Association between emergence delirium and brain status parameters in children undergoing general anesthesia: A prospective observational study

Introduction: Emergence delirium is a common postoperative neurological complication in children after general anesthesia. There is no valid tool to predict emergence delirium. Wavelet index, pain threshold index, anxiety index, and comfort index are real-time brain status parameters extracted from the electroencephalogram, which have recently been developed. The aim is to evaluate the association between real-time brain status parameters during emergence and emergence delirium in children undergoing general anesthesia.

Methods: One hundred and thirty patients between 3 and 6 years of age undergoing dental surgery under general anesthesia were enrolled in the study. Real-time electroencephalogram data were recorded at four different time points from end of anesthetics administration (T1), end of surgery (T2), extubation (T3), and response (eye opening, movement) to verbal stimulation (T4). Each patient was assessed for emergence delirium using the Pediatric Anesthesia Emergence Delirium scale. Receiver operating characteristics curves and the associated areas under the curves were computed to analyze the ability of wavelet index, pain threshold index, anxiety index, and comfort index to predict emergence delirium.

Results: One hundred and sixteen patients were included for final analysis. During recovery from general anesthesia, brain status parameters increased significantly from T1 (wavelet index, 59.5 ± 6.2; pain threshold index, 61.7 ± 5.3; anxiety index, 9.2 ± 2.5; comfort index, 21.6 ± 8.7) to T4 (wavelet index, 67.4 ± 9.4; pain threshold index, 73.2 ± 9.1; anxiety index, 38.6 ± 11.2; comfort index, 66.1 ± 16.5; p < .001). To predict emergence delirium, areas under the curves [95% CI] for anxiety index were 0.84 [0.75-0.93] (p < .001), and comfort index was 0.89 [0.81-0.96] (p < .001). The Pediatric Anesthesia Emergence Delirium scale scores of 37 patients were higher than 10 indicating emergence delirium, and the incidence of emergence delirium was 31.90%. The sensitivity and specificity of anxiety index with corresponding cutoff values in predicting emergence delirium were 73.0% and 89.9%, and the sensitivity and specificity of comfort index in predicting emergence delirium were 91.9% and 83.5%. The best cutoff values for anxiety index and comfort index to predict emergence delirium were 46.5 and 68.5, respectively. The areas under the curves [95% CI] of wavelet index to predict emergence delirium were 0.43 [0.31-0.35] (p = .27), while the areas under the curves [95% CI] of pain threshold index to predict emergence delirium were 0.49 [0.37-0.62] (p = .73).

Discussion: Both anxiety index and comfort index derived from electroencephalogram wavelet analysis were associated with emergence delirium in pediatric patients undergoing general anesthesia for dental surgery. Wavelet index and pain threshold index were not associated with emergence delirium during general anesthesia for dental surgery in children.

Conclusions: AnXi and CFi might be used to guide anesthesiologists to identify and intervene ED in children.