Responses of Processed Electroencephalographic Indices to Ascending and Descending Simulated Remimazolam Effect-Site Concentrations: A Prospective Exploratory Interventional Study

Background/Objectives: Temporal characteristics of concentration–response loops for processed electroencephalographic indices during changes in remimazolam effect-site concentration (Ce) remain unclear. Therefore, this prospective, exploratory interventional study aims to compare the magnitude of concentration–response loops between the ascending and descending phases of simulated remimazolam Ce for the bispectral index (BIS) and patient state index (PSI). Methods: Twenty patients were administered remimazolam at a constant infusion rate of 3.5 mg/kg/h for 5.5 min, followed by infusion discontinuation. BIS and PSI values were continuously recorded from initiation of remimazolam infusion until 16 min, and remimazolam Ce was simulated using the Schüttler–Eisenried pharmacokinetic model. The area between the ascending and descending concentration–response curves (ABC) was calculated for each index within the overlapping Ce range. ABC was compared between BIS and PSI using a paired analysis, and a patient-level bootstrap was used to estimate the 95% confidence interval (CI) for the paired difference. Results: All 20 patients were included in the final analysis. The mean paired difference in ABC (BIS − PSI) was −6.254 (95% CI: −7.945 to −4.563), with a bootstrap median difference of −6.241 (95% CI: −8.752 to −4.133). The gap between the ascending and descending concentration–response curves was greater for PSI than for BIS. Conclusions: BIS and PSI showed different concentration–response patterns during changes in simulated remimazolam Ce, with PSI showing greater separation between the ascending and descending phases. These exploratory findings do not establish direct pharmacodynamic hysteresis or indicate superiority of BIS for assessing hypnotic state.