Propofol Versus Remimazolam in Gastrointestinal Endoscopy: A Comprehensive Systematic Review and Meta-Analysis of Randomized Controlled Trials

Abstract

Background

Propofol is the most commonly used intravenous anesthetic for endoscopic surgery, although it comes with various adverse effects. Research indicates that Remimazolam, which is considered a safe general anesthetic, is being increasingly adopted as an alternative to propofol in clinical settings. Our meta-analysis sought to evaluate whether the rate of adverse reactions associated with Remimazolam in gastrointestinal endoscopic procedures is acceptable, and whether its surgical success rate is at least comparable to that of propofol.

Methods

Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we searched five electronic databases (PubMed, Scopus, Cochrane Library, Embase, and Web of Science) to identify eligible studies published up to January 2025. Using R version R.4.4, we reported outcomes as risk ratios (RRs) or mean differences (MDs) and confidence intervals (CIs). A P-value of ≤ 0.05 is considered statistically significant.

Results

Our meta-analysis included 37 studies encompassing 8533 patients. The results, analyzed using a random effects model, demonstrated no statistically significant difference in induction time between the two sedative groups, with an overall MD of 0.11 min (95% [- 0.10; 0.31], p = 0.2977). Patients in the Remimazolam group experienced longer total sedation times than those receiving Propofol, with an MD of 1.84 min (95% CI [0.61; 3.07], p = 0.0033). The pooled analysis indicated low heterogeneity (I² = 4.2%). There was no statistically significant difference between the two sedatives in procedure time, with an MD of 0.16 (95% CI: [−0.13; 0.45], p = 0.2689). The pooled analysis displayed moderate heterogeneity (I² = 43.5%, p = 0.0052).

Conclusion

Remimazolam has demonstrated safety, showing a reduced incidence of hypotension, bradycardia, respiratory depression, and hypoxemia compared to propofol. However, efficacy outcomes including the induction time and total procedure time were comparable between the two groups. However, the results were heterogeneous, which could be due to variability in the perfumed procedures, co-analgesics or dosing.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10620-026-09678-z.

Introduction

Gastrointestinal (GI) diseases are highly prevalent and represent a global health burden [1]. Thus, there is mounting attention to early diagnostic and therapeutic interventions such as endoscopy [2]. Endoscopic procedures are a cornerstone in the early detection and ongoing management of common GI conditions, such as peptic ulcers, gastroesophageal reflux disease (GERD), inflammatory bowel diseases (IBD), strictures, and various GI malignancies [3].

Moreover, these minimally invasive approaches are milestones not only for visualization and biopsy of the GI tract but also to enable subsequent therapeutic interventions [4, 5]. GI procedures are inherently irritating and often associated with abdominal discomfort, cramping, gagging, and choking sensations. Their fear and anxiety may limit patients’ readiness to undergo an endoscopy because of the expected discomfort and pain, which may also interfere with the endoscopist's ability to perform the procedure appropriately [6].

Endoscopies require sedation to relieve patients' discomfort and facilitate the endoscopist’s technical operation [7]. Sedation is an essential step in the success of endoscopy, as it enhances patient comfort, reduces anxiety, and improves the quality of the technique [8]. As a result, an emerging focus is on optimizing sedative agents during endoscopic procedures, and among these agents [9]. The best analgesics and sedatives should have a high safety profile, maximize cost-effectiveness, act rapidly with rapid offset, and not accumulate [10]. Nowadays, sedatives and analgesic medications include opioids, propofol, midazolam, etomidate, and dexmedetomidine [9].

Propofol enhances the activity of gamma-aminobutyric acid (GABA) at the GABA-A receptor and inhibits the N-methyl-D-aspartate (NMDA) receptor. As a result, it achieves hyperpolarization of neuronal membranes, leading to sedation, hypnosis, and anxiolysis [11]. It is commonly used as it offers a rapid onset of action, a short duration, and effective deep sedation [11]. On the other hand, it has potential challenges, including dose-dependent circulatory depression, hypotension, respiratory suppression, injection pain, lack of analgesic properties, and possible allergies [11].

Remimazolam is a promising new intravenous ultra-short-acting benzodiazepine with a highly selective affinity ligand for the benzodiazepine site on GABA receptors. It stimulates the chloride channel opening, induces cell membrane hyperpolarization, and interferes with GABA reuptake like midazolam, but with a greater potency and maximum effect [12]. Moreover, rapid independent organ metabolism into inactive metabolites makes it safe for hepatic and renal impairment [12].

Although recent evidence supports Remimazolam's superior safety profile, several gaps in knowledge and uncertainties remain regarding its actual efficacy and other emerging side effects [13, 14]. Thus, we conducted our study to compare the safety and efficacy of Remimazolam with Propofol, providing adequate clinical evidence of debatable outcomes.

Methods

This systematic review and meta-analysis was reported using the format recommended by the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [15]. We followed the Cochrane Handbook of Systematic Reviews and Meta-analyses [16] steps strictly. The research protocol was submitted to the International Prospective Registry of Systematic Reviews (PROSPERO) database with this ID: CRD42025638248.

Search Strategy and Study Selection

The included studies were selected through a comprehensive search across five electronic databases (PubMed, Web of Science, Cochrane, Scopus, Embase) from inception to Jan 9, 2025, using the following keywords: Remimazolam, Propofol, endoscopy, and other alternative terms. We provided a detailed search strategy and the different terms in the supplementary material. All the extracted studies from the systematic databases search were subsequently exported into EndNote Reference Library to remove the duplicates. Then, two authors independently screened the retrieved studies based on our eligibility criteria by titles and abstracts. We solved the conflicts through full-text screening by another author. The references for the included studies were screened manually for more relevant studies.

Eligibility Criteria

Inclusion Criteria

We included all studies conforming to the following criteria: All randomized controlled trials (RCTs) included patients underwent endoscopic procedures (Colonoscopy, gastroscopy, etc.) comparing Remimazolam to Propofol.

Exclusion Criteria

We excluded observational studies including: cohort, case–control, case series, and case reports, reviews, conference abstracts, editorial correspondence, book chapters, animal and laboratory experiments, and in vitro inquiries. We also excluded studies that use combined therapy (such as propofol or remimazolam together or plus another sedative) and non-English publications.

Data Extraction

Two authors independently extracted the outcomes of interest for our systematic review and meta-analysis, including induction time, sedation time, sedation success, procedure time, time to discharge, time to recovery, duration of Post-anesthesia Care Unit (PACU) stay, time to extubation, time to eye-opening, and the satisfaction of the patient, endoscopist, and anesthetist. The summary and study characteristics included study ID, age, sex, sample size, type of procedure, Remimazolam dose, Propofol dose, combined analgesia, body mass index (BMI), the American Society of Anesthesiologists (ASA) grade, oxygen saturation, heart rate (HR), and comorbidities. Disagreements were resolved through discussion. Primary outcomes included Induction time (min), Procedure time (min), Sedation success, Time to recovery (min), and Adverse events. A detailed definition of each outcome is provided in each included study (Table S1). We utilized WebPLotDigitizer to extract the data from figures. Data was extracted by two blind authors; the third author resolved any disagreement.

Risk of Bias in Individual Studies

Two authors independently conducted the quality assessment of the included RCTs using the Cochrane assessment tool designed for (RCTs) (ROB2) [17]. All included studies were evaluated based on their randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection bias within the reported results. We comprehensively screened all included studies and subsequently rated them as “low risk,” “moderate risk,” or “high risk” of bias.

Statistical Analysis

A meta-analysis will quantitatively synthesize the results from the studies included, utilizing suitable statistical methods. Depending on heterogeneity, treatment effect estimates will be pooled using either random (restricted maximum likelihood) or fixed-effects models (Mantel–Haenszel method). The heterogeneity is considered significant if the I² ≥ 50%, or the P-value < 0.05. Subgroup and sensitivity analyses will examine heterogeneity sources and evaluate the robustness of the findings. Publication bias will be assessed using funnel plots and Egger's test. All analyses will be performed using R software version 4.4, maintaining a significance threshold of p < 0.05.

Results

Literature Search Results

Database searches identified 2304 studies. After removing duplicates, 1500 articles were included for further evaluation. Following a detailed abstract and full-text screening of these studies, 37 studies were included in our meta-analysis [12, 18–53], as shown in the PRISMA flow diagram Fig. 1.

Fig. 1.

PRISMA flow diagram

Included Studies' Baseline Characteristics

The thirty-seven studies encompassed a total of 8533 patients. The included studies were published between 2020 and 2024 and conducted in China and South Korea. Upper GI procedures were the most commonly performed ones. Remimazolam doses ranged from 0.05 mg/kg to 0.4 mg/kg, while Propofol doses ranged from 0.5 mg/kg to 2.5 mg/kg. The most common co-analgesics were Sufentanil, Alfentanil, Fentanyl, and Remifentanil, Table 1,2.

Table 1.

