Regional cerebral oxygen saturation metrics for predicting postoperative delirium in elderly patients undergoing non-cardiac, non-neurological surgery: a propensity-matched cohort study
摘要
Regional cerebral oxygen saturation reflects cerebral oxygenation, but it remains unclear which cerebral oxygenation metric is most strongly associated with postoperative delirium during non-cardiac surgery.
MethodsThis retrospective observational cohort study analyzed patients aged ≥ 65 years undergoing non-cardiac, non-neurosurgical procedures under general anesthesia between January 2020 and December 2024. Baseline regional cerebral oxygen saturation, the absolute lowest intraoperative value, and the percentage decrease from baseline were evaluated. Postoperative delirium was assessed within 7 days after surgery using the Confusion Assessment Method. Propensity score matching and multivariable analyses were performed to assess discriminatory performance.
ResultsPostoperative delirium occurred in 18% of patients. In ROC analysis, baseline rcSO₂ (AUC 0.78; 95% CI 0.71–0.85; P < 0.001), absolute lowest rcSO₂ (AUC 0.74; 95% CI 0.67–0.81; P = 0.002), and rate of rcSO₂ decrease (AUC 0.70; 95% CI 0.63–0.77; P = 0.005) were associated with POD. In conditional logistic regression, lower baseline rcSO₂ (OR 1.9 per 10% decrease; 95% CI 1.4–2.6; P < 0.001), lower absolute lowest rcSO₂ (OR 1.6; 95% CI 1.2–2.2; P = 0.003), and higher rate of rcSO₂ decrease (OR 1.4 per 10% increase; 95% CI 1.1–1.8; P = 0.008) were associated with POD. In mediation analysis, the indirect effect via hemoglobin was ACME − 0.12 (bootstrapped 95% CI − 0.18 to − 0.06).
ConclusionLower baseline regional cerebral oxygen saturation is most strongly associated with postoperative delirium in older patients undergoing non-cardiac surgery, while intraoperative declines are influenced by potentially modifiable perioperative factors.