Phase-Specific Electroencephalography Monitoring: A Dynamic Approach to Predict Postoperative Delirium During Aortic Surgery

Abstract

Background: Static or averaged electroencephalography (EEG) metrics may fail to capture dynamic cerebral changes during surgery. We assessed the EEG features during aortic arch surgery with cardiopulmonary bypass (CPB) and total circulatory arrest (TCA) to identify key EEG predictors of postoperative delirium (POD). Methods: This retrospective study analyzed intraoperative EEG data from 233 patients across 5 phases: pre-CPB, CPB initiation, TCA, post-TCA, and post-CPB. The predictive potential of EEG parameters was assessed using logistic regression, and phase-specific nomogram models were developed. The primary analysis included emergency cases; elective cases were included in sensitivity analyses. Results: POD occurred in 78 patients (44.8%). Phase-specific models showed high predictive performance. Independent predictors of POD included reduced alpha power during the post-CPB phase (odds ratio [OR]=0.76, 95% CI: 0.67-0.87, P<0.001) and lower delta power during TCA (OR=0.87, 95% CI: 0.79-0.96, P=0.031). In addition, elevated BSR during CPB initiation (OR=1.53, 95% CI: 1.24-1.89, P<0.001) and post-TCA (OR=1.37, 95% CI: 1.11-1.70, P=0.008) predicted increased POD incidence. A similar, but nonsignificant, observation for alpha power was observed during the pre-CPB phase (P=0.103). Calibration plots showed strong agreement between predicted and observed outcomes. Conclusions: Phase-specific EEG monitoring reliably predicted POD during aortic arch surgery. Predictive factors varied across intraoperative phases, underscoring the dynamic sensitivity of EEG to physiological changes. Comprehensive, phase-specific EEG assessment may improve risk stratification and perioperative management, though large prospective studies are needed to confirm these findings.