Summary of the included

Study ID	Country	Study design	Total sample size	Type of procedure	Remimazolam dose	Propofol dose	analgesia combined	Inclusion criteria	Exclusion criteria	Conclusion
X. Cui 2023	China	RCT	218	Upper GI Endoscopy	R1: 0.2 mg/kg R2: 0.3 mg/kg R3: 0.4 mg/kg	2.0 mg/kg	Lidocaine hydrochloride 0.1 g	Aged between 18 and 60 years, undergoing routine gastroscopic diagnosis and treatment, ASA physical status of Class I or II, 18 kg/m2 < BMI < 30 kg/m2, expected gastroscopic operation time not more than 15 min, Clear understanding and voluntary participation in the study with informed consent	Requiring complex endoscopic techniques, scheduled for intubation, respiratory management difficulties (modified Mahalanobis score IV), uncontrolled hypertension (systolic ≥ 160 mmHg, diastolic ≥ 100 mmHg), sitting systolic blood pressure ≤ 90 mmHg, pregnant/lactating women, allergic to benzodiazepines, propofol, dyclonine, or their components, deemed inappropriate by the investigator	Satisfactory efficacy can be obtained with higher concentrations of remimazolam tosilate in patients undergoing upper gastrointestinal endoscopy with ASA grade I or II Remimazolam tosilate has less effect on hemodynamics, but as the dose is progressively increased, the adverse reactions caused by higher concentrations (0.4 mg/kg) are also significantly increased, such as nausea, vertigo, and prolonged sedation recovery time The general applicability of remimazolam tosilate needs to be confirmed by further clinical trials with other types of surgery
J. Guo 2022	China	RCT	77	GI endoscopy	0.15 mg/kg	1.5 mg/kg	Alfentanil 5 µg/kg	Aged ≥ 65 years ASA grade I-II BMI of 18.0 to 30.0 kg/m 2	Severely difficult airways, apnoea syndrome, uncontrolled hypertension, abnormal liver or kidney function, history of opioid or analgesic abuse, contraindications to benzodiazepines, opioids, or propofol, changes in endoscopic procedures due to gastrointestinal bleeding or perforation	Compared with propofol, Remimazolam tosilate (RT) can be safely and effectively used for gastrointestinal endoscopy sedation in elderly patients. It is associated with a lower incidence of sedation-related adverse reactions, especially haemodynamic events and respiratory depression. However, the number of supplemental doses after successful induction may increase slightly with RT
B. Hu 2022	China	RCT	346	Gastroscopy	0.2 mg/kg	1.5 mg/kg	Sufentanil 0.1 µg/kg	Patients aged ≥ 65 years American Society of Anesthesiologists (ASA) grade I–III Body mass index (BMI) between 18 and 30 kg/m 2 Scheduled for elective gastroscopy with sedation anesthesia	Severe cardiorespiratory instability (e.g., acute heart failure, myocardial infarction, obstructive sleep apnea, acute respiratory infection), requires tracheal intubation (including laryngeal mask), history of renal or hepatic dysfunction, neurological diseases, known allergy or contraindication to benzodiazepines, opioids, propofol, egg products, or soybeans, history of alcohol or opioid addiction, inability to complete the survey, expected difficult airways	Compared with propofol, the incidence of respiratory depression and other sedation-related adverse events caused by RT are significantly lower. RT may be a suitable alternative sedative agent for elderly patients undergoing gastroscopy due to its non-inferior efficacy and higher safety profile
S. Li 2023	China	RCT	116	Colonoscopy	0.05–0.15 mg/kg	0.5–1.5 mg/kg	Sufentanil 0.1 µg/kg	Patients undergoing painless colonoscopy Ages 18–65 years ASA PS I-II (American Society of Anesthesiologists Physical Status classification I or II)	Patient refusal, history of brain surgery, diabetes with fasting blood glucose ≥ 11.1 mmol/L, recent myocardial infarction or unstable angina, bradycardia (HR ≤ 50 bpm), malignant arrhythmia, third-degree AV block, systolic BP ≤ 90 mmHg, mental disease or cognitive dysfunction, substance abuse, abnormal coagulation, bleeding tendency, liver or renal dysfunction, allergy to tested drugs, pregnant/breastfeeding, participation in other trials within 3 months	compared to propofol, remimazolam could offer more stable hemodynamics in adult patients undergoing painless colonoscopy at the same depth of sedation. Despite the continuous pump infusion, remimazolam tosilate still has higher circulatory stability than propofol in painless colonoscopy. Besides, remimazolam might significantly decrease the incidence of adverse effects such as respiratory depression, nausea, and arrhythmia. Therefore, we conclude that remimazolam is an effective and safe agent in patients undergoing outpatient colonoscopy
D. Li 2024	China	RCT	166	GI endoscopy	0.2 mg/kg	1.5 to 2.0 mg/kg	Alfentanil 7 µg/kg	Aged between 18 and 75 years American Society of Anesthesiologists (ASA) classification I to II Mallampati classification I to II Body mass index (BMI) between 18 and 30 kg/m 2 Expected gastrointestinal endoscopy duration ≤ 20 min	Severe heart, lung, or metabolic disease (liver/kidneys), gastrointestinal obstruction or dyskinesia, narcotic or sedative abuse, hypersensitivity to emulsions, opioids, or benzodiazepines, procedure time > 20 min	Compared to propofol with alfentanil, remimazolam with alfentanil can be used safely and effectively for sedation in patients undergoing gastrointestinal endoscopy, with less impact on the patient's circulatory and respiratory systems and a lower incidence of adverse events
F. Liu 2023	China	RCT	235	GI endoscopy, Colonoscopy, Gastroscopy	0.1 mg/kg	1.5 mg/kg	Sufentanil 5 ug	Patients aged 60–80 years old American Society of Anesthesiologists (ASA) physical status I-II Body mass index (BMI) of 18–30 kg/m 2 Signed informed consent	Diagnosed heart or lung disease (arrhythmia, heart failure, COPD, etc.), previous hypotension (systolic BP ≤ 90 mmHg), bradycardia (HR < 50 bpm), hypoxemia (SpO2 < 90%), contraindications to gastroscopy, underlying diseases requiring oxygen, upper respiratory infection, allergy to propofol or sedatives, expected procedure time > 40 min, non-consent to participate	For patients aged 60–80, compared with propofol, the application of remimazolam for sedation during gastrointestinal endoscopy significantly reduced the incidence of hypoxemia. Remimazolam increases the safety of gastrointestinal endoscopy in elderly patients and may be a new option for sedation medication
Y. Tan 2021	China	RCT	99	Upper GI endoscopy	R1: 0.1 mg/kg R2: 0.2 mg/kg	1.0–1.5 mg/kg	Lidocaine 0.2 g, Butorphanol 0.01 mg/kg	Patients aged > 60 years undergoing upper GI endoscopy American Society of Anesthesiologists physical status I or II Sufficient Chinese language proficiency to complete neuropsychological assessments	Clinically significant cardiorespiratory instability, renal or hepatic dysfunction, history of drug/alcohol abuse, psychiatric illness, neurological disease, auditory/visual disturbances affecting neuropsychological assessments, pre-existing memory or cognitive impairment, allergy to anesthetic agents, anesthesiologist refusal of anesthetic use	Remimazolam tosilate (RT) at a dose of 0.1 mg/kg is a safe and effective sedative for upper GI endoscopy in patients of advanced age. The addition of RT as an adjunct to opiate sedation for upper GI endoscopy can lead to more stable perioperative hemodynamics and achieve acceptable neuropsychiatric functions. To minimize adverse impacts on cognitive function, 0.1 mg/kg RT should be considered for sedation in elderly patients undergoing upper GI endoscopy
A. Wei 2023	China	RCT	304	Upper GI endoscopy	RT/R + P;: 0.2/0.1 mg/kg	P/R + P: 2/0.5 mg/kg	Fentanyl 0.5 µg/kg	Aged 18–80 years; 2. ASA status I or II; 3. BMI 18–30 kg/m 2	1. BP > 180/110 or < 90/60; 2. Difficult airway/abnormal anesthesia history; 3. Severe heart, brain, lung, liver, or kidney disease	Co-administration had fewer adverse events than propofol monotherapy, better sedation, and higher endoscopist satisfaction vs remimazolam
