Age-Stratified ED95 of Remimazolam Tosilate for Moderate Sedation in Pediatric Patients: A Biased Coin Design Study Authors: Bingqing Zhu, Wenjun Meng, Kunyun Pan, Li Peng, Xu Chen, Sheng Wang, Chengwei Yang Published: 2026 ABSTRACT AbstractBackgroundProviding safe sedation in pediatric patients remains challenging because of age-dependent physiological variations. This study aimed to identify the age-stratified 95% effective dose (ED95) of remimazolam tosilate for achieving moderate sedation in children.MethodsThis study enrolled children aged 1–12 years with ASA physical status I–II scheduled for elective surgery. Participants were stratified into three age groups: toddlers (1–3 years), preschoolers (3–6 years), and school-aged children (6–12 years), with 40 patients in each group. The initial dose of remimazolam tosilate was 0.2 mg·kg−1, with subsequent 0.05 mg·kg−1 adjustments using a biased coin design up-and-down sequential method based on patient response. Sedation depth was evaluated using the Modified Observer’s Assessment of Alertness and Sedation (MOAA/S) scale. An MOAA/S score ≤ 3 at 3 minutes after drug administration was defined as successful sedation. ED95 values were estimated using isotonic regression. Perioperative variables, including sedation onset time, operation time, total anesthesia time, and wake-up time, were recorded. Adverse events, including hypotension, bradycardia, and oxygen desaturation, were monitored during the observation period.ResultsOf the 142 children screened, 120 completed the study. The estimated ED95 (95% confidence interval) of remimazolam tosilate was 0.415 mg·kg−1 (0.311–0.437) in toddlers, 0.389 mg·kg−1 (0.258–0.440) in preschoolers, and 0.362 mg·kg−1 (0.228–0.396) in school-aged children. Toddlers had the longest median wake-up time (32.5 min; IQR: 25.3–40.8). Hypotension was reported in 0.8% of patients (n=1), whereas bradycardia and oxygen desaturation were each observed in 5.8% of patients (n=7).ConclusionThis study establishes age-specific ED95 reference doses of remimazolam tosilate for moderate sedation in children. Although remimazolam demonstrated an overall favorable safety profile, individualized dose titration remains essential, particularly given the 17.5% incidence of bradycardia observed in toddlers, which warrants closer monitoring.Trial RegistrationChinese Clinical Trial Registry (ChiCTR2300075986).Keywords: remimazolam, children, sedation, 95% effective dose Introduction Achieving safe and effective sedation in pediatric patients remains a major challenge in anesthetic practice. Children demonstrate age-dependent pharmacokinetic and pharmacodynamic characteristics that differ substantially from those of adults. They are inherently more susceptible to sedation-related adverse events.1 These differences arise from developmental changes in body composition, organ maturation, and drug metabolism. As a result, sedative dosing in children must be carefully adjusted to account for these physiological factors.2 The selection of pediatric sedative agents, therefore, prioritizes drugs with a rapid onset, short duration of action, and a favorable safety profile, particularly for brief procedures that require prompt recovery. Remimazolam is a novel ultra-short-acting benzodiazepine that is rapidly metabolized by nonspecific tissue esterases into an inactive metabolite.3 Pharmacokinetically, it is characterized by a plasma-effect site equilibration time of approximately 2.1 minutes and a context-sensitive half-time of approximately 11 minutes, even after prolonged infusion.4 This rapid onset and offset profile, together with minimal respiratory depression, stable hemodynamic effects, and non-irritating intravenous administration, offers clear advantages for procedural sedation and anesthesia in adults,5,6 highlighting its potential as a promising option for pediatric sedation.7 Despite these advantages, the application of remimazolam in pediatric anesthesia remains limited. Although the pharmacological properties of remimazolam have been well defined in adults, age-specific dosing recommendations for pediatric patients remain unavailable.8,9 Owing to the developmental immaturity of the autonomic nervous system and the evolving expression of GABA receptors, children may be more susceptible to hemodynamic instability and variable sedative responses during benzodiazepine administration.10–12 As a result, growing attention has been directed toward optimizing remimazolam dosing across different pediatric clinical settings. Despite the frequent requirement for moderate sedation in pediatric