Association between burst suppression and postoperative delirium in older cardiac surgery patients during fixed-dose propofol induction: a prospective observational study

Burst suppression is associated with postoperative delirium (POD) in older adults undergoing surgery. It is unclear whether burst suppression results from excessive anesthetic dosage or is predetermined by neurophysiological characteristics. We aimed to examine the association between burst suppression and POD following a fixed propofol induction. A prospective observational study was conducted in patients aged 65 and older undergoing cardiac surgery with cardiopulmonary bypass (CPB). Standardized induction with intravenous propofol (0.5 mg/kg) was performed. Frontal EEGs were recorded at baseline, post-induction, before CPB, and during CPB, and spectral estimates were calculated. The presence of POD was evaluated using the Confusion Assessment Method for the ICU and chart review. Burst suppression was quantified according to the isoelectric rate (peak absolute amplitude ≤ 5 µV for ≥ 1 s). Univariable and multivariable logistic regression, ROC curve generation, and AUC analyses were used. 50% of the patients developed POD (37/74). Patients presenting POD were older (75 (69–78) versus 71 (68–73) years, p = 0.012). There were no differences in spectral analyses between POD (+) and POD (-) groups (EEG-analytic cohort n = 62). Univariable logistic regression showed associations between age (OR = 1.15 (95% CI: 1.03–1.29, p = 0.01)) and burst suppression (OR = 1.23 (1.00-1.76, p = 0.046)) with POD at induction. However, in a multivariable model including burst suppression, age, MiniCog, and preoperative C-reactive protein, burst suppression was no longer a significant predictor of POD (OR = 1.17 (0.94–1.79), p = 0.15). Increased burst suppression following a fixed-dose propofol induction shows a hypothesis-generating association with a higher incidence of POD in older patients undergoing cardiac surgery with cardiopulmonary bypass. However, this association was not independent of other perioperative risk factors in multivariable analysis. Neurophysiological responses during anesthetic induction could serve as early markers of susceptibility to postoperative delirium. The trial was registered on clinicaltrials.gov prior to patient enrollment (NCT04713644, Date of registration: 01-19-2021).