B. Zhou 2024	China	RCT	102	ERCP	0.2–0.4 mg/kg	1–2.5 mg/kg	Remifentanil hydrochloride 0.3 µg/kg/h	ASA Class I-III, adult patients undergoing ERCP under MAC, and operation time between 30 min and 4 h	Allergic or contraindicated to contrast agents, remimazolam, propofol, or other drugs; BMI ≥ 27 kg/m 2; cardiac insufficiency (EF < 48%) or severe obstructive ventilatory dysfunction; pregnant or lactating women; mental disorders or epilepsy; and difficult airways (Mallampati > III) or high risk of regurgitation/aspiration	Remimazolam-based MAC for ERCP exhibited non-inferior efficacy compared to propofol-based MAC, with fewer circulatory and respiratory adverse events
J. Lee 2023	South Korea	RCT	110	ERCP	5.0 mg	0.5 mg/kg		> 18 years old, scheduled for diagnostic or therapeutic ERCP with or without a CBD stone or bile duct obstruction, American Society of Anesthesiologists Physical Status Classification System risk class of 1–3, a body mass index of < 40 kg/m 2, provision of informed consent	Known sensitivity to benzodiazepines, flumazenil, or a medical condition contraindicating these agents, chronic use of benzodiazepines for any indication, pregnancy, or a positive screening result for drugs of abuse	Remimazolam is not inferior to propofol in achieving successful ERCP completion without rescue medication. Incidences of adverse events were comparable. Remimazolam is a safe and effective alternative to propofol for ERCP sedation
X. Xiao 2024	China	RCT	80	ERCP	0.2 mg/kg	1.5 mg/kg	Remifentanil 0.15 ug/kg/min	Adult patients (≥ 18 years old) with ASA physical status I to III scheduled for elective therapeutic ERCP under general anesthesia	Emergency ERCP, uncontrolled hypertension (> 160/100 mm Hg), severe cardiovascular disease, an anticipated difficult airway, morbid obesity, drug abuse, communication difficulties, recent respiratory infections, allergy, intolerance to general anesthetics, breastfeeding or pregnant women, and participation in other clinical studies within the previous 3 months	Remimazolam produced a milder change in MAP during induction and a smaller AUB through the procedure compared to propofol during GA for therapeutic ERCP. Remimazolam may be considered a better choice for complex endoscopic procedures in the outpatient setting, with the potential advantage of stable hemodynamics
D. Chen 2024	China	RCT	240‬	Gastroscopy	0.170 mg/kg	1.112 mg/kg	Sufentanil 0.1 µg/kg	Aged 65–80 years, BMI 16–35 kg/m2, able to undergo routine gastroscopy, ASA grade I–II, Mallampati class I–II, no serious heart, lung, brain, liver (CHILD A), kidney (GFR > 90), or other organ diseases, no schizophrenia or severe depression	Allergic to propofol, sufentanil, benzodiazepines, or study components; myasthenia gravis; depressive state or schizophrenia; severe respiratory diseases (e.g., COPD, bronchial asthma, obstructive sleep apnea); poorly controlled severe hypertension; HR < 50 bpm with positive atropine test or third-degree AV block; sinus tachycardia (HR > 120 bpm) or severe arrhythmia; pyloric or high intestinal obstruction; long history of alcohol or drug abuse; contraindications for intravenous anesthesia	In summary, the ED95 of 0.1 µg/kg sufentanil combined with remimazolam or propofol in inhibiting responses during gastroscopy among the elderly were 0.163 mg/kg (95% CI 0.160–0.170 mg/kg) and 1.042 mg/kg (95% CI 1.007–1.112 mg/kg), respectively. Moreover, when 0.170 mg/kg remimazolam and 1.112 mg/kg propofol were administered, there was no signifcant diference in the anesthetic efective rate between the two groups in inhibiting body movement response during gastroscopy in elderly patients. However, remimazolam enabled stabler circulatory function and fewer adverse events. Terefore, it is more benefcial than propofol when applied in gastroscopy among the elderly
B. Deng 2024	China	RCT	702	Colonoscopy	0.15 mg/kg	1.5 mg/kg		Age 65–85 years, ASA classification I–II, BMI 18.5–23.9 kg/m2	Impaired liver or kidney function, history of alcohol or opioid abuse, recent use of benzodiazepines or analgesics (within 3 months), known allergies or contraindications to trial medications, patients requiring intubation or at high risk for airway complications (e.g., Mallampati score 3 or 4), presence of psychiatric conditions or uncooperative attitude	Remimazolam combined with esketamine offers superior hemodynamic stability and significantly reduces adverse events rates compared to propofol plus esketamine in elderly patients undergoing painless colonoscopies, while maintaining effective sedation
S. Lin 2024	China	RCT	228	Colonoscopy	0.2 mg/kg	1 mg/kg	Sufentanil 0.05 μg/kg	Eligible participants were adults aged 65 years or older with an American Society of Anesthesiologists (ASA) physical status classification of I–III who were scheduled for elective outpatient colonoscopy	Refusal to participate, allergy to study drugs, inability to ambulate independently, history of alcohol or substance abuse, cognitive impairment, psychiatric or neurological disorders, BMI ≥ 30 kg/m2, inability to communicate in Mandarin, need for therapeutic endoscopic interventions, severe cardiopulmonary, hepatic, or renal dysfunction	In elderly patients undergoing colonoscopy, remimazolam demonstrated comparable cognitive recovery to propofol, with higher patient satisfaction and a more favorable safety profile. Remimazolam may be the preferred alternative to propofol for procedural sedation in this vulnerable population
L. Lu 2024	China	RCT	351‬	GI Endoscopy	0.15 mg/kg	1 mg/kg	Fentanyl 1 µg/kg	Age ≥ 65 years, ASA I–II, BMI 18.0–30.0 kg/m2, scheduled for gastrointestinal endoscopy under sedation	Allergy to anesthetics, alcohol/drug abuse, cognitive or psychiatric disorders, respiratory infections, airway challenges, severe cardiovascular conditions, complex endoscopic procedures, recent participation in other studies, inability or unwillingness to cooperate	This non-inferiority study revealed that in elderly outpatients undergoing gastrointestinal endoscopy, remimazolam achieved recovery times comparable to propofol, with fewer associated complications
Q. Sun 2024	China	RCT	113	Colonoscopy	A = 0.2 mg/kg B = 0.25 mg/kg	2.0 mg/kg	Sufentanil 5 µg	ASA Grade I–II, BMI 18–28 kg/m2, age 18–65 years, undergoing routine colonoscopy	Pregnant or lactating women and patients (including men) with birth plans within 3 months; Patients with obvious respiratory and circulatory dysfunction, abnormal blood routine and blood biochemical indexes before operation; People suffering from serious neuropsychiatric diseases; Take benzodiazepines or opioids intermittently in the last three months or every day within one month; Allergies or contraindications to benzodiazepines, opioids or drugs used in this study and their drug components; Patients who were judged to have difficulty in respiratory tract management: improved Markov classification III and above; Before anesthesia induction, BIS < 90	In this trial, remimazolam (Remimazolam besylate) provided safe and effective sedation for colonoscopy and reduced the occurrence of adverse reactions such as hypotension, respiratory depression, and injection pain The initial dose of remimazolam of 0.25 mg/kg and the supplementary dose of 2.5 mg/kg can achieve satisfactory sedation induction and maintenance effects
