Clinical Predictors of Emergence Delirium in Children: A Prospective Cohort Study

Abstract Objective. To determine the incidence and identify independent clinical predictors of emergence delirium (ED) in children aged 2-12 years. Material and methods. A prospective observational study included 56 children aged 2-12 years undergoing elective surgery under general anaesthesia. Preoperative anxiety (m-YPAS), induction behaviour (4 point scale), anaesthesia duration, opioid use, and postoperative pain (FLACC) were assessed. ED was diagnosed when the maximum PAED score was [≥]12. Results. The incidence of ED was 55.4% (31/56). Univariate analysis with false discovery rate (FDR) correction identified significant associations with ED for anaesthesia duration (q=0.002), induction behaviour (q=0.007), and surgery type (q=0.027). Multivariable logistic regression revealed three independent predictors: induction behaviour (category 3 vs 1) - odds ratio (OR) 14.2 (95% CI 2.6-78.1); anaesthesia duration (per minute) - OR 1.07 (95% CI 1.02-1.13); opioid use - OR 12.1 (95% CI 1.3-113.0). The model showed good discriminatory ability: area under the ROC curve (AUC) = 0.83 (95% CI 0.72-0.94). Conclusion. Emergence delirium in children aged 2-12 years without pharmacological premedication occurs in 55.4% of cases. The strongest independent predictors are adverse induction behaviour, longer anaesthesia duration, and intraoperative opioid use. The derived model can be used for personalised risk stratification of ED. Keywords: emergence delirium; children; risk factors; PAED; prediction model.