Study Design and Population

We conducted a randomized, double-blind, placebo-controlled prospective clinical trial from August to November 2022 at the Affiliated Hospital of Qingdao University. The study was conducted in accordance with the principles outlined in the Declaration of Helsinki and adhered to the reporting guidelines provided by the Standards for Reporting Interventions in Clinical Trials of Acupuncture [18] and the Consolidated Standards of Reporting Trials statement [19]. Ethics approval: The ethics committee of the Affiliated Hospital of Qingdao University approved this study (ID: QYFYKYLL916711920), which was also registered on the clinical trial website (NCT05485064). All patients provided informed consent before enrollment.

Inclusion criteria were patients aged 18 years to 80 years who were scheduled for gastrointestinal endoscopy were included. Exclusion criteria were patients with neuropsychiatric disorders, severe anxiety, severe depression, etc., who could not complete this study; those with American Society of Anesthesiologists (ASA) class > III, such as those with severe heart, liver, kidney, brain, or lung diseases; those who did not receive TEAS or who were allergic to acupoint electrodes or benzodiazepines; pregnant or lactating women; those with a history of long-term alcohol or drug abuse; and those who were unable to provide informed consent. Additionally, patients who were aware of or had previously received TEAS treatment were excluded because patients in group P were aware that TEAS was not guaranteed to be included in the grouping, which could lead to the failure of double-blind control [9, 17].

Division of Researchers

The study staff members included three gastroenterologists, three endoscopic nurses, two research assistants, and one anesthesiologist. The gastroenterologists had the same proficiency level (more than 3000 gastrointestinal endoscopies were performed). Research assistant 1 recruited, screened, obtained informed consent, randomized, grouped, collected general information, placed acupuncture electrodes at acupoints, and adjusted the stimulation parameters of the patients. Researcher assistant 1 was the only member who was aware of the study groups. The administration of moderate sedation, monitoring of vital signs, and evaluation of adverse events during the examination were carried out by the anesthesiologist. Research assistant 2 was responsible for performing the behavioral pain scale for non-intubated patients (BPS-NI) [13, 20] assessments at each time point during the examination, documenting adverse events and procedure duration, overseeing patient recovery, conducting satisfaction surveys, and conducting follow-up via telephone.

Randomization and Blindness

After the patients had an appointment for gastrointestinal endoscopy, researcher assistant 1 invited the patients to participate in this clinical trial, obtained their informed consent, and randomized them into the fentanyl plus remimazolam group (group C), TEAS plus remimazolam group (group E), and placebo-TEAS plus remimazolam group (group P) by using a computer-generated random numbers table. To ensure that patients in group E or group P were unaware of the grouping, all patients had no prior experience or knowledge of TEAS and were advised that they may not feel anything in the acupoints where the electrodes were placed [9, 17].

Intervention and Procedure

All patients were required to undergo fasting and water deprivation for more than 8 h and complete routine bowel preparations associated with the colonoscopy. Venous access was established after the patients were hospitalized on the examination day. Subsequently, normal saline was continuously administered.

Before Sedation

Patients were directed by research assistant 1 to enter a dedicated induction room and assume a left lateral position on an examination cart. Acupuncture electrodes were applied to the bilateral Neiguan (LI4), Hegu (PC6), and Zusanli (ST36) acupoints for patients in groups C and P. A Hwato acupuncture therapy device was connected [Specification: SDZ-III, Suzhou Medical Limited Company, Jiangsu Food and Drug Administration (Approval) No. 2270675, 2017], and the parameters were set at dense wave frequencies of 2 and 100 Hz with automatic shifting, maximum tolerable stimulation intensity of the patient (current within 2–3 mA), and pre-operation stimulation time of group E (30 min)]. Throughout the entire examination, the stimulation parameters remained at their default settings [10, 11]. Patients in group P had acupuncture electrodes attached to them, similar to those in group E, but actual electrical stimulation was not administered [14, 15]. Patients in group C waited 30 min between inductions and were administered 50 µg (fentanyl) intravenously by research assistant 1 before they entered the examination room. Research assistant 1 shielded all patients before they entered the examination room to ensure double blinding [14].

Sedation

After entering the examination room, all patients were administered 10 g of lidocaine hydrochloride mucilage (lidocaine 0.2 g) for adequate local anesthesia of the oral cavity and pharynx; then, they were administered an initial dose of remimazolam (5 mg) [remimazolam tosilate for injection, Jiangsu Hengrui Medicine Co., Ltd., China, national medicine permission number: H20190034] intravenously within 1 min and received top-up doses (2.5 mg/dose) dividedly to reach the Modified Observer’s Assessment of Alertness/Sedation (MOAA/S) scale of 3–4 [2]. The MOAA/S scores were assessed at regular intervals of 2 min. Continuous administration of supplemental oxygen at a flow rate of 2–4 l/min was provided through a nasal cannula, while the anesthesiologist administered intravenous normal saline. Vital signs, including electrocardiogram, heart rate (HR), noninvasive blood pressure (NIBP), and oxygen saturation (SpO2), were continuously monitored. If the patient reached the target depth of sedation and was still unable to tolerate the stimulation of gastrointestinal endoscopy (BPS-NI > 9 points), the case was designated a treatment failure, and 50–100 mg of propofol was administered intravenously as a rescue sedative medication by the anesthesiologist.