S. Dong 2023	China	RCT	505	ERCP	0.3 mg/kg	1.5 to 2 mg/kg	alfentanil 10 μg/kg	ASA-I to III adults undergoing elective ERCP under deep sedation	patients with abnormal renal function, previous abnormal surgical anesthesia recovery history, hypertension (> 160/95 mmHg when the patient admission to the operating room), BMI < 18 or BMI > 30, taking monoamine oxidase inhibitor or antidepressant within 15 days, chronic pain with long-term use of analgesics or psychotropic substances, alcohol abusers, with known drugs allergy, anticipated difficult airway, pregnant or parturient women, involved in other drug trials within three months, or could not communicate well with the researchers	this randomized, controlled trial involving a comparison of remimazolam and propofol when administer in combination with alfentanil in patients undergoing elective ERCP under deep sedation showed that the incidence of hypoxia (SpO2 < 90%) was significantly lower in the remimazolam-alfentanil group compared to the propofol-alfentanil group. Remimazolam-alfentanil also appeared to possess some hemodynamic advantages over propofol-alfentanil. Future studies should compare the two sedative-hypnotic drugs when administered with small doses of fentanyl for procedural sedation
K. Lu 2022	China	RCT	400	upper GI endoscopy	300 mg/h	3000 mg/h	fentanyl 50 μg	Elderly patients aged 65–85 years scheduled for upper gastrointestinal endoscopy	ASA IV +, BMI < 18 or > 30 kg/m2, tracheal intubation or difficult airways, acute respiratory infection, uncontrolled hypertension or hypotension, hemoglobin < 80 g/L, suspected GI bleeding, perforation, or obstruction, recent drug abuse/alcoholism, psychiatric disorders, allergy to study medications, recent participation in other trials, procedure time > 30 min, investigator-determined ineligibility	Remimazolam was associated with a lower rate of hypotension in elderly patients undergoing upper gastrointestinal endoscopy under deep sedation/ anaesthesia than propofol
Y. Qiu 2022	China	RCT	56‬	ESD	0.3 mg/kg	2.0 mg/kg	sufentanil 0.5 µg/kg	Male and female subjects aged 18–80 years, ASA I–III, BMI 18–30 kg/m2	Uncontrolled hypertension or hypotension, coronary heart disease, heart failure, severe respiratory disease, severe bradycardia, heart block, arrhythmias, coagulopathy, end-stage liver or kidney disease, emergent surgeries, peripheral artery disease, investigator-determined ineligibility, and unsuitable for rapid extubation with ESD for superficial pharyngeal carcinoma	Remimazolam has better hemodynamic stability than propofol in adult patients undergoing ESD surgery. The benefits of remimazolam on hemodynamic stability and hypotension prevention may be partly contributed to its better preservation of cardiac output
W. Shi 2022	China	RCT	161	Gastroscopy	0.33 mg/kg	2.5 mg/kg	alfentanil 5 μg/kg	The study inclusion criteria were as follows: age 18 to 60 years, body mass index of 18 to 25 kg/m2, and American Society of Anesthesiologists grade (ASA) I to II	Contraindications to general anesthesia, cardiac function > grade III (NYHA), history of severe cardiovascular disease, acute respiratory infection, asthma attack, severe laryngeal disease, uncontrolled hypertension (SBP ≥ 180 mmHg, DBP ≥ 110 mmHg), alcoholism, long-term sedative or opioid use, liver, kidney, or renal dysfunction, psychiatric disorders, pregnancy or lactation, procedures > 30 min	The ED95 of remimazolam was 0.33 mg/kg when it was combined with alfentanil (5 μg/kg) for gastroscopy sedation. The sedation strategy of remimazolamalfentanil has noninferior efficacy, fewer adverse effects, and a better postoperative recovery process than propofol-alfentanil for patients undergoing gastroscopy
J.W Choe 2024	South Korea	RCT	400	EUS	10.8 ± 7.7 mg	88.0 ± 49.1 mg		Age ≥ 18 years, undergoing diagnostic EUS, ASA I–II, ability to provide informed consent	Exclusion criteria encompassed those who declined participation, intended to undergo therapeutic EUS procedures, were pregnant, were allergic to benzodiazepines or propofol, had a history of severe allergic reactions to any medication, or had contraindications to EUS. Eligible patients were randomly assigned to receive either remimazolam or propofol as sedatives	Remimazolam was superior to propofol in terms of safety and efficacy during EUS examinations
k. Zhang 2023	China	RCT	261	GI endoscopy	0.1 mg/kg	1 mg/kg		Age 18–65 years, BMI 30–40 kg/m2, ASA PS grade II–III, no history of cardiac insufficiency (left ventricular ejection fraction > 55%, E/A > 1, tricuspid annular plane systolic excursion > 13 mm), able to communicate normally	Exclusion criteria were known allergy to planned medications, severe hypertension, history of myocardial infarction, severe hepatic or renal dysfunction, mental illness, pregnancy, epileptic disorder, elevated intracranial pressure, use of medication that affects the central nervous system, or modified Mallampati class IV. Secondary exclusion criteria were termination of the clinical protocol for any reason and patients request to withdrawal	concluded that compared with propofol combined with esketamine, remimazolam combined with esketamine can reduce the incidence of severe hypoxemia during gastrointestinal endoscopy in obese patients
X. Liu 2021	China	RCT	232	Colonoscopy	0.15 mg/kg	0.1 ml/kg	Fentanyl citrate 0.5 µg/kg	Patients who underwent colonoscopy with sedation anesthesia who were aged between 65 and 75 years were recruited	Contraindications to endoscopy (e.g., respiratory infection, asthma, liver dysfunction), ASA ≥ III, tracheal cannula requirement, sensitivity to study drugs, adrenocortical insufficiency, porphyria, chronic corticosteroid use, difficult airways, hemodynamic instability (SBP < 90 or > 180 mmHg, DBP > 110 mmHg, SpO2 < 90%), sleep apnea, procedure > 30 min, drug abuse, participation in other trials in last 3 months	Remimazolam may have non-inferior efficacy and a higher safety profile than Etomidate-propofol in elderly outpatients undergoing colonoscopy, which suggests that RT may be more suitable for elderly outpatients undergoing colonoscopy
Y.Cao 2022	China	RCT	148	Gastroscopy	0.107 mg/kg	2 mg/kg	Sufentanil 0.15 μg/kg	Liver cirrhosis, aged 18–65 years, ASA physical status II to III, Child–Pugh class A	History of respiratory or cardiovascular disease, renal dysfunction, known allergy to medications, inability to complete satisfaction survey, refusal to participate	remimazolam tosilate combined with adjuvant sufentanil is a better alternative to propofol with sufentanil for sedation in gastroscopy for Chinese patients with liver cirrhosis
Shao.Chen 2020	China	RCT	384	Colonoscopy	Initial dose 5 mg, then up to 5 top-up doses of 2.5 mg	Initial dose 1.5 mg/kg, then up to 5 top-up doses of 0.5 mg/kg	Fentanyl 1 ug/kg	Patients undergoing diagnostic or therapeutic colonoscopy	patients requiring endotracheal intubation or a laryngeal mask, or those undergoing complex endoscopic procedures. Individuals with acute heart failure, unstable angina pectoris, or myocardial infarction within six months prior to screening were excluded	Remimazolam tosilate is non-inferior in sedation efficacy while safer than propofol in patients undergoing colonoscopy
H.Chen 2024	China	RCT	60	GI endoscopy	11.2 ± 4.4 mg	140.7 ± 43.2 mg	sufentanil	Elderly patients aged 65–95 years, BMI 19–27 kg/m2, ASA physical status I to III, undergoing gastrointestinal endoscopy	Known allergy to study drugs, chronic use of sedative or opioid analgesics, severe respiratory disease (breath-hold test < 20 s), heart failure (ejection fraction < 40%)	Remimazolam with flumazenil antagonism results in shorter recovery time for elderly patients undergoing gastrointestinal endoscopy compared to propofol
SH.Chen 2020	China	RCT	378	Upper GI endoscopy	Initial dose 5 mg, up to 5 top-up doses of 2.5 mg	Initial dose 1.5 mg/kg, up to 5 top-up doses of 0.5 mg/kg	lidocaine viscous oral liquid 10 g (0.2 g lidocaine) and fentanyl (0.5 μg/kg)	Aged 18–60 years, ASA physical status I–II, BMI 18– < 30 kg/m2, undergoing upper GI endoscopy expected to take ≤ 30 min	Patients with uncontrolled hypertension, pregnant or lactating, history of drug abuse/alcoholism, known sensitivity to study drugs, participating in other trials	Remimazolam tosilate has a non-inferior sedation success rate compared with propofol, allows faster recovery, and has a superior safety profile