procedures,13,14 most existing studies have focused on loss of consciousness or alternative routes of administration,15,16 leaving limited evidence to guide moderate sedation practices. Since the dose requirement for remimazolam is influenced by the intended depth of sedation,17 establishing dosing protocols specifically designed for moderate sedation is clinically important. In clinical practice, the 95% effective dose (ED95) is regarded as the “potency dose of real interest”,18 as this high-percentile dose helps reduce sedation failure and the need for rescue interventions following the initial dose.13,19 This is particularly crucial for pediatric procedural sedation, given that children are typically less cooperative. In light of these considerations, the primary aim of this study was to determine the ED95 of remimazolam tosilate required to achieve moderate sedation in pediatric patients across age groups, using an up-and-down sequential method with a biased coin design. A secondary objective was to assess the preliminary safety and efficacy of remimazolam tosilate in the pediatric population. Methods Ethics and Registration This prospective clinical trial was approved by the Medical Research Ethics Committee of the First Affiliated Hospital of the University of Science and Technology of China (USTC) (Chairperson: Zuojun Shen; approval No. 2023KY-050; March 22, 2023). The study was registered with the Chinese Clinical Trial Registry (ChiCTR2300075986) on September 21, 2023. Patient recruitment commenced in October 2023, and the trial was completed in August 2024. This study is reported in accordance with the Consolidated Standards of Reporting Trials Dose-finding Extension (CONSORT-DEFINE) guidelines.20 Written informed consent was obtained from the legal guardians of all participants before enrollment. All study procedures complied with the Declaration of Helsinki and relevant clinical trial regulations. Inclusion and Exclusion Criteria Children aged 1–12 years scheduled to undergo elective surgery under general anesthesia were eligible for inclusion if they were classified as American Society of Anesthesiologists (ASA) physical status I or II. Exclusion criteria included a known allergy to remimazolam, hepatic or renal dysfunction, cardiovascular or respiratory disease, neurological disorders or developmental delay, and preoperative use of sedatives or anticonvulsant medications. Dose Allocation and Stopping Rules The ED95 of remimazolam was estimated using a biased coin design (BCD) up-and-down sequential method,13,21 managed by an independent data researcher blinded to patient management and clinical outcomes. A fixed step size of 0.05 mg·kg−1 was applied,22 and the sedation response of the preceding patient determined the dose administered to each successive patient. If the previous participant experienced sedation failure, the remimazolam dose for the subsequent participant was automatically increased by 0.05 mg·kg−1. A biased coin draw was executed only after successful sedation was confirmed at the 3-minute mark: the next participant received the same dose with a probability of 0.95, or a dose reduced by 0.05 mg·kg−1 with a probability of 0.05.18 Enrollment continued until at least 40 patients were included in each age group. Blinding To maintain strict blinding and ensure data integrity, the responsibilities for dose calculation, drug preparation, and outcome assessment were clearly separated. The designated data researcher calculated the required dose using the predefined BCD algorithm and communicated this information directly to an unblinded anesthesia nurse. The nurse prepared remimazolam in a standard syringe labeled only with the patient’s identification number. Sedation depth was evaluated by one of two designated anesthesiologists, both of whom were fully blinded to the dosing regimen and to the outcomes of previous patients. To avoid inadvertent unblinding related to differences in injection volume, the assessors were positioned away from the injection site and focused solely on patient responsiveness. Furthermore, both assessors completed standardized training before study initiation, during which consistency in MOAA/S scoring criteria was established using video-based instruction and simulated clinical scenarios. Anesthesia Protocol and Study Management A total of 120 participants were enrolled and stratified into three age groups: toddlers (1–3 years, n = 40), preschoolers (3–6 years, n = 40), and school-aged children (6–12 years, n = 40). All participants adhered to standard preoperative fasting