Abstract
BACKGROUND: Postoperative cognitive dysfunction (POCD) is a common complication in elderly surgical patients and has been associated with excessive anesthetic depth. Electroencephalogram (EEG)-guided anesthesia provides real-time cerebral monitoring (e.g., bispectral index [BIS]), but its effect on POCD remains inconclusive across randomized controlled trials (RCTs). METHODS: We conducted a systematic review and meta-analysis following PRISMA guidelines, searching PubMed, Embase, Cochrane Library, and Web of Science for RCTs evaluating EEG-guided anesthesia versus standard care in elderly surgical patients. Primary outcome was POCD incidence; secondary outcomes included cognitive scores across acute (1–7 days), subacute (1–3 months), and chronic (≥ 6 weeks) phases. Risk of bias was assessed using the Cochrane Tool. Pooled odds ratios (ORs) and standardized mean differences (SMDs) were calculated with fixed/random-effects models. Trial sequential analysis (TSA) and sensitivity analyses validated evidence robustness; funnel plots and Egger’s test evaluated publication bias. RESULTS: Ten RCTs (4,367 patients) were included. EEG-guided anesthesia significantly reduced POCD incidence by 22% (pooled OR = 0.78, 95% CI: 0.69–0.90, P < 0.001, I(2) = 0.0%), with TSA confirming conclusive evidence after reaching the required information size (3,437 patients) and crossing the efficacy boundary. Subacute follow-ups (1–3 months) showed improved verbal fluency (WMD = 1.2, P = 0.009) and delayed recall (WMD = 0.8, P = 0.03) in EEG-guided groups, primarily with BIS monitoring, while acute-phase scores were heterogeneous and long-term (≥ 6 weeks) global cognitive scores did not differ. Sensitivity analyses and funnel plots indicated no significant publication bias or result instability. Non-cardiac surgeries demonstrated consistent benefits, whereas cardiac surgery data were limited. CONCLUSIONS: Intraoperative EEG-guided anesthesia—particularly using BIS monitoring—reduces POCD incidence and improves subacute cognitive outcomes in elderly patients, likely by avoiding excessive anesthetic depth and optimizing hemodynamics. While long-term effects on global cognition remain unproven, these findings support EEG monitoring as a valuable adjunct in high-risk populations, particularly for major non-cardiac surgery. Standardized POCD assessment, personalized strategies, and long-term follow-ups are needed to refine clinical guidelines and understand persistent cognitive trajectories. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12871-025-03297-3.