SummaryBackground Bispectral index (BIS) and entropy monitors have been proposed for use in children, but research has not supported their validity for infants. However, effective monitoring of young children may be even more important than for adults, to aid appropriate anesthetic dosing and reduce the chance of adverse consequences. This prospective study aimed to investigate the relationships between age and the predictive performance of BIS and entropy monitors in measuring the anesthetic drug effects within a pediatric surgery setting. Methods We concurrently recorded BIS and entropy (SE/RE) in 48 children aged 1 month–12 years, undergoing general anesthesia with sevoflurane and fentanyl. Nonlinear mixed effects modeling was used to characterize the concentration–response relationship independently between the three monitor indicators with sevoflurane. The model's goodness-of-fit was assessed by prediction-corrected visual predictive checks. Model fit with age was evaluated using absolute conditional individual weighted residuals (|CIWRES|). The ability of BIS and entropy monitors to describe the effect of anesthesia was compared with prediction probabilities (PK) in different age groups. Intraoperative and awakening values were compared in the age groups. The correlation between BIS and entropy was also calculated. Results |CIWRES| vs age showed an increasing trend in the model's accuracy for all three indicators. PK probabilities were similar for all three indicators within each age group, though lower in infants. The linear correlations between BIS and entropy in different age groups were lower for infants. Infants also tended to have lower values during surgery and at awakening than older children, while toddlers had higher values. Conclusions Performance of both monitors improves as age increases. Our results suggest a need for the development of new monitor algorithms or calibration to better account for the age-specific EEG dynamics of younger patients.