guidelines, including abstinence from clear fluids for 2 hours and from solid food for 4–8 hours. Intravenous cannulation was performed in the ward, and intravenous fluid administration was initiated before anesthesia. No pre-anesthetic medications were administered before entry into the operating room. Parental presence was permitted during the preoperative period and transfer. Baseline vital signs, defined as T0, including oxygen saturation (SpO2), heart rate (HR), and noninvasive blood pressure (NIBP), were recorded in the preoperative holding area. Upon arrival in the operating room, patients were placed in the supine position, and standard monitoring was continuously applied, including upper arm NIBP, electrocardiography (ECG), SpO2, and end-tidal carbon dioxide (PETCO2). Preoxygenation with 100% oxygen at a flow rate of 2 L·min−1 was administered before remimazolam injection. Remimazolam tosilate (Jiangsu Hengrui Medicine Co., Ltd., Lianyungang, China) was diluted in 0.9% saline to 0.5 mg·mL−1 and administered manually as a standardized single intravenous bolus over 20–30 seconds. To minimize the risk of missing an effective dose in pediatric patients, the first child in each age group received an initial dose of 0.2 mg·kg−1.5,23 Doses for subsequent patients were determined by the designated data researcher according to the predefined BCD algorithm. Sedation depth was evaluated using the Modified Observer’s Assessment of Alertness and Sedation (MOAA/S) scale (Supplementary Table S1).13 Assessments were performed every 10 seconds following remimazolam administration or at any time sedation was considered adequate. Successful sedation was defined as an MOAA/S score ≤ 3 achieved and maintained until the end of the 3-minute observation period; otherwise, it was classified as sedation failure. In the event of failure, rescue sedation was administered using sevoflurane (4–8% inspired concentration in 100% oxygen) via face mask for 1–2 minutes until an MOAA/S score ≤ 3 was achieved.24 In the event of an allergic reaction or other unexpected adverse event, immediate treatment was provided and documented. At the same time, the dosing sequence for subsequent patients in the same age group continued unchanged. After completion of the observation period, anesthesia was induced with sevoflurane (3–5%), sufentanil (0.3 μg·kg−1), and cisatracurium (0.1 mg·kg−1) to facilitate either endotracheal intubation or laryngeal mask airway placement.25 Anesthesia was then maintained with sevoflurane at an end-tidal concentration of 2–4% in a 50% oxygen–air mixture, with adjustments made to keep heart rate and mean arterial pressure within 20% of baseline values.26 Vital signs were recorded at 1 minute (T1) and 3 minutes (T2) after remimazolam administration. Sedation onset time and adverse events during the observation period, including hypotension, bradycardia, and oxygen desaturation, were documented. Operation time and total anesthesia time were recorded at the end of surgery, and wake-up time was assessed as an exploratory outcome. Sedation onset was defined as the interval from remimazolam administration to the first MOAA/S score ≤ 3. Wake-up time was defined as the interval from the end of surgery to recovery of consciousness, indicated by an MOAA/S score ≥ 5. Hypotension was defined as a systolic blood pressure < 70 mmHg + (2 × age in years). Bradycardia was defined as a heart rate < 95 beats·min−1 in toddlers, < 80 beats·min−1 in preschoolers, and < 60 beats·min−1 in school-aged children.15 Episodes of hypotension or bradycardia requiring treatment were managed with vasoactive agents or anticholinergics, as appropriate. Oxygen desaturation was defined as SpO2 < 95%, with jaw thrust performed immediately as the initial intervention. Positive-pressure ventilation was initiated in cases of apnea or when SpO2 decreased below 90% despite this initial maneuver.27 Statistical Analysis Based on evidence from previous simulation studies,18,21 a sample size of 20–40 participants is generally sufficient to yield stable target-dose estimates using isotonic regression. Accordingly, 40 evaluable patients were included in each age group to ensure robust estimation of the ED95. The ED95 was calculated using isotonic regression with the pool-adjacent-violators algorithm (PAVA), assuming a monotonic dose–response relationship. Ninety-five percent confidence intervals (CIs) for the ED95 were derived using a bias-corrected percentile bootstrap approach based on 1,000 resampling iterations. Continuous variables were assessed for normality using the Shapiro–Wilk test. Non-normally