F.Shi 2022	China	RCT	76	Endoscopic variceal ligation	induction bolus:0.2 mg/kg maintenance dose: 1.0–2.0 mg/kg/h	induction bolus: 2 mg/kg, maintenance dose: 4–10 mg/kg/h	Fentanyl citrate	Patients with ASA physical status II or III undergoing endoscopic variceal ligation	Patients with hemorrhagic shock, alcohol use in the past 2 weeks, or refusal to sign informed consent	Remimazolam tosilate provides satisfactory anaesthetic effects with rapid recovery and decreased incidence of hypotension and low SpO2, making it a safer option than propofol for cirrhotic patients
X.Wang 2022	China	RCT	480	Colonoscopy	Initial dose: 7 mg Top-up doses: 2.5 mg	Initial dose: 1.5 mg/kg. Top-up doses: 0.5 mg/kg	Fentanyl 50 µg	Aged 18–65 years, BMI 18–28 kg/m2, ASA physical status I–III, scheduled for diagnostic or therapeutic colonoscopy, voluntarily provided written informed consent	Allergy or contraindication to benzodiazepines, opioids, propofol, flumazenil, naloxone or their components, severe respiratory disease, abnormal liver or kidney function, difficult airway (modified Mallampati ≥ III), craniocerebral injury, mental illness or cognitive impairment, resting ECG heart rate < 50 bpm, QTc male ≥ 470 ms, female ≥ 480 ms, third-degree AV block, severe arrhythmia, or moderate to severe heart valve disease	Remimazolam besylate had a better safety and tolerability profile and similar sedative efficacy to propofol in patients undergoing a diagnostic or therapeutic colonoscopy in China, suggesting that remimazolam besylate has potential as a sedative agent for colonoscopy
Y.Xin 2022	China	RCT	114	Colonoscopic polypectomy	RT1: 0.1 mg/kg RT2: 0.15 mg/kg RT3: 0.2 mg/kg Additional 2.5 mg doses of remimazolam given as needed	2 mg/kg	Alfentanil 10 μg/kg	Aged 18 to 80 years, American Society of Anesthesiologists (ASA) status I or II, Operation time 20 to 60 mi, Body Mass Index (BMI) 18.5 to 23.9 kg/m2	Emergency operation, high risk of reflux aspiration, allergy to benzodiazepines or opioids, sedative, analgesic, or antidepressant use within 24 h, pregnant or breastfeeding, abnormal liver or kidney function, history of drug abuse, recent participation in other clinical studies	Remimazolam combined with alfentanil has a non-inferior sedative effect compared to propofol with alfentanil during colonoscopic polypectomy. Moreover, this combination of two short-acting drugs may be a safer alternative, due to the lower incidence of hypotension and faster time to full alert in remimazolam groups
L.Cui 2023	China	RCT	113	Gastroscopy and Colonoscopy	0.3 mg/kg	2 mg/kg	Sufentanyl 0.1 μg/kg	Aged 18–70 years, undergoing gastrointestinal endoscopy (gastroscopy or colonoscopy), ASA classification grade I–II, BMI 18–30 kg/m2, completed informed consent	"Patients with preoperative alcohol or related drug consumption were excluded from this study."	Remimazolam tosilate combined with propofol can be used for painless endoscopy, with almost the same safety as propofol. Remimazolam tosilate produces a low incidence of adverse reactions and is a safe anaesthetic option for painless endoscopies
C.Xu 2022	China	RCT	914	Gastroscopy	0.21 (0.18–0.25) mg/kg,	2.00 (1.00–2.00) mg/kg	Afentanil 10 μg/Kg	Aged 18–85 years, undergoing gastroscopy without sedation, ASA classification grade III	patients undergoing emergency surgery; patients who need en dotracheal intubation for general anesthesia; cases with incomplete data	RB combined with alfentanil can provide safe and effective sedation for patients undergoing painless gastroscopy. RB and alfentanil for injection can avoid large hemodynamic fluctuations and deep sedation, and have fewer adverse reactions than propofol
Y.Yao 2022	China	RCT	132	Colonoscopy	intial dose: 0.2 mg/kg, 6 mg if insufficient	initial dose: 1 mg/kg, 30 mg if insufficient	sufentanyl 5 μg	age(18 to 65) years with an ASA physical status of I to II undergoing ambulatory colonoscopy	a history of allergy to study medications, weight less than 40 kg, BMI greater than 30 kg m 2 obstructive sleep apnoea, chronic use of opioids or benzodiazepines, inability to ambulate without assistance, preexisting cognitive impairment, a suspected or diagnosed disorder of the lower gastrointestinal tract required advanced therapeutic endoscopy, a history of drug or ethanol abuse, pregnancy or lactating female patients	remimazolam provides a noninferior discharge time and associated with less injection pain, lower risks of hypotension and bradycar dia, and improved patient satisfaction
E.Ye 2023	China	RCT	129	Gastroscopy	0.2 mg/kg	2.0 mg/kg	remifentanil 0.2 μg/kg	age ≥ 65 years, undergoing painless gastroscoy, ASA I to II, BMI(18–28 kg/m2)	allergy to benzo diazepines, opioids, propofol, lidocaine, upper respiratory tract infection within 2 weeks, use of benzodiazepines or opioids in the last 3 months, epilepsy, any mental illnesses, a history of addiction to anal gesics or hypnotics, severe cardiac dysfunction (NYHA cardiac functions grade 3–4), a history of myocardial or cerebral infarction within 6 months, arrhythmia, severe renal (serum creatinine ≥ 2 mg/dL or 176.8 μmol/L), or liver dysfunction (Child–Pugh scale grade C)	the effective dose(95%) of remimazolam is a safer alternative than propofol when inducing the same sedation depth
R.Zhou 2024	China	RCT	360	Colonoscopy and Gastroscopy	0.2–0.3 mg/kg	1.5–2 mg/kg	sufentanil (0.06–0.08 μg/kg)	Aged 18–65 years, ASA status I–II, respiration rate 12–18 cycles/min, heart rate 50–100 beats/min, systolic blood pressure ≥ 90 mmHg, diastolic blood pressure ≥ 60 mmHg, SpO2 > 95% on room air	allergies to the related drugs or drug adjuvants, contraindications to deep sedation or anesthesia, medication of opioid analgesics within 1 month, drinking more than 14 units of alcohol per week (1 unit = 360 mL beer or 45 mL spirits with 40% alcohol or 150 mL wine), difficult airway, pregnant or lactating women, serious mental illness or emotional disorder, and disability to finish the interviews	remimazolam was less likely to contribute to intraoperative hypotension and improvement of sleep quality of the 1 st post-operative day than propofol
T.Yang 2024	China	RCT	240	Gastroscopy	0.2–0.25 mg/Kg	1.5–2.0 mg/kg	Dyclonine hydrochloride mucilage	age = (18–65 years), undergoing gastrointestinal endoscopy with sedation, BMI (18 and 28 kg/m2), ASA(I or II)	pregnant or lactating women, patients requiring endoscopic treatment (such as polypectomy and endoscopic submucosal dissection), recent use of drugs affecting gastrointestinal motility (such as morpholine, mosapride, and M1 receptor agonists/inhibitors) or sedative drugs, allergy to benzodiazepines or propofol, and history of abdominal surgery (including gastrointestinal surgery or pelvic surgery)	Remimazolam tosilate mildly inhibits the serum motilin and gastrin levels, potentially facilitating the recovery of gastrointestinal motility
W.Li 2023	China	RCT	150	Colonoscopy	0.15–0.25 mg/kg	1–2 mg/kg	lidocaine 0.5 mg/kg and esketamine 0.15 mg/kg	age (35–65) years, undergoing outpatient colonoscopy, ASA (I or II)	patients with abnormal liver or renal function; known cardiovascular, respiratory or endocrine disease; actively taking a monoamine oxidase inhibitor, sedative, analgesic, hypnotic, antipsychotic, antiemetic or antidepressant; addiction to tobacco or alcohol; known history of drug allergies; an anticipated difficult airway; body mass index (BMI) > 30 kg/m2, high risk of nausea and vomiting; pregnant or breastfeeding; involved in other clinical trials within the past three months; psychosocial disease or cognitive dysfunction; and the inability to cooperate or communicate	Remimazolam besylate has a similar efficacy to propofol when combined with subanesthetic doses of esketamine during outpatient colonoscopy. Remimazolam besylate combined with esketamine resulted in less injection pain and more stable hemodynamics, although it prolonged induction and awakening time

