EEG-Derived Index Predicts Postoperative Delirium in Elderly Patients With Hip Fracture: A Prospective Study From a Tertiary Medical Center.

INTRODUCTION: Postoperative delirium (POD) remains challenging to predict in elderly hip fracture patients. This study posited that a newly developed electroencephalogram-derived metric, termed the Delirium Index (DELi), has the potential to elucidate latent information pertaining to the preoperative predisposing factors of POD. METHODS: A prospective cohort of 144 elderly patients undergoing hip fracture surgery was enrolled. DELi scores were derived from preoperative electroencephalography (EEG) using wavelet analysis. POD was diagnosed using the Confusion Assessment Method (CAM). Predictive performance was evaluated via ROC analysis, logistic regression, and bootstrapping validation. RESULTS: DELi demonstrated strong correlation with CAM scores (r = 0.516, p < 0.0001) and good predictive accuracy for POD (AUC = 0.791, 95% CI: 0.715-0.867). A composite model integrating DELi, Montreal Cognitive Assessment (MoCA), and FRAIL scores achieved excellent discrimination (AUC = 0.922, 95% CI: 0.878-0.965), with 85.9% sensitivity and 86.3% specificity. CONCLUSIONS: The DELi index, combined with cognitive and frailty assessments, provides a practical tool for preoperative POD risk stratification in elderly hip fracture patients. TRIAL REGISTRATION: ChiCTR2200060389.