distributed data were presented as median (interquartile range, IQR). Group comparisons for non-normally distributed continuous variables were performed using the Kruskal–Wallis test; significant results (P < 0.05) were further examined using post hoc Mann–Whitney U-tests. Categorical variables were analyzed using Fisher’s exact test, followed by post hoc pairwise comparisons when overall significance was observed (P < 0.05). Effect sizes were reported to quantify the magnitude of group differences. For binary outcomes, Cramér’s V was used, with 95% CIs estimated by bootstrap resampling (1,000 iterations). For continuous outcomes analyzed with the Kruskal–Wallis test, effect sizes were calculated using the epsilon-squared (ϵ2) statistic. Time-dependent changes in hemodynamic parameters across groups were analyzed using a generalized linear mixed-effects model with a Gamma distribution and log-link function. This model incorporated subject-specific random intercepts and fixed effects for group, time, and the group-by-time interaction. Bonferroni correction was applied to adjust for all multiple pairwise comparisons, with unadjusted P values multiplied by the number of comparisons to obtain adjusted values. All statistical analyses were conducted using IBM SPSS Statistics version 28.0 (Armonk, NY, USA) and R software version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria). All tests were two-sided, and statistical significance was defined as P < 0.05. Results General Patient Characteristics A total of 142 eligible pediatric patients were assessed for enrollment. Twenty-two were excluded, including 18 whose parents declined participation and four who did not meet the inclusion criteria (two with an ASA physical status ≥ III, one with pre-existing cardiac disease, and one with intellectual disability). The final study cohort comprised 120 children, with 40 participants per age group (Figure 1). Baseline demographic characteristics are summarized in Table 1. CONSORT diagram of the study. Table 1Demographic Data of the Patients in Three GroupsCharacteristicsToddlers (n=40)Preschoolers (n=40)School-Aged Children (n=40)Age (yr)1.9 (1.6–2.8)4.4 (3.8–5.3)7.0 (6.5–9.0)Weight (kg)13.0 (11.0–14.5)18.0 (17.0–20.9)24.3 (21.5–34.0)BMI (kg·m−2)16.7 (15.5–18.5)15.4 (14.5–16.6)16.5 (14.6–18.5)Sex, n (%)Male37 (92.5)25 (62.5)30 (75.0)Female3 (7.5)15 (37.5)10 (25.0)ASA physical status, n (%)I25 (62.5)29 (72.5)28 (70.0)II15 (37.5)11 (27.5)12 (30.0)Primary diagnosis, n (%)Inguinal hernia20 (50.0)26 (65.0)17 (42.5)Superficial masses3 (7.5)1 (2.5)5 (12.5)Hydrocele16 (40.0)9 (22.5)5 (12.5)Cryptorchidism/Concealed penis/Phimosis1 (2.5)4 (10.0)11 (27.5)Appendicitis0 (0.0)0 (0.0)2 (5.0)Note: Data are presented as median (IQR), or number (%).Abbreviation: BMI Body Mass Index, ASA American Society of Anesthesiologists. Demographic Data of the Patients in Three Groups Note: Data are presented as median (IQR), or number (%). Abbreviation: BMI Body Mass Index, ASA American Society of Anesthesiologists. Dose–Response Relationship for Remimazolam The observed and PAVA-adjusted sedation response rates at each remimazolam dose were summarized for the three age groups (Table 2 and Supplementary Figure S1). Although the unadjusted observed responses occasionally departed from a strictly monotonic pattern, the PAVA-adjusted estimates demonstrated a consistent increasing dose–response relationship across all groups (Supplementary Figure S1). Individual dose assignments and corresponding sedation outcomes, classified as success or failure, are graphically illustrated in Figure 2. The estimated ED95 of remimazolam demonstrated a stepwise decrease with increasing age: 0.415 mg·kg−1 (95% CI: 0.311–0.437) in toddlers, 0.389 mg·kg−1 (95% CI: 0.258–0.440) in preschoolers, and 0.362 mg·kg−1 (95% CI: 0.228–0.396) in school-aged children. Table 2Observed and PAVA-Adjusted Response Rates with Remimazolam (Isotonic Regression Method)Assigned Dose (mg·kg−1)Number SedatedNumber TestedObserved Response RatePAVA-Adjusted Response RateToddlers0.20010.0000.0030.25010.0000.0200.30120.5000.1480.35020.0000.5910.4014150.9330.9230.4519191.0000.989Preschoolers0.20010.0000.2190.25120.5000.4630.30340.7500.7270.3513140.9290.8910.40780.8750.9620.4511111.0000.987School-aged children0.20010.0000.5100.25890.8890.7210.30010.0000.8650.35230.6670.9410.4026261.0000.974Abbreviation: PAVA, pool-adjacent-violators algorithm. Observed and PAVA-Adjusted Response Rates with Remimazolam (Isotonic Regression Method) Abbreviation: PAVA, pool-adjacent-violators algorithm. Determination of the 95% effective