Table 2.

Baseline characteristics of the included studies

Study ID	Groups	Sample size	Age, mean(SD)	Male, n(%)	BMI, mean(SD)	ASA grade	Oxygen saturation, mean(SD)	HR, mean(SD)	Comorbidities n(%)
									Hypertension	Diabetes
Cui 2023	RT 1	54	48.6 (11.9)	33 (61)	24 (2.6)	I-II	NA	81.4 (13.4)	NA	NA
	RT 2	55	49.6 (10.3)	29 (54)	24 (2.7)	I-II	NA	79.4 (12.7)	NA	NA
	RT 3	54	44.6 (12.2)	25 (46)	23 (3.1)	I-II	NA	80.9 (12.4)	NA	NA
	P	55	46.9 (11.4)	38 (69)	24 (4.6)	I-II	NA	78.6 (11.5)	NA	NA
Guo 2022	RT	39	70.4 (3.9)	22 (57.9)	23 (3)	I-II	NA	67.0 (4.1)	16 (41)	4 (10.2)
	P	38	69.1 (4)	25 (64.1)	23 (3.4)	I-II	NA	67.0 (4.1)	16 (42)	5 (13.1)
Hu 2022	RT	173	70.11 (7.37)	69 (39.8)	22.75 (3.15)	I-II	97.67 (0.75)		65 (37.6)	30 (17.3)
	P	173	69.92 (7.57)	72 (41.6)	22.73 (3.23)	I-II	98 (1.5)		53 (39.8)	25 (14.5)
Li 2023	RT	58	48.64 (8.983)	35 (60.3)	23.18 (4.212)	I-II	99.8 (0.7)	99.7 (0.9)	NA	NA
	P	58	47.19 (11.917)	31 (53.4)	24.71 (4.52)	I-II	99.7 (0.9)	99.8 (0.7)	NA	NA
Li 2024	RT	83	54.1 (11.2)	43 (51.8)	22.8 (3.2)	I-II	NA	NA	29 (34.9)	12 (14.5)
	P	83	52.7 (12)	40 (48.2)	23.5 (3)	I-II	NA	NA	28(33.7)	15 (18.1)
Liu 2023	RT	107	67.6 (5.7)	51 (47.7)	23.7 (3.0)	I-II	99 (1.5)	78 (10.5)	NA	NA
	P	109	67.5 (4.9)	51 (46.8)	24.0 (2.6)	I-II	98.67 (1.5)	76 (8.9)	NA	NA
Shi 2022	RT	38	52.74 (4.93)	16 (42.1)	NA	II-III	NA	NA	NA	NA
	P	38	51.61 (5.48)	18 (47.3)	NA	II-III	NA	NA	NA	NA
Tan 2021	RT1	33	66.4 (4.8)	19 (57.6)	22.7 (3.0)	I-II	NA	NA	15 (45.5)	6 (18.2)
	RT2	33	65.5 (5.2)	22 (66.7)	23.4 (3.9)	I-II	NA	NA	11 (33.3)	6 (18.1)
	P	33	66.2 (5.0)	21 (63.6)	23.2 (3.0)	I-II	NA	NA	14 (42.4)	5 (15.2)
Wei 2023	RT	111	44 (13)	50 (45)	22.8 (2.8)	I-II	98 (1)	75 (12)	NA	NA
	RT + P	96	47 (13)	49 (51)	22.6 (2.6)	I-II	99 (1)	74 (12)	NA	NA
	P	97	46 (13)	48 (49)	23.5 (2.9)	I-II	99 (1)	75 (12)	NA	NA
Zhou 2024	RT	51	61.24 (12.15)	23 (45)	22.06 (3.03)	I-II-III	NA	NA	21 (41.2)	10 (19.6)
	P	51	59.98 (12.36)	26 (51)	21.73 (2.68)	I-II-III	NA	NA	23 (45.1)	12 (23.5)
Lee 2023	RT	55	65 (38.06)	32 (58.2)	23.53 (13.55)	I-II-III	NA	NA	28 (50.9)	21 (38.2)
	P	55	63 (49.48)	32 (58.2)	27.97 (19.79)	I-II-III	NA	NA	24 (43.6)	19 (34.5)
Xiao 2024	RT	40	54 (13.84)	22 (55)	20.93 (2.77)	I-II-III	97 (1.54)	77 (10)	0 (0)	NA
	P	40	59 (12.3)	24 (60)	21.47 (2.92)	I-II-III	96.67 (2.31)	78(8)	2 (5)	NA
D. Chen 2024	RT	122	71.9 (5.0)	66 (54)	22.2 (3.0)	I–II	99.29 (0.88)	76.95 (12)	NA	NA
	P	118	71.7 (5.1)	62 (53)	22.5 (2.9)	I–II	99.51 (0.71)	77.31 (14.8)	NA	NA
B. Deng 2024	RT	358	74.68 (6.03)	156 (44)		I–II	NA	78.24 (5.45)	244 (68.16)	91 (25.42)
	P	344	75.21 (5.98)	166 (48)		I–II	NA	77.25 (4.71)	215 (62.50)	67 (19.48)
S. Lin 2024	RT	114	70.67 (6.01)	59 (51.8)	22.9 (2.7)	I–II–III	NA	NA	50 (43.9)	36 (31.6)
	P	114	71.00 (5.26)	63 (55.3)	22.7 (2.6)	I–II–III	NA	NA	46 (40.4)	32 (28.1)
L. Lu 2024	RT	174	69.33 (3.74)	77 (42.8)	24.21 (2.54)	I–II	97 (1)	77.54 (9.02)	85 (47.2)	44 (24.4)
	P	177	69.33 (3.74)	76 (42.2)	24.01 (3.00)	I–II	97 (1)	76.57 (10.0)	88 (48.9)	50 (27.8)
Q. Sun 2024	RT (A)	37	52.5 (12.83)	23 (62.2)	22.9 (2.81)	I–II	99.5 (0.80)	75.0 (14.82)	NA	NA
	RT (B)	38	47.7 (15.01)	16 (42.1)	22.7 (2.63)	I–II	99.3 (0.94)	75.5 (13.27)	NA	NA
	P	38	52.5 (12.36)	24 (63.2)	22.6 (3.32)	I–II	99.4 (0.79)	72.4 (72.4)	NA	NA
S. Dong 2023	RT	250	66.33 (11.93)	144 (57.6)	24.00 (2.28)	I–II–III	99.78 (0.492)	73.30 (8.50)	83(33.2)	18(7.2)
	P	255	67.00 (12.67)	156 (61.2)	23.64 (2.44)	I–II–III	99.84 (0.496)	72.89 (7.46)	101(39.6)	27(10.6)
K. Lu 2022	RT	200	70.6 (4.7)	78 (39.0)	22.2 (2.5)	I–II–III	99.1 (1.3)	74.8 (9.8)	62 (31.0)	14 (7.0)
	P	200	70.1 (4.5)	83 (46.5)	22.2 (2.3)	I–II–III	99.2 (1.2)	75.4 (10.7)	68 (34.0)	16 (8.0)
Y. Qiu 2022	RT	28	62.8 (7.1)	21 (75)	23.2 (3.5)	I–II–III			7 (25)	3 (10.7)
	P	28	64.7 (8.9)	19 (68)	22.3 (2.9)	I–II–III			13 (46.4)	3 (10.7)
W. Shi 2022	RT	81	42.03 (10.51)	35 (43)	22.43 (1.90)	I–II	99.9 (0.48)	75.8 (11.43)	NA	NA
	P	80	45.45 (11.75)	43 (54)	22.66 (1.99)	I–II	100 (0.45)	77.34 (12.03)	NA	NA
J.W Choe 2024	RT	200	57.7 (12.7)	105 (52.5)	23.9 (3.8)	I–II	NA	NA	55 (27.5)	33 (16.5)
	P	200	55.3 (13.1)	91 (45.5)	23.9 (3.3)	I–II	NA	NA	40 (20)	32 (16)
k. Zhang 2023	RT	131	44.67 (8.25)	62 (47.3)	34.50 (3.60)	II-III	97.00 (1.50)	NA	40 (30.5)	16 (12.2)
	P	130	43.33 (9.00)	77 (59.2)	34.57 (3.60)	II-III	97.00 (1.50)	NA	29 (22.3)	18 (13.8)
X. Liu 2021	RT	115	68.9 (2.6)	54	25.4 (2.1)	I–II	NA	NA	38	21
	P	117	69.1 (2.8)	58	24.8 (2.2)	I–II	NA	NA	42	25
Y.Cao 2022	RT	74	47.6 (5.8)	42 (56.8)	NA	II-III	NA	NA	NA	NA
	P	74	48.5 (7.1)	45 (60.8)	NA	II-III	NA	NA	NA	NA
Shao.Chen 2020	RT	194	44.47 (11.67)	73 (37.63)	23.19 (2.92)	I-II	NA	72.25 (9.52)	NA	NA
	P	190	44.43 (11.37)	88 (46.32)	23.21 (2.84)	I-II	NA	73.09 (9.40)	NA	NA
H. Chen 2024	RT	30	68.0 (3.9)	11 (36.7)	23.0 (2.0)	I-II-III	NA	NA	NA	NA
	P	30	69.5 (4.4)	9 (30)	23.3 (3.3)	I-II-III	NA	NA	NA	NA
SH Chen;2020	RT	189	40.78 (11.10)	78 (41.27)	23.00 (3.05)	I-II	NA	73.70 (8.57)	NA	NA
	P	189	41.69 (10.48)	69 (36.51)	23.15 (2.88)	I-II	NA	72.80 (9.10)	NA	NA
F.Shi 2022	RT	38	52.74 (4.93)	16 (42.1)		II-III	NA	NA	NA	NA
	P	38	51.61 (5.48)	18 (47.4)		II-III	NA	NA	NA	NA
X.Wang 2022	RT	360	43 (15.63)	154 (43.1)	22.81 (3.13)	I-III	NA	74.30 (10.42)	NA	NA
	P	120	46 (13.88)	56 (46.7)	22.86 (2.63)	I-III	NA	73.97 (10.13)	NA	NA
Y.Xin (2022)	RT 1	28	54.11 (12.08)	20 (71.42)	22.20 (1.71)	I-II	NA	NA	NA	NA
	RT 2	30	49.70 (8.95)	23 (76.6)	22.43 (1.77)	I-II	NA	NA	NA	NA
	RT 3	29	52.24 (9.80)	19 (65.5)	22.40 (1.67)	I-II	NA	NA	NA	NA
	P	27	56.00 (10.13)	17 (63)	22.28 (1.61)	I-II	NA	NA	NA	NA
L.Cui (2023)	RT	40	48.40(11.20)	20(50)	22.98 (2.38)	I-II	NA	NA	8(20)	4(10)
	P	38	48.40 (12.56)	21(52.5)	22.98 (3.01)	I-II	NA	NA	8(21)	4(10.5)
C.Xu 2022	RT	457	52.69 (13.12)	211 (46.17)	23.97 (3.40)	II	NA	NA	NA	NA
	P	457	52.56 (12.69)	209 (45.7)	23.84 (3.50)	II	NA	NA	NA	NA
Y.Yao 2022	RT	66	48.67 (11.37)	29 (44)	22.47 (3.46)	I-II	99 (1.52)	75 (11)	9 (14)	4(6)
	P	66	47.67 (12.88)	34 (51)	22.47 (3.49)	I-II	99.33 (1.52)	73 (14)	11 (17)	6(9)
E.Ye et al. 2023	RT	64	68.67 (4.55)	39 (60.9)	22.27 (3.79)	I-II	NA	NA	NA	NA
	P	65	68.33 (3.79)	36 (55.4)	22.33 (2.65)	I-II	NA	NA	NA	NA
R.Zhou 2024	RT	120	49.65 (8.90)	54 (45)		I-II	NA	NA	NA	NA
	P	120	48.72 (9.97)	66 (55)		I-II	NA	NA	NA	NA
T.Yang 2024	RT	119	44 (11)	50 (41.6)	23.3 (2.6)	I-II	NA	NA	NA	NA
	P	121	45 (12)	55 (45.8)	23.2 (2.6)	I-II	NA	NA	NA
W.Li 2023	RT	75	50.28 (6.09)	44 (58.6)	25.73 (2.00)	I-II	NA	NA	20 (26.67)	12 (16)
	P	75	50.36 (7.75)	45 (60)	25.26 (1.75)	I-II	NA	NA	22 (29.33)	14 (18.67)

RT Remimazolam Tosilate, P Propofol, BMI Body Mass Index, HR Heart Rate, ASA American Society of Anesthesia

Quality Assessment of the Included Studies

Our assessment revealed that 22 RCTs had a low risk of bias, 14 RCT had some concerns, and one RCT showed a high risk of bias (Fig. 2).

Fig. 2.

Risk of bias of the included randomized controlled trials

Meta-Analysis

Primary Outcomes

Induction Time

Twenty-three studies reported on induction time after administering either Remimazolam or propofol. Using a random effects model, the analysis showed no statistically significant difference between the two sedative groups, with an overall MD of 0.11 min (95% [- 0.10; 0.31], p = 0.2977), Fig. 3. When examining subgroups based on Modified Observer's Assessment of Alertness/Sedation (MOAA/S) score criteria for induction, the results in studies reporting a MOAA/S score of 2 or 3 showed a statistically significant difference, with the MD found to be 0.24 min (95% CI: [0.14; 0.35], p = 0.0002). For studies with an MOAA/S score of 1, there was no statistically significant difference between Remimazolam and propofol, with an MD of −0.16 min (95% CI: [−0.72; 0.40], p = 0.5328). The pooled analysis showed significant heterogeneity for both overall effects, with I² = 97.9%, and subgroup analyses, with I² = 98.2% for MOAA/S scores of 2 or 3 and I² = 97.5% for the MOAA/S score of 1, Fig. 3. A leave-one-out sensitivity analysis was conducted. No study influenced the effect estimate or drove the elevated heterogeneity excessively, Fig.S1.