Gastrointestinal Endoscopy

After reaching the predetermined depth of sedation, the gastroenterologist performed gastrointestinal endoscopy according to the order of gastroscopy and colonoscopy. The gastrointestinal endoscope used was an Olympus CV-290 [Olympus Corporation, China Food and Drug Administration (Approval) No. 3221606, 2017], and the OD of the gastroscope and colonoscope was 9.8 and 13.2 mm, respectively.

Recovery Process

Following the completion of the gastrointestinal endoscopy, research assistant 1 transferred patients to the induction room to remove acupuncture electrodes and associated equipment. Subsequently, patients were moved to the recovery room for ongoing monitoring of vital signs. The discharge criteria were based on the modified Aldrete score, with a minimum threshold of ≥ 9 [21].

Outcomes

The primary endpoints were patient satisfaction, physician satisfaction, and pain scale score during the examination. The secondary endpoints were time of recovery, recovery of normal behavioral function and discharge, incidence of adverse reactions, and dose of remimazolam.

Satisfaction Survey and Follow-up Questionnaire

The content and format of the satisfaction survey and follow-up questionnaire were improved based on Cohen (2004) [22], Megan (2006) [23], and Levitzky (2012) [24]. At the conclusion of procedural sedation, research assistant 2 conducted satisfaction surveys with gastroenterologists using a clinical satisfaction with sedation instrument (CSSI). Similarly, patients were administered satisfaction surveys with a sedation instrument (PSSI) before discharge. Both surveys utilized a 100-mm visual analog scale (VAS) for assessment, with scores ranging from 0 (not at all satisfied) to 100 (completely satisfied). After completing the surveys, the patients were discharged with paper questionnaires of follow-up surveys. Research assistant 2 also contacted discharged patients by phone within 1 week to complete the follow-up questionnaires. They reported their overall satisfaction with the procedural sedation on the 100-mm VAS of the follow-up questionnaire. In addition, they also reported the time when they returned to normal activities (such as work, swimming, etc.) after the examination, the occurrence of nausea and emesis after discharge (based on a 100-mm VAS), procedure recall, and willingness to undergo procedural sedation again.

Pain Response Scale

Research assistant 2 used the BPS-NI to evaluate patients’ pain scale at each timepoint of gastrointestinal endoscopy (Table 1) [9]. The scale comprises three components, facial expression, body movement, and vocalization, each assessed on a scale of 1 (no pain) to 4 (most severe pain). The overall score ranged from 3 (no pain) to 12 (most severe pain).

Table 1 Timepoints of gastrointestinal endoscopyIndicators Related to Time

The recovery time was the time from gastrointestinal endoscopy to MOAA/S = 5 (evaluated three times consecutively). The discharge time was the time from full alertness to discharge. The time to resume normal behavioral function was the time from discharge to return to normal (patients resuming work, swimming, driving, etc.), which was obtained through the follow-up questionnaire.

The induction time referred to the duration from the commencement of the intravenous administration of remimazolam by the anesthesiologist to the attainment of a pre-established level of sedation. The procedure time is the total time to complete gastrointestinal endoscopy (from the insertion of the gastroscope to the withdrawal of the colonoscope from the anus).

Adverse Reactions During Examination and Recovery

The anesthesiologist monitored and recorded adverse events, such as respiratory depression, hypoxemia, abnormal blood pressure, reflux or aspiration, sedation failure, and the use of rescue sedative medication [2, 22]. Hypoxemia and severe hypoxemia were defined as SpO2 < 90% and < 85%, respectively. Hypotension and bradycardia were defined as a decrease in mean arterial pressure and a heart rate ≥ 25% from the baseline value, respectively. Apnea was defined as an adverse event of respiratory arrest lasting ≥ 30 s or < 6 bpm. Adverse events (nausea, emesis, or dizziness) were observed and recorded by research assistant 2 in the recovery room.

Dose of Remimazolam

The dose of remimazolam used during gastrointestinal endoscopy in the three groups was recorded separately by researcher assistant 2.

Statistical Methods

SPSS (version 26.0) software was used to statistically analyze all the data. The Kolmogorov test was used to analyze the normal distribution of all the data. Analysis of variance was used to analyze normally distributed measurement data, and the data are expressed as the mean ± standard deviation (x̅ ± SD). Pairwise comparisons between the three groups were conducted using the least significant difference t test (LSD-t). The Kruskal‒Wallis rank-sum test was employed to analyze nonnormally distributed measurement data, with the results presented as medians (ranges). For pairwise comparisons between the three groups, the Bonferroni correction was utilized. Qualitative data were analyzed using either a chi-square test or Fisher’s exact test, and the data are reported as frequencies or percentages. A nonparametric rank-sum test was used to analyze multiple sets of hierarchical data, and the data are expressed as quantities or percentages. Statistical significance was determined at P < 0.05. We incorporated Bonferroni adjustments for multiple testing. There were three pairwise comparisons in this study, so the adjusted α was 0.05/3 = 0.017.

Sample Size Calculation

PASS (version 15) was used to calculate the sample size. The mean score of satisfaction was based on the data of the phase III study evaluating the efficacy and safety of remimazolam (CNS 7056) compared with placebo and midazolam in colonoscopy, which was 9.6 (Brice VAS satisfaction questionnaire) [2]. A 5-mm difference in the PSSI was detected with 90% power at a significance level of α = 0.05, so approximately 30 patients were required in each group. Considering a loss to follow-up and dropout rate of 15%, the sample size increased by 15%, and the final target sample size was 35 patients per group. Therefore, this study required 105 participants.