dose (ED95) of remimazolam tosilate required to achieve moderate sedation in pediatric patients across different age groups. The x-axis denotes the sequence of patient enrollment, and the y-axis indicates the administered dose of remimazolam tosilate. Filled squares represent successful sedation, whereas open squares indicate sedation failure. The ED95 is depicted by a bold dashed line, and the shaded area represents the corresponding 95% confidence interval (CI). Hemodynamic Parameters and Adverse Events Hemodynamic variables were compared among the three age groups at predefined time points (Figure 3). Heart rate showed no significant change over time (P = 0.173; Figure 3A and Supplementary Table S2). Although significant time effects were detected for MAP and SpO2 (P < 0.001), the observed changes in MAP were modest, with median values decreasing slightly from T0 to T2 (toddlers: 66.0 to 63.0 mmHg; preschoolers: 72.5 to 68.5 mmHg; school-aged children: 79.0 to 72.0 mmHg; Figure 3B and Supplementary Table S2). Oxygen saturation remained above 95% in the majority of patients throughout the observation period (Figure 3C). No significant time-by-group interaction was observed for any of the measured parameters. Hemodynamic parameters across the three age groups. Panels show (A) heart rate (HR), (B) mean arterial pressure (MAP), and (C) oxygen saturation (SpO2) measured at predefined time points. Half-violin plots illustrate the distribution density, scatter plots represent individual observations, and box plots display the median (central line), interquartile range (box), and range (whiskers). T0 denotes baseline; T1, 1 minute after remimazolam administration; and T2, 3 minutes after remimazolam administration. Detailed numerical data are provided in Supplementary Table S2. Adverse events occurring within 3 minutes of remimazolam administration are summarized in Table 3. Hypotension was observed in 1 patient (0.8% overall), who was in the school-aged group. Oxygen desaturation occurred in 7 patients (5.8% overall), including 2 toddlers, 2 preschoolers, and 3 school-aged children. The incidence of hypotension and oxygen desaturation did not differ significantly among the three age groups. Bradycardia was documented in 7 patients (5.8% overall), all of whom were in the toddler group. The incidence of bradycardia in toddlers (17.5%) was significantly higher than that observed in preschoolers and school-aged children (adjusted P = 0.036 for both comparisons). All episodes of hypotension and bradycardia resolved after administration of a single dose of a vasoactive agent. Similarly, all cases of oxygen desaturation were successfully managed with a jaw-thrust maneuver without requiring positive-pressure ventilation. Table 3Perioperative Data and Adverse Events in Three GroupsParametersToddlers (n=40)Preschoolers (n=40)School-Aged Children (n=40)Effect Size (Estimate)P valuePerioperative data Sedation onset time (s)36.0 (28.0–41.0)38.5 (30.0–44.5)38.0 (33.25–45.0)0.0010.343† Operation time (min)12.5 (9.5–15.0)11.5 (10.0–15.0)15.0 (10.0–25.0)0.0210.110† Anesthesia time (min)20.0 (15.5–25.0)20.5 (16.3–28.8)23.5 (18.0–33.5)0.0080.223† Wake-up time (min)32.5 (25.3–40.8)a,b17.0 (15.0–20.5)16.5 (13.5–20.0)0.396< 0.001†Adverse events, n (%) Hypotension0 (0.0)0 (0.0)1 (2.5)0.130, (0.112–0.260)1.000‡ Bradycardia7 (17.5)c0 (0.0)0 (0.0)0.352, (0.211–0.479)< 0.001‡ Oxygen desaturation2 (5.0)2 (5.0)3 (7.5)0.05, (0.018–0.253)1.000‡Notes: Data are presented as median (IQR) or number (%). aP < 0.001 versus preschoolers, calculated by Mann–Whitney U-test with Bonferroni correction; bP < 0.001 versus school-aged children, calculated by Mann–Whitney U-test with Bonferroni correction; cP = 0.036 versus preschoolers and school-aged children, calculated by Fisher’s exact test with Bonferroni correction; †Kruskal–Wallis test, effect size reported as ϵ2; ‡Fisher’s exact test, effect size reported as Cramér’s V with 95% CI. Perioperative Data and Adverse Events in Three Groups Notes: Data are presented as median (IQR) or number (%). aP < 0.001 versus preschoolers, calculated by Mann–Whitney U-test with Bonferroni correction; bP < 0.001 versus school-aged children, calculated by Mann–Whitney U-test with Bonferroni correction; cP = 0.036 versus preschoolers and school-aged children, calculated by Fisher’s exact test with Bonferroni correction; †Kruskal–Wallis test, effect size reported as ϵ2; ‡Fisher’s exact test, effect size reported as Cramér’s V with 95% CI. Perioperative Outcomes Perioperative variables were compared among the three age groups (Table 3). Sedation