Objective: Incontrovertible disease markers are absent in delirium. This study investigated the usefulness of quantitative electroencephalography (qEEG) in diagnosing delirium.
Methods: This retrospective case-control study reviewed medical records and qEEG data of 69 age/sex-matched patients (delirium group, n=30; control group, n=39). The first minute of artifact-free EEG data with eyes closed was selected. Nineteen electrodes' sensitivity, specificity, and correlation with delirium rating scale-revised-98 were analyzed.
Results: On comparing the means of absolute power by frontal, central, and posterior regions, the delta and theta powers showed significant differences (p<0.001) in all regions, and the magnitude of the absolute power was higher in the delirium group than in the control group; only the posterior region showed a significant (p<0.001) difference in beta power. The spectral power of theta at the frontal region (area under the curve [AUC]=0.84) and theta at the central and posterior regions (AUC=0.83) showed 90% sensitivity and 79% specificity, respectively, in differentiating delirious patients and controls. The beta power of the central region showed a significant negative correlation with delirium severity (R=-0.457, p=0.011).
Conclusion: Power spectrum analysis of qEEG showed high accuracy in screening delirium among patients. The study suggests qEEG as a potential aid in diagnosing delirium.
Keywords: Biomarker; Delirium; Disease severity; Electroencephalography; Retrospective study.