Study design and participants
This prospective dose-response study was conducted in a multi-discipline ICU with 26 beds. The research was carried out in accordance with the Ethics Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Permission number: 0445-01), and informed consent were obtained prior to study enrolment.
Eighty-seven postoperative patients admitted to the integrated ICU of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology between November 1, 2020 and September 30, 2021 were screened using the following inclusion and exclusion criteria. The inclusion criteria included: (1) patients aged between 18 and 80 years old; (2) patients admitted to ICU immediately after surgery with general anesthesia and invasive mechanical ventilation. The exclusion criteria included: (1) allergy to benzodiazepines; (2) myasthenia gravis and coma (glasgow coma scale (GCS) score < 8); (3) pregnant or lactating women; (4) patients with uncorrected hemorrhagic or cardiogenic shock; (5) body mass index (BMI)＜18 kg/m2, or＞35 kg/m2; (6) chronic renal failure or chronic liver damage; (7) alcohol dependence, drug use, mental illness or severe cognitive impairment; and (8) requiring deep sedation due to clinical consideration (RASS<–3). A total of 23 patients met these criteria and were included in the study (Fig. 1).
Sedation assessment and interventions
In this study, the Narcotrend index (NTI) was used to assess the depth of sedation, and the Richmond Agitation-Sedation Scale (RASS) was also recorded in order to better titrate the dose of remimazolam within 24 hours of the admission to the ICU after surgery.
The NTI is an automatic electroencephalogram (EEG) recording system designed to assess the depth of sedation and anesthesia , which has been proven to be equivalent to bispectral Index (BIS) [14-16]. The skin on the patient’s forehead was prepared with alcohol gauze to keep impedances less than 4 kΩ, then placed three electrode pads on the forehead to connect to the machine with a distance of at least 8cm between any two electrodes. The Narcotrend (MonitorTechnick, Bad Bramstedt, Germany) was used to assess the level of sedation: NTI 0–12 (burst suppression), 13–36 (deep anesthesia), 37–64 (general anesthesia), 65–79 (moderate sedation), 80–94 (light sedation), 95–100 (awake).
The RASS is a 10 points subjective scale that designed to assess the level of sedation and agitation, which can be easy to perform and needs minimal training. The RASS has been widely used in ICU patients and its reliability and validity have already been confirmed . 5 negative scores is assigned to the level of tranquility (–5 = unarousable, –4 = deep sedation, –3 = moderate sedation, –2 = light sedation, –1 = drowsy), 0 score for being alert and calm, and 4 positive scores are assigned to the level of agitation (+1 = restless, +2 = agitated, +3 = very agitated, +4 = combative).
After inclusion, the researchers must confirm that the patient was connected to ECG monitor and placed protective restraints. At that time, any other sedatives would be stopped (if the patients had), but the analgesics retained (if necessary, use sufentanil or nabuphine to make the patient achieve an effective level of analgesia with the target of the score of Critical-care Pain Observation Tool (CPOT) = 0). After confirming that the patient can breathe spontaneously, the researcher would adjust the ventilator to constant positive airway pressure (CPAP) + pressure support ventilation (PSV) mode.
Before the administration of remimazolam besylate, the patients should be awake from general anesthesia (NTI>95, RASS of 0–+1). The remimazolam besylate (36mg, Jiangsu Hengrui Pharmaceutical Co. Ltd.) was formulated into 1 mg/ml with 0.9% saline, and administered with a micro-injection pump. The infusion doses of remimazolam besylate were calculated using the theoretical ideal weight of the patient [X + 0.91 (height–152.4), where X is 50.0 for men and 45.5 for women]. The loading dose of remimazolam besylate started from 0.02 mg/kg, which was injected within one minute. A supplementary bolus of 0.005 mg/kg of remimazolam would be administered until the RASS of –3––1 and NTI of 65–94 were obtained with the lowest possible dose. The initial maintaining dose of remimazolam besylate was 0.2 mg/kg/h, followed by increments or subtractions of 0.05 mg/kg each time until the RASS of –3––1 and NTI of 65–94 were achieved, which should be maintained for 30–120 minutes. If the RASS and NTI are significantly inconsistent, the RASS would be the main reference.
The clinical data of the patients involved in this study included: (1) basic demographic data: age, sex, height, weight, body mass index, education level, and comorbidities, etc.; (2) anesthesia and surgical data: American Society of Anesthesiologists (ASA) class, type of surgery, total surgical procedure time, total anesthesia time, anesthetics used during anesthesia; (3) monitors parameters: systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), heart rate (HR), respiratory rate (RR), tidal volume (Vt), and oxygen saturation (SpO2); and (4) usage and drip speed of remimazolam. All data were checked by two researchers.
The primary outcomes are the optimal dosing of remimazolam besylate for a single intravenous infusion and subsequent optimal dosing for continuous pumping. Secondary outcomes are the changes of hemodynamics and respiratory parameters (SBP, DBP, MAP, HR, RR, Vt, and SpO2) before and after drug use.
Categorical variables were described as number (%). Continuous variables were described using mean (SD) if they were normally distributed, or median (interquartile range, IQR) if they were not. The data were compared using a one-way repeated measures ANOVA. The correlation of Narcotrend Index and RASS was analyzed by Spearman’s Related Coefficient and the prediction probability of NTI for RASS was calculated by PKMACRO software based on the statistical approach proposed by Smith et al . P < 0.05 was considered statistically significant. The data collected were all analyzed using SPSS version 20.0 software (SPSS, Tokyo, Japan).