onset time, surgical duration, and total anesthesia time did not differ significantly across the three groups (all P > 0.05), and the associated effect sizes were small. In comparison, wake-up time differed significantly among groups (ϵ2 = 0.396). Toddlers showed a longer recovery period, with a median wake-up time of 32.5 minutes (25.3–40.8), compared with 17.0 minutes (15.0–20.5) in preschoolers and 16.5 minutes (13.5–20.0) in school-aged children. Discussion This study established preliminary age-specific estimates of the ED95 of remimazolam tosilate required to achieve moderate sedation in pediatric patients using a BCD up-and-down sequential method. The estimated ED95 values were 0.415 mg·kg−1 (95% CI: 0.311–0.437) in toddlers aged 1–3 years, 0.389 mg·kg−1 (95% CI: 0.258–0.440) in preschoolers aged 3–6 years, and 0.362 mg·kg−1 (95% CI: 0.228–0.396) in school-aged children aged 6–12 years. Overall, remimazolam achieved effective sedation with a favorable safety profile across all age groups. However, given the distinct physiological vulnerability of toddlers, our clinical findings underscore the need for tailored management strategies for this population, including continuous monitoring and extended post-sedation observation. The optimal depth of remimazolam sedation depends on the specific procedure being performed.15,28,29 Previous studies have shown that an MOAA/S score ≤ 3 provides effective and safe preoperative sedation in pediatric patients,13,14,24 whereas deeper levels of sedation are associated with a higher risk of adverse events, including hypotension.30 Given the need to balance safety and clinical effectiveness, moderate sedation has become a widely adopted target in pediatric practice. However, the remimazolam doses required to reliably achieve this level of sedation across different pediatric age groups have not yet been clearly established. Therefore, this study defined successful sedation as achieving an MOAA/S score ≤ 3. Previous studies have reported varying ED95 values for remimazolam in pediatric patients. For example, Cai et al reported higher ED95 values when loss of consciousness was used as the target endpoint,15 likely reflecting the deeper level of sedation required in that study. These findings support a clear dose–depth relationship for remimazolam, in which deeper sedation requires higher drug doses. Although this pharmacological principle is well established, the present study provides novel age-stratified reference values for moderate sedation, offering practical guidance for pediatric settings where maintaining moderate sedation is clinically desirable. Overall, these results emphasize that remimazolam dosing in children should be individualized based on patient age and the intended depth of sedation to optimize procedural safety and efficacy. Recent pharmacological data indicate that remimazolam reaches its peak effect approximately 2.5 minutes after administration, with an equilibration half-life of about 2.1 minutes.4,31 Given the single-bolus administration used in this study, a 3-minute observation period was selected to encompass the expected peak pharmacodynamic response. Accordingly, the safety outcomes reported in this study primarily reflect the tolerability of remimazolam during the 3-minute window. The BCD method enables efficient estimation of the ED95 by concentrating observations near the target dose range.32 This strategy accounts for pharmacodynamic variability within the upper tail of the tolerance distribution, particularly among patients who are less sensitive to the drug and therefore require higher doses.18 As a result, ED95 estimation provides clinically reliable guidance for achieving procedural success in the majority of patients.13 The ED95 values for remimazolam identified in this study are intended to support clinicians in developing dosing strategies that minimize sedation failure and reduce the need for rescue interventions.19 This is especially important in pediatric practice, where reducing repeated drug administration may help lower the cumulative risk of adverse events and psychological distress. In adult populations, remimazolam has been associated with transient increases in HR.22,33,34 In comparison, pediatric studies have reported reductions in HR in a subset of children following remimazolam administration.24,29 In this study, overall HR did not show significant time-dependent changes (Supplementary Table S2). However, the incidence of bradycardia was significantly higher in toddlers (17.5%) compared with preschoolers and school-aged children (Table 