Authors: Van der A J et al.
Anesthesiology, March 9, 2026, 10.1097/ALN.0000000000006026
This prospective multicenter cohort study investigated whether quantitative electroencephalographic (EEG) patterns exist before or after postoperative delirium (POD) in older surgical patients. Postoperative delirium is a common complication after major surgery in elderly patients and is associated with increased morbidity, prolonged hospitalization, and long-term cognitive decline. Although EEG abnormalities are well documented during the acute delirium episode, it has remained unclear whether identifiable neurophysiologic markers exist before delirium develops or whether persistent EEG changes remain months afterward.
The investigators enrolled patients aged 65 years or older undergoing major elective surgery who had no prior cognitive impairment. Resting-state EEG recordings were obtained before surgery and again three months after surgery. Non-surgical control participants were also enrolled to account for potential age-related EEG changes unrelated to surgery. Quantitative EEG analyses focused on measures known to change during delirium, including spectral power, functional connectivity, spectral variability, and signal complexity.
Among 379 enrolled surgical patients, 330 had EEG recordings of sufficient quality for analysis. Postoperative delirium developed in 59 patients, representing approximately 18% of the cohort. Fifty-seven non-surgical participants served as controls.
The most important finding was the identification of a preoperative EEG difference in patients who later developed delirium. Specifically, these patients had significantly lower beta-band amplitude-based functional connectivity before surgery compared with patients who did not develop POD. This suggests that certain patterns of baseline neural network organization may reflect a vulnerability to delirium.
Beyond this single measure, however, no other baseline EEG metrics differed significantly between groups. Measures such as relative power, phase-based connectivity, spectral variability, and signal complexity did not distinguish future delirium patients from others.
Equally notable were the findings at the three-month follow-up. The investigators found no persistent EEG abnormalities associated with either surgery or postoperative delirium. Quantitative EEG characteristics at three months were similar across groups, including patients who had experienced delirium and those who had not. This suggests that resting-state brain network function may recover or compensate over time after an episode of delirium.
These findings are important because they challenge the assumption that postoperative delirium necessarily produces lasting electrophysiologic network disruption detectable by resting-state EEG months later. Instead, the results suggest that pre-existing vulnerability may play a larger role than long-term structural or functional brain injury in many patients.
The authors conclude that reduced beta-band connectivity may represent a neurophysiologic vulnerability marker for postoperative delirium, but further studies are needed to confirm its predictive value and clinical utility.
What You Should Know
Postoperative delirium occurs in roughly 15% to 25% of older patients undergoing major surgery and is linked to worse outcomes and long-term cognitive decline.
This study suggests that subtle preoperative brain network differences may predispose certain patients to delirium.
Reduced beta-band functional connectivity before surgery was the only EEG measure associated with later delirium.
Importantly, the study found no persistent EEG abnormalities three months after surgery, suggesting that brain network activity may recover after delirium.
The findings support the concept that delirium vulnerability may be driven more by preexisting neurophysiologic susceptibility than by permanent postoperative brain injury.
Key Points
Prospective multicenter study of 330 surgical patients aged 65 years or older.
Postoperative delirium occurred in 18% of patients.
Lower preoperative beta-band amplitude-based connectivity was associated with later delirium.
No other baseline EEG measures predicted POD.
At three months after surgery, no persistent EEG changes were observed in patients who developed delirium.
The findings suggest that delirium may reflect preexisting neural vulnerability rather than lasting electrophysiologic injury.
Thank you to Anesthesiology for allowing us to summarize this article.