Fig. 3.

Induction time

Total Sedation Time

Seven studies reported on total sedation time after administering either Remimazolam or Propofol. Using a fixed effects model, the analysis results showed a significant difference in the total sedation times between the two groups, with an MD favoring Propofol over Remimazolam of 1.50 min (95% CI [0.79; 2.22], p < 0.0001), while random effects model yielded an MD of  1.84 min (95% CI [0.61; 3.07], p = 0.0033), indicating that patients in the Remimazolam group had longer total sedation times compared to those receiving Propofol, Fig.S2. The pooled analysis demonstrated low heterogeneity, with I² = 4.2%.

Sedation Success

Twenty-four studies reported on sedation success after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed no significant difference in sedation success of Remimazolam and Propofol. The overall RR is 0.99 (95% CI [0.98; 1.00], p = 0.1644). The pooled analysis showed substantial heterogeneity (I² = 66%, p < 0.0001), Fig. 4. A leave-one-out sensitivity analysis was conducted; this analysis varied slightly in the overall degree of heterogeneity, with I² ranging from 57.6% to 67.4% across different scenarios, and did not significantly alter the overall risk ratio, Fig.S3.

Fig. 4.

Sedation Success

Procedure Time

Thirty-two studies reported on procedure time after administering either Remimazolam or Propofol. Using a random effects model, the analysis showed no statistically significant difference between the two sedatives, with an MD of 0.16 (95% CI: [−0.13; 0.45], p = 0.2689). The pooled analysis indicated moderate heterogeneity (I² = 43.5%, p = 0.0052), Fig. 5.

Fig. 5.

Procedure time

Time to Recovery

Twenty-six studies reported the time to recover after administering either Remimazolam or Propofol. Results analyzed using a random effects model showed no statistically significant difference between the two sedatives, with an overall MD in time to recovery between Remimazolam and Propofol of −1.16 min (95% CI: [−2.47 to 0.16]), (p = 0.0818), Fig. 6. The pooled analysis indicated significant heterogeneity with (I² = 99.6%, p = 0). A leave-one-out sensitivity analysis was conducted, and no study was found to influence the effect estimate or drive the elevated heterogeneity excessively, Fig.S4.

Fig. 6.

Time to recovery

Time to Discharge

Twelve studies reported the time to discharge after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed no statistically significant difference between the two sedatives, with a pooled MD for time to discharge of −1.15 min (95% CI: [−2.78; 0.47], p = 0.1644), Fig.S5. The pooled analysis demonstrated significant heterogeneity (I² = 98.8%, p < 0.0001). A leave-one-out sensitivity analysis was conducted, and no study was found to influence the effect estimate or drive the elevated heterogeneity excessively. The sensitivity of results remained high, with I² values ranging from 86.9% to 98.9%, Fig.S6.

Awake Time

Ten studies reported on awake time after administering either Remimazolam or Propofol. Using a random effects model, the analysis showed no significant difference between the drugs. The pooled MD for awake time was 0.31 min (95% CI −0.47; 1.09], p = 0.4383). The pooled analysis revealed significant heterogeneity (I² = 94.6%, p < 0.0001), Fig.S7. A leave-one-out sensitivity analysis was conducted, and no study was found to influence the effect estimate or drive the elevated heterogeneity excessively, Fig.S8.

Patient Satisfaction

Seven studies reported on patient satisfaction after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed Remimazolam significantly improved patient satisfaction. With an RR of 1.06 (95% CI [0.1.00; 1.11], p = 0.0404), Fig.S9. The pooled analysis revealed significant heterogeneity, with I² = 77.3%, p = 0.0002. A leave-one-out sensitivity analysis was performed, omitting one study at a time. This sensitivity analysis retained the overall high degree of heterogeneity, with I² values ranging from 70.7% to 82.7% across different scenarios, Fig.S10.

Endoscopist Satisfaction

Five studies reported on endoscopic satisfaction after administering either Remimazolam or Propofol. Using a fixed effects model, the analyzed results showed no significant difference in endoscopic satisfaction between Remimazolam and Propofol, with an RR of 1.01 (95% CI [0.97; 1.05], p = 0.7475), Fig.S11. The pooled analysis indicated low heterogeneity with (I² = 20.6%, p = 0.2835).

Adverse Events

A detailed definition of each adverse event as mentioned in each included study, Supplementary Table (S1).

Hypotension

Thirty-five studies reported on hypotension rates after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed that Remimazolam was associated with a significantly lower risk of hypotension compared with Propofol. The overall RR is 0.42 (95% CI [0.36; 0.50], p < 0.0001), which favors Remimazolam over Propofol in preventing hypotensive episodes, Fig.S12. The pooled analysis demonstrated substantial heterogeneity (I² = 67.2%, p < 0.0001). After conducting a sensitivity analysis, heterogeneity was best reduced by excluding the study by C. Xu 2022, leading to moderate heterogeneity with (I² = 45.9), Fig.S13.

Bradycardia

Twenty-three studies reported on bradycardia after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed that sedation with Remimazolam resulted in a significantly lower risk of bradycardia than Propofol. The overall RR = 0.42 (95% CI [0.32; 0.56], p < 0.0001), which favors Remimazolam over Propofol in preventing bradycardia episodes, Fig.S14. The pooled analysis showed low heterogeneity (I² = 24.6%, p = 0.1399).

Postoperative Nausea and Vomiting (PONV)

Eighteen studies reported on the occurrence of nausea and vomiting after administering either Remimazolam or Propofol. Using a random effects model, the analyzed results showed no statistically significant difference between the two sedatives, with an overall RR of 1.18 (95% CI [0.86; 1.62], p = 0.3139), Fig.S15. The pooled analysis indicated no heterogeneity (I² = 0%, p = 0.6869).

Respiratory Depression

Fifteen studies reported on respiratory depression following the administration of either Remimazolam or Propofol. Using a random effects model, the analyzed results showed that patients receiving Remimazolam had a significantly lower risk of experiencing respiratory depression compared to those receiving Propofol, with an RR = 0.37 (95% CI [0.26; 0.52], p < 0.0001), which favors Remimazolam, Fig.S16. The pooled analysis indicated moderate heterogeneity with I² = 36.9%, p = 0.0810.

Hypoxemia

Twenty-four studies reported hypoxemia following the administration of either Remimazolam or Propofol. Using a random effects model, the analyzed results showed a significant difference in the incidence of hypoxemia between the two drug groups, with an overall RR of 0.42 (95% CI [0.33; 0.55], p < 0.0001), suggesting that the use of Remimazolam is associated with a lower risk of hypoxemia compared to Propofol, Fig.S17. The pooled analysis indicated moderate heterogeneity, with I² = 45.5%, p = 0.0086.

Pain of Injection

Twenty-seven studies reported on bradycardia following the administration of either Remimazolam or Propofol. Using a random effects model, the analyzed results showed that the incidence of injection pain was significantly lower with Remimazolam compared to Propofol, with an overall risk ratio (RR) of 0.11 (95% CI [0.07; 0.18], p < 0.0001), demonstrating a substantial advantage for Remimazolam in minimizing pain upon injection, Fig.S18. The pooled analysis revealed considerable heterogeneity, with I² = 69.8%, p < 0.0001.

Body Movement

Twelve studies reported on body movement after administering either Remimazolam or Propofol. Using a random effects model, analyzed results showed that the pooled risk ratio for body movement associated with Remimazolam compared to Propofol is 1.43 (95% CI: [0.96; 2.14]), suggesting a potential increased risk associated with Remimazolam, although this result did not reach statistical significance (p = 0.08), Fig.S19. The pooled analysis showed significant heterogeneity with (I² = 80.9%, p < 0.0001). A leave-one-out sensitivity analysis was conducted; the sensitivity analysis maintained the overall high degree of heterogeneity with (I² ranging from 70.7% to 82.7%) throughout different scenarios and did not significantly alter the overall risk ratio, Fig.S20.

Additional Analyses

A subgroup analyses for the primary outcomes based on the type of endoscopic procedure used (gastroscopy, colonoscopy, or ERCP) were performed. The subgroup analysis based on these types of endoscopic procedures did not include EUS or polypectomy, as only one study utilized these types of endoscopic procedures.

Induction Time

Remimazolam showed a statistically significant increase in induction time compared to propofol during gastroscopy (p = 0.0009, I² = 94.8%) and colonoscopy (p = 0.0018, I² = 93.3%). In contrast, no significant differences were noted in ERCP (p = 0.0923, I² = 90.5%), Fig.S21.

Total Sedation Time

Remimazolam showed a statistically significant increase in total sedation time compared to propofol in ERCP (p = 0.0010, I² = 0%), Fig.S22.

Procedure Time

The subgroup analysis indicated no statistically significant difference in procedure time between the two sedatives, with (p = 0.8067, I² = 46.8%) in gastroscopy, (p = 0.8080, I² = 63.3%) in colonoscopy, and (p = 0.2748, I² = 66.9%) in ERCP, Fig.S23.