3). Although the precise effects of remimazolam on the autonomic nervous system in toddlers remain unclear, pharmacological studies in pediatric populations should carefully account for developmental and sex-related differences.11 From a physiological perspective, toddlers are in a stage of ongoing autonomic maturation. Compared with older children, they demonstrate a developing sympathetic–parasympathetic balance,12 which may partially contribute to their greater susceptibility to bradycardia. Furthermore, regarding sex distribution, the toddler group in this study had the highest proportion of male participants (92.5%) (Table 1). Previous studies involving other sedative agents have identified male sex as a potential risk factor for bradycardia during pediatric sedation.35 Therefore, the combined effects of autonomic functional immaturity and the sex imbalance within our cohort may have contributed to the higher incidence of bradycardia observed in this age group. Given these considerations, continuous electrocardiographic monitoring and immediate availability of anticholinergic agents are recommended when using remimazolam in toddlers, even during short procedures. Further large-scale investigations are needed to confirm these observations and clarify the underlying mechanisms. Consistent with adult studies,5,36 remimazolam did not cause clinically meaningful hemodynamic instability in the pediatric population. Although statistically significant time effects were observed for MAP and SpO2, the magnitude of these changes was small and unlikely to be of clinical relevance. Previous pediatric investigations of remimazolam for anesthesia induction reported no significant respiratory depression.15 In contrast, our study recorded an oxygen desaturation incidence of 5.8% (n = 7). This discrepancy may be explained by the use of more restrictive criteria for defining respiratory depression in previous studies. Although broader criteria were applied in this study, the overall frequency of oxygen desaturation was low and the severity of these events was mild, indicating acceptable clinical tolerability. A plausible explanation for this favorable respiratory profile is remimazolam’s high selectivity for the GABAA receptor, which may result in less suppression of respiratory drive than other sedative agents.29,37,38 Regarding the onset of action, no significant differences were observed among the three age groups in this study (Table 3). Previous investigations employing an MOAA/S score ≤ 1 as the study endpoint reported longer onset times for remimazolam than those observed here.29 In comparison, the study adopting an MOAA/S threshold of ≤ 3 has reported onset times comparable to our findings.24 These observations indicate that remimazolam offers a rapid onset of action for pediatric procedures requiring moderate sedation. Although drug administration was standardized to 20–30 seconds, manual bolus injection may still introduce variability in peak drug concentrations and onset profiles compared with pump-controlled infusion. Toddlers demonstrated a significantly longer wake-up time than preschoolers and school-aged children (Table 3). However, this delay is likely influenced by several confounding factors, including the combined effects of other anesthetic agents and the more cautious post-anesthesia care practices typically applied to younger children at our institution. Furthermore, increased remimazolam requirements secondary to preoperative crying and agitation in toddlers may be an additional contributing factor; however, estimating the ED95 under these realistic clinical stress conditions ultimately provides more practical dosing guidance for routine pediatric practice.13,14,24 Future studies incorporating standardized anxiety assessments are warranted to further refine dosing strategies across varying emotional states. Consistent with findings from adult studies,5,22 the ED95 values identified in this study demonstrated a stepwise decrease with increasing age. The higher ED95 observed in toddlers is likely related to an increased volume of distribution associated with their greater total body water content.15,29 Interestingly, although toddlers required the highest weight-adjusted dose to achieve the target sedation level, they concurrently exhibited a higher incidence of bradycardia and significantly longer recovery times (Table 3). This seemingly paradoxical clinical profile indicates that the requirement for a higher dose in toddlers does not necessarily reflect greater physiological tolerance. Therefore, these findings emphasize the