Background: Intraoperative isoelectric electroencephalography (EEG) has been associated with hypotension and postoperative delirium in adults. This international prospective observational study sought to determine the prevalence of isoelectric EEG in young children during anesthesia. The authors hypothesized that the prevalence of isoelectric events would be common worldwide and associated with certain anesthetic practices and intraoperative hypotension. Methods: Fifteen hospitals enrolled patients age 36 months or younger for surgery using sevoflurane or propofol anesthetic. Frontal four-channel EEG was recorded for isoelectric events. Demographics, anesthetic, emergence behavior, and Pediatric Quality of Life variables were analyzed for association with isoelectric events. Results: Isoelectric events occurred in 32% (206 of 648) of patients, varied significantly among sites (9 to 88%), and were most prevalent during pre-incision (117 of 628; 19%) and surgical maintenance (117 of 643; 18%). Isoelectric events were more likely with infants younger than 3 months (odds ratio, 4.4; 95% CI, 2.57 to 7.4; P < 0.001), endotracheal tube use (odds ratio, 1.78; 95% CI, 1.16 to 2.73; P = 0.008), and propofol bolus for airway placement after sevoflurane induction (odds ratio, 2.92; 95% CI, 1.78 to 4.8; P < 0.001), and less likely with use of muscle relaxant for intubation (odds ratio, 0.67; 95% CI, 0.46 to 0.99; P = 0.046]. Expired sevoflurane was higher in patients with isoelectric events during preincision (mean difference, 0.2%; 95% CI, 0.1 to 0.4; P = 0.005) and surgical maintenance (mean difference, 0.2%; 95% CI, 0.1 to 0.3; P = 0.002). Isoelectric events were associated with moderate (8 of 12, 67%) and severe hypotension (11 of 18, 61%) during preincision (odds ratio, 4.6; 95% CI, 1.30 to 16.1; P = 0.018) (odds ratio, 3.54; 95% CI, 1.27 to 9.9; P = 0.015) and surgical maintenance (odds ratio, 3.64; 95% CI, 1.71 to 7.8; P = 0.001) (odds ratio, 7.1; 95% CI, 1.78 to 28.1; P = 0.005), and lower Pediatric Quality of Life scores at baseline in patients 0 to 12 months (median of differences, –3.5; 95% CI, –6.2 to –0.7; P = 0.008) and 25 to 36 months (median of differences, –6.3; 95% CI, –10.4 to –2.1; P = 0.003) and 30-day follow-up in 0 to 12 months (median of differences, –2.8; 95% CI, –4.9 to 0; P = 0.036). Isoelectric events were not associated with emergence behavior or anesthetic (sevoflurane vs. propofol). Conclusions: Isoelectric events were common worldwide in young children during anesthesia and associated with age, specific anesthetic practices, and intraoperative hypotension. Editor’s Perspective: What We Already Know about This Topic: What This Article Tells Us That Is New:

Purpose: Ciprofol (HSK3486 injectable emulsion) is a 2.6-disubstituted phenol derivative that acts similarly to propofol with a fast onset, quick, and stable recovery. Moreover, remimazolam is a new benzodiazepine with a faster onset and recovery than midazolam, and more stable hemodynamics than propofol. We found no relevant literature that directly compared these two drugs. The primary objective of this study was to compare the recovery of HSK3486 with that of remimazolam during successful general anesthesia. Patients and Methods: This prospective, randomized, double-blind study included 93 patients who had undergone fundus surgery. The patients were divided into three groups: HSK3486 (H group), remimazolam-saline (RS group), and remimazolam-flumazenil (RF group). The primary outcome was the time from drug withdrawal to eye opening, and the secondary outcome was the time from drug withdrawal to laryngeal mask removal, the time for Aldrete modified score ≥ 9, intraoperative hemodynamic changes, post-operative adverse events (nausea, vomiting, pain, delirium, re-sedation), and the Quality of Recovery-15 score before surgery and at 1, 7, and 14 days post-operatively. Results: The RF group showed the shortest eye-opening time: 3.8 (2.8) min vs 13.3 (5.2) min for the H group, P < 0.001; and 11.5 (5.5) min for the RS group, P < 0.001. And laryngeal mask removal time was 5.2 (3.1) min for the RF group vs 14.1 (5.2) min for the H group, P < 0.001; and 12.4 (5.5) min for the RS group, P < 0.001. There were no significant differences in other post-operative outcomes. Conclusion: The incorporation of flumazenil with remimazolam for total intravenous anesthesia provided rapid and reliable recovery of consciousness, although there were no significant visible differences between HSK3486 and Remimazolam-saline.

Processed electroencephalography (EEG) indices used to guide anesthetic dosing in adults are not validated in young infants. Raw EEG can be processed mathematically, yielding quantitative EEG parameters (qEEG). We hypothesized that machine learning combined with qEEG can accurately classify expired sevoflurane concentrations in young infants. Knowledge from this may contribute to development of future infant-specific EEG algorithms. Frontal EEG collected from infants ≤ 3 months were time-matched as one-minute epochs to expired sevoflurane (eSevo). Fifteen qEEG parameters were extracted from each epoch and eight machine learning models combined the qEEG to classify each epoch into one of four eSevo levels (%): 0.1–1.0, 1.0–2.1, 2.1–2.9, and > 2.9. 64 epochs formed the post hoc SHAP dataset to determine the qEEG that contributed most to the model. The remaining epochs were randomly split 50 times into 80/20 training/testing sets. Accuracy and F1-score determined model performance. 42 infants provided 4574 epochs. The top classifiers K-nearest neighbors, default multi-layer perceptron, and support vector machine achieved 67.5–68.7% accuracy. Burst suppression ratio and entropy β were the top contributors to the models. Post hoc analysis performed without burst suppression ratio yielded similar prediction performance. In young infants, machine learning applied to qEEG predicted eSevo levels with moderate success. Burst suppression ratio, the most important contributor, represented an efficient EEG feature that encapsulated underlying EEG changes seen on other qEEG features. These results provided insight into EEG parameter selection and optimal machine learning models used for future development of infant-specific EEG algorithms.