Sedation Success

Remimazolam showed a statistically significant increase in sedation success compared to propofol during colonoscopy (p = 0.0018, I² = 44.3%). In contrast, no significant differences were noted in gastroscopy (p = 0.0858, I² = 59.1%), Fig.S24.

Awake Time

The subgroup analysis revealed a statistically significant difference in awake time between the two sedatives, favoring propofol in ERCP (p = 0.0212, I² = 77.8%). In contrast, no significant differences were observed in gastroscopy (p = 0.9717, I² = 98%) and colonoscopy (p = 0.7364, I² = 93.8%), Fig.S25.

Publication Bias Assessment

Detailed publication bias assessment using funnel plots is available in Fig.S26.

Meta-Regression

We performed a meta-regression analysis by including age as a covariate. There was a significant relationship between age and the outcome of induction time (regression coefficient [RC] −0.0161; P = 0.02391). For this outcome, older age was associated with a tendency toward no significant difference between groups, indicating a reduction in the effect of Remimazolam compared to propofol. There was no significant relationship between age and the relative efficacy of Remimazolam compared with propofol for other outcomes, such as time to recovery ([RC] −0.0271; P = 0.6297), time to discharge ([RC] −0.0856; P = 0.2148), awake time ([RC] 0.0188; P = 0.7367), procedure time ([RC] −0.0033; P = 0.7821), and total sedation time ([RC] −0.0767; P = 0.5667). See Fig.S27.

Discussion

Over the years, multiple gastrointestinal endoscopic procedures were developed, including esophagogastroduodenoscopy (EGD), endoscopic retrograde cholangiopancreatography (ERCP), endoscopic ultrasound (EUS), colonoscopy, wireless capsule endoscopy (WCE), double balloon enteroscopy (DBE), and flexible sigmoidoscopy. Each of the previously mentioned procedures has its own set of indications. In a recent European Society of Gastrointestinal Endoscopy (ESGE) position statement, GI endoscopy is proposed to have a great screening potential for gastric tract cancers, including colorectal carcinoma and gastric cancer [54, 55].

Multiple pharmacological sedatives and analgesics are used for endoscopic sedation, most commonly propofol, opioids, and benzodiazepines, such as midazolam [9]. Propofol is an intravenous hypnotic drug, initially approved for general anesthesia. Multiple reports suggested that propofol is superior to traditional methods of analgesia (as midazolam and opioids), with shorter recovery times, much quicker onset, and lower adverse events, including nausea and vomiting [56]. This sedative is relatively contraindicated in patients with chronic kidney disease and liver cirrhosis, as it is metabolized and excreted by the liver and kidneys, respectively. The main challenge is the associated cardiovascular side effects, resulting in hypotension. Other main reported adverse events are respiratory depression and pain on injection [57]. Remimazolam, a benzodiazepine agent activating γ-aminobutyric acid type A (GABA-A) receptor, was first approved by the Food and Drug Administration (FDA) in 2020. It is considered a short-acting sedative with reported positive results in both general and procedural anesthesia [58]. Its quick onset, predictable duration, stable hemodynamics, and independence from organ function make it a viable and safe option for procedural anesthesia, especially endoscopic procedures. The exact safety profile of Remimazolam is not currently available, as it is a newly approved drug; however, hypoxia, bradycardia, and hypotension were reported in some cases [59]. A recent meta-analysis conducted on older patients reported no significant difference between Remimazolam and other sedatives regarding endoscopic success, nausea, vomiting, or recovery time. Nonetheless, lower rates of hypoxia, hypotension, bradycardia, respiratory depression, and pain on injection were reported in the Remimazolam group [60].

When comparing propofol and Remimazolam in elderly patients undergoing gastrointestinal endoscopy procedures, higher efficacy was reported in propofol sedated groups, with quicker onset of action and greater sedation success; on the other hand, Remimazolam was associated with a better safety profile with lower rates of bradycardia, hypoxemia, and injection pain [61]. When both drugs are combined individually with a short-acting opioid, Remimazolam was superior to propofol regarding safety in GI endoscopic procedures, with no statistically significant difference in efficacy outcomes [14]. Another meta-analysis comparing the safety of monotherapy sedation by Remimazolam or propofol reported a lower incidence of respiratory depression, hypoxemia, bradycardia, injection pain, and dizziness in the Remimazolam-sedated group [13].

Our meta-analysis reported the results from thirty-seven trials comparing propofol and Remimazolam in patients undergoing various GI endoscopy procedures. Our analysis showed no statistically significant difference between the two drugs in sedation success, procedure times, time to discharge, time to recovery, awake time, and endoscopic satisfaction. However, the propofol and Remimazolam groups reported shorter total sedation time, respectively. Treated patients stated higher satisfaction after Remimazolam administration. A better safety profile was observed in Remimazolam groups with lower rates of hypotension, bradycardia, depression, hypoxemia, injection pain, and respiratory depression during sedation, except for post-operative nausea and vomiting rates and body movements, where there was no statistically significant difference between the two sedatives studied.

Jiang's research on how Propofol inhibits respiration suggests that it may bind to β3, which is linked to respiratory depression and unconsciousness [62]. In contrast, remimazolam primarily binds to receptor subtypes associated with the β2 subunit [63], potentially leading to less impact on respiratory function during use. Nonetheless, further research is needed to fully understand the differences among various GABA subtypes.

Endoscopic retrograde cholangiopancreatography (ERCP), a combination of endoscopy and fluoroscopy, is used to diagnose and treat pancreaticobiliary conditions [64]. Propofol monotherapy sedation was reported to be safe and effective in ERCP procedures, associated with quicker recovery and lower costs, depending on patient selection and endoscopists’ skill [65]. Due to Remimazolam’s previously observed lower cost and superior safety profile [66], we conducted a subgroup analysis between the two proposed sedatives in ERCP patients. There was no statistically significant difference between propofol and Remimazolam in induction time, procedure times, endoscopic and patient satisfaction, with better awake time in propofol-sedated groups.

Colonoscopy enables imaging for the large intestine, with a high capability of diagnosing and treating various related diseases, including colorectal cancer, lower GI bleeding, and inflammatory bowel disease (IBD) [67]. Propofol was associated with high efficacy and safety after colonoscopy, with patients being able to drive one hour after the procedure. These results render the procedure more favourable to the general population for screening of colorectal carcinoma, decreasing its associated mortality [68]. Remimazolam was reported to be more sensitive in sedation induction than midazolam, with no statistically significant difference between the two drugs on colonoscopy completion and overall procedural sedation [69]. Our subgroup analysis reported shorter induction time and higher sedation success in the Remimazolam-sedated group. There was no statistically significant difference between the two drugs in procedure time, time to discharge, and awake time.

Gastroscopy (EGD, upper GI endoscopy), visualizing the mucosal surface of the esophagus, stomach, and proximal duodenum, is indicated in achalasia, Barrett’s esophagus, esophageal cancer, gastric cancer, ampullary adenoma, and multiple other pathologies [70]. Propofol had higher patient and physician satisfaction, shorter recovery time, and lower hypoxia rates when compared with midazolam [71]. Due to the previously stated advantages of Remimazolam, we conducted a subgroup analysis reporting a significantly more extended induction and total sedation time in Remimazolam groups compared to propofol, with no statistically significant difference between the two groups in terms of procedure time, time to discharge, sedation success rates, awake time, time to recovery, endoscopic, and patient satisfaction rates.

The strength of our study lies in following the PRISMA guidelines, using reliable risk of bias assessment tools, and accurate analysis. However, we faced multiple limitations: Heterogeneity exceeded 90% in some outcomes, which may be due to the variability in performed procedures, outcome definitions, doses, or the use of co-analgesics. Additionally, all the included clinical trials are based in China, indicating that the results are based on an Asian population, with no indication of their viability in other populations. The definition of the (MOAA/S) and its cutoff point for induction were not consistent across the included studies, which limited our ability to generalize the results, along with the inability to conduct a subgroup analysis for other forms of endoscopy, including EUS, due to the low number of available studies. Finally, we could not include non-English studies as we can not trust online translation websites in evidence synthesis.

Our results support the routine use of Remimazolam for sedation in multiple endoscopic procedures due to its proven safety profile. We recommend future trials with a prespecified MOAA/S cutoff score and larger, more diverse populations to update the guidelines accurately.

Conclusion

Sedation remains the most challenging part of any procedure, either for the physician or the patients. Remimazolam has demonstrated its safety, with a lower incidence of hypertension, bradycardia, respiratory depression, and hypoxemia compared to propofol. However, our results showed no statistically significant difference between the two groups in the efficacy-related outcomes. Further trials should investigate these two options on different populations, with standardization of the confounders such as co-analgesics, dosing, or the performed procedure.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

ALY, ZWE, and NH: conceptualization and methodology. ALY, ZWE, NH, AMR, ARS, KAK, MNH: investigation and data curation. AAA and AFY: formal analysis. BME, BH, AAM, YMH: Writing—Original Draft. AAA, YN, DAE: Supervision. ALY, AAA: Project administration. AAA, AG, and RM: Writing—Review & Editing. All authors read and approved the final content.

Funding

Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). No funding was received.

Data Availability

No datasets were generated or analysed during the current study.

Declaration

Competing interests

The authors declare no competing interests.