Retrospective case-control study of intraoperative EEG correlates in postoperative delirium

Objectives This retrospective study investigated the association between perioperative SedLine EEG parameters and the incidence of postoperative delirium (POD) in elderly patients undergoing breast cancer surgery. Methods A total of 80 elderly patients, aged 60 years or older, who underwent breast cancer surgery between January and December 2024, were included in the analysis. Patients were divided into the POD group and the NPOD group according to whether they developed postoperative delirium. Intraoperative EEG parameters captured by the SedLine monitor included the patient state index (PSI), spectral edge frequency (SEF), burst suppression (BS), and burst suppression ratio (BSR). Raw EEG recordings spanning the entire surgical procedure were preprocessed using EEGLAB. For spectral analysis, power spectral density (PSD) within each frequency band was computed by averaging across all frequency points in that band. Results Of the 80 patients, 18 (22.5%) developed POD, while 62 (77.5%) did not. Patients in the POD group had significantly lower preoperative MMSE scores and were less likely to have comorbid hypertension. No significant differences were observed in PSI values between the two groups. However, SEF values (both left and right hemispheres) were significantly lower in the POD group, which also showed a higher incidence of BS. Especially in BSR exceeded 10% for over 1 min. Multivariate logistic regression identified low MMSE scores, absence of hypertension, and decreased SEF-L/SEF-R as independent predictors of POD ( P 0.05). Further EEG spectral analysis conducted with MATLAB revealed significantly reduced α-band power (8–13 Hz) in the POD group, especially at frontal electrodes Fp1, Fp2, F7, and F8 ( P 0.05), along with diminished β-band power in the same regions. Conclusion Lower intraoperative SEF values and reduced α- and β-band EEG power were associated with an increased risk of POD. These EEG markers might serve as early predictors for identifying patients at high risk of developing POD.