Perioperative dynamic EEG patterns correlated with changes in cognitive subdomains

Objective: Surgery and anesthesia can be associated with postoperative neurological injury, especially in individuals with diminished cognitive reserve. Timely identification of at-risk patients enables deployment of preventive strategies. Despite consistent recommendations, perioperative cognitive risk assessment remains largely overlooked. We explored which dynamic electroencephalogram (EEG) patterns are correlated with cognitive subdomains significantly affected by surgery and anesthesia. Methods: This was a secondary analysis of two prospective observational studies that included patients undergoing elective laparoscopic abdominal or pelvic surgery under general anesthesia. Cognition was assessed using the Mini-Mental State Exam, 2nd edition (MMSE-2) and the Montreal Cognitive Assessment (MoCA) within one week before and up to 48 h after surgery. Frontal EEG was recorded pre-induction to extubation. Cognitive subdomains and EEG changes were explored using linear regression and accuracy analyses. Results: < 0.001 AUC = 0.75 [0.6-0.89]) and processing speed (MMSE-2, p < 0.001 AUC = 0.85 [0.72-0.95]) were significantly affected postoperatively. Delta power at the end of maintenance, changes in theta, alpha power, aperiodic slope, offset were correlated with processing speed changes. Conclusion: Dynamic EEG patterns were correlated with processing speed changes. Integrating EEG into perioperative risk assessment may contribute to early identification of vulnerability. Significance: This study demonstrates the correlation between changes in perioperative EEG and cognitive subdomains.