudinal cohort study on cardiac arrest survivors in six Dutch hospitals. Resting-state nineteen-channel EEG was recorded during hospital admission. We selected eyes closed segments to calculate alpha-to-theta ratio, peak frequency, center of gravity, and global and local imaginary coherency. The primary outcome measure was Montreal Cognitive Assessment (MoCA) obtained during hospital admission. We assessed the correlations between the MoCA and qEEG measures with Pearson’s correlation coefficient. We performed multinomial logistic regression analyses to examine the added predictive value of qEEG measures for prediction of cognitive impairment (MoCA<26). Five-fold validation was used for model validation. Results: We included 77 cardiac arrest survivors, of whom 53 had cognitive impairment. We found a positive significant correlation between the alpha-to-theta ratio and the MoCA (r=0.60 p<0.01), as well as between the peak frequency and the MoCA (r=0.56 p<0.01). There was a negative significant correlation between the center of gravity and the MoCA (r=-0.48 p=<0.01). No significant correlation between the mean imaginary part of coherency and MoCA was found in the theta band (r=-0.09 p=0.43) and alpha band (r=0.09 p=0.41). The imaginary part of coherency differed significantly in some electrode pairs in the theta and alpha band between patients with and patients without cognitive impairment. The multinomial logistic regression analyses resulted in five models with different explanatory variables. The model with center of gravity as explanatory variable showed the highest sensitivity (92%) and specificity (50%). Conclusion: Lower Alpha-to-theta ratio, lower peak frequency, and higher center of gravity correlated significantly with poorer cognitive functioning during hospital admission in cardiac arrest survivors. These qEEG measures hold potential to add to prediction of cognitive functioning as individual explanatory variable or in combination with demographic and clinical parameters.
Item Type:Essay (Master)
Faculty:TNW: Science and Technology
Programme:Technical Medicine MSc (60033)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page