importance of extended post-sedation monitoring in this age group. In contrast, among children older than 3 years, remimazolam’s rapid onset and offset characteristics support its suitability for high-throughput pediatric procedures. The relationship between BIS values and the sedative efficacy of remimazolam remains unclear, limiting the utility of BIS monitoring in this setting.29,39,40 Similarly, the Narcotrend Index has demonstrated only a weak correlation with MOAA/S scores,4 rendering it unsuitable for reliable assessment of sedation depth during remimazolam administration. Practical considerations, including poor compliance among awake pediatric patients, further restrict the feasibility of objective electroencephalographic monitoring in this population. Therefore, sedation depth in this study was assessed exclusively using the MOAA/S scale. However, it must be acknowledged that the MOAA/S scale has not been formally psychometrically validated in toddlers aged 1–3 years. At this developmental stage, differentiating sedation-induced unresponsiveness from age-appropriate behavioral noncompliance or limited receptive language abilities is inherently challenging. Although the use of MOAA/S in young children is supported by previous pediatric studies,16,18,19 this limitation may reduce the precision of ED95 estimates in the toddler group. Future investigations should aim to confirm these findings using sedation assessment tools specifically designed and validated for younger pediatric populations. Several limitations of this study should be considered when interpreting the results. First, only healthy children with ASA physical status I–II were included, and the estimated ED95 values may therefore not be directly generalizable to children with ASA physical status III or higher. Second, the lack of specific validation of the MOAA/S scale for toddlers may introduce assessment bias. Third, the marked sex imbalance in the toddler group represents a potential confounding factor that may have influenced the estimated dose requirements. Finally, the study protocol primarily evaluated sedative efficacy during the peak effect period, and the relatively short observation window may have precluded the detection of delayed adverse events occurring after this timeframe. Despite these limitations, our findings provide a valuable basis for clinical practice. Future studies should include multicenter validation across more diverse pediatric populations and incorporate extended monitoring periods to confirm and further refine these age-stratified dosing recommendations. Conclusions This study establishes age-specific ED95 estimates of remimazolam tosilate as reference values for achieving moderate sedation in pediatric patients. Although remimazolam demonstrated effective sedation with an overall favorable safety profile, individualized dose titration and careful monitoring remain essential. The 17.5% incidence of bradycardia observed in toddlers further highlights the need for closer surveillance in this age group. Acknowledgments The authors are grateful to their colleagues in the Department of Anesthesiology at The First Affiliated Hospital of USTC (Hefei, Anhui, China) for their assistance in patient recruitment and data collection during this study. Funding Statement This study was supported by the National Natural Science Foundation of China (Grant No. 82101289), Wu Jieping Medical Foundation (Grant No. 320.6750.2020-21-13), Hubei Chen Xiaoping Science and Technology Development Foundation (Grant No. CXPJJH124001-2481) and Fundamental Research Funds for the Central Universities (Grant No. WK9110000169). Data Sharing Statement The datasets generated and analyzed during the current study are available from the corresponding author (Chengwei Yang, yangcw1108@ustc.edu.cn) upon reasonable request. Ethics Approval and Consent to Participate This study received approval from the Medical Research Ethics Committee of the First Affiliated Hospital of USTC (approval number: 2023KY-050). It was registered in the Chinese Clinical Trial Registry (registration number: ChiCTR2300075986). All procedures performed in this study involving human participants complied with the Declaration of Helsinki and applicable clinical trial guidelines. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors declare that they have no competing interests. References Associated Data Data Availability Statement The datasets generated and analyzed during the current study are available from the corresponding author (Chengwei Yang, yangcw1108@ustc.edu.cn) upon reasonable request.