INTRODUCTION
Remimazolam is a new drug that belongs to the family of ultrashort-acting benzodiazepines [1]. It is used in many countries as a sedative in non-operative room anesthesia procedures as well as for intravenous induction and maintenance of general anesthesia in operating theaters. Like midazolam, it acts on γ-aminobutyric acid (GABA) A receptors to exert its action. Its effect can be reversed using the drug flumazenil [2]. Remimazolam has a rapid onset of action and recovery and shows very little accumulation after continuous intravenous infusion, making it suitable for maintenance of anesthesia. Systematic reviews and meta-analyses have been published comparing remimazolam to propofol or midazolam for procedural sedation [3,4]. Total intravenous anesthesia (TIVA) with remimazolam has been shown to decrease the incidence of respiratory depression, hypoxemia, bradycardia, nausea, and vomiting. Postoperative nausea and vomiting (PONV) is a major adverse reaction of general anesthesia and still affects 20-30% of patients, even in cases employing enhanced recovery of surgery protocols [5]. With the advent of newer drugs and anesthesia machines, the safety of anesthesia techniques has improved substantially, with an emphasis on the quality of postoperative recovery. Randomized controlled trials (RCTs) have been conducted to compare the effects of remimazolam with those of inhalational anesthetics on postoperative functional recovery. Given the distinct pharmacological profile and growing clinical use of remimazolam, a focused evaluation of its benefits in the postoperative period in comparison with inhalational anesthesia is warranted. Therefore, we conducted a systematic review and meta-analysis, including trial sequential analysis (TSA), to determine whether the newer drug remimazolam reduces the incidence of PONV in comparison with inhalational anesthesia in patients undergoing general anesthesia.
MATERIALS AND METHODS
The study was conducted in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [6]. The protocol for this systematic review was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42024544157.
Data sources
We searched MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), Web of Science, Google Scholar, and Scopus from their inception until August 2024. A comprehensive search strategy was used on the basis of the keywords “Remimazolam” “Inhalational anesthetic” “Sevoflurane” “Isoflurane” “Desflurane” “Postoperative” “Recovery” separated by Boolean operators AND and OR. Two authors (KM and RS) independently screened the titles and abstracts of relevant articles that fulfilled the eligibility criteria from the databases. All the references of the selected articles were also scanned to identify other relevant articles. Disagreements between the two authors were resolved by consensus with a third author (AK). The search equation is detailed in Supplementary File 1.
Study selection
Our inclusion criteria were based on the following PICOS framework:
Population (P): Patients in the age group 18-70 years undergoing surgery under general anesthesia.
Intervention (I): Remimazolam use for induction and maintenance of anesthesia.
Comparison (C): Inhalational agent use for induction and maintenance of anesthesia.
Outcome (O): The primary outcome was PONV assessed at 24 h postoperatively. Secondary outcomes were the incidence of use of rescue antiemetics, pain scores, and incidence of use of rescue analgesics in the postoperative period until 24 h after surgery. The time to extubation, time to reach Aldrete score ≥ 9, post-anesthesia care unit (PACU) length of stay, and assessment of adverse effects such as intraoperative hypotension, postoperative pulmonary complications, postoperative urinary retention, incidence of acute kidney injury (AKI), postoperative delirium, and Quality of Recovery-15 (QoR-15) scores were also analyzed.
Study design (S): RCTs
Exclusion criteria: We excluded articles reporting abstracts alone, protocols, non-randomized studies, non-human studies, quasi-RCTs, case reports or case series, and articles published in languages other than English. The selection process was recorded in a PRISMA flow diagram [7].
Data screening and extraction
Two authors (AK and KP) independently extracted data for each study that included the authors, year, country, sample size, study design (time, dose, and method of administration of general anesthesia), inclusion and exclusion criteria, type of surgery, inhalational anesthetic used, intraoperative use of remifentanil, and postoperative outcomes reported by the studies. Disagreements were resolved by a third author (RS).
Risk-of-bias analysis
Two authors (AK and KP) conducted a risk-of-bias assessment for the studies included in the meta-analysis. Any disagreements were resolved by a third author (RS) using the Revised Cochrane Risk-of-Bias Tool for Randomized Trials (RoB 2) [8]. This tool evaluates five key domains: randomization process, allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), deviations from the intended intervention, missing outcome data (attrition bias), measurement of outcomes, selection of reported results (reporting bias), and other potential biases.
Quality of evidence
The level of certainty of the evidence was assessed with the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system, which considered the study limitations, consistency of effect, imprecision, indirectness and publication bias [9]. A GRADE evidence table was generated using the GRADE software available online, and the outcomes were rated as high, moderate, low, or deficient quality. The quality of evidence was downgraded by one or two levels when serious or very serious deficiencies were found.
Publication bias was assessed using a funnel plot. Egger’s regression test was performed to quantitatively examine the presence of small-study effects. Sensitivity analysis was also performed using the leave-one-out method, in which studies were sequentially excluded to evaluate the robustness of the pooled estimates against potential publication bias.
Statistical analysis
For dichotomous variables, we recorded the sample size and number of events. Mean and standard deviation (SD) values were noted for continuous variables. Continuous data reported as medians and interquartile ranges were converted to mean and SD values using the formula provided in the Cochrane Handbook [10]. The analysis of dichotomous variables was reported with relative risk (RR) and a 95% confidence interval (CI). For continuous variables, we used the standardized mean difference (SMD). Statistical significance was set at P < 0.05. Review manager (RevMan 5.4.1, Cochrane) was used for the analysis. Heterogeneity among trials was quantified with Higgins’ and Thompsons’ I2. Regardless of the I2 value, we used a random-effects model. An I2 value exceeding 50% indicated substantial heterogeneity. To assess the robustness of our findings, we also performed TSA using the TSA software (0.9.5.10 Beta, Copenhagen Trial Unit) [11]. For this analysis, we set a two-sided type I error of 5% and power of 80% based on the RR reduction derived from low-risk-of-bias trials. TSA-adjusted boundaries were constructed using a random-effects model, and the cumulative Z-curve was assessed against these thresholds. If the cumulative Z-curve crossed the monitoring boundary or reached the required information size (RIS), the result was considered conclusive.
RESULTS
In total, 230 articles were identified using the search strategy described above. Of these, 166 articles were screened for eligibility. Eighteen full-text articles were reviewed. Three articles were removed since they reported studies involving children [12-14], and one article did not specify whether an inhalational anesthetic was used in the remimazolam group [15]. Studies in which remimazolam was used in both the intervention and control groups were excluded, since they would not allow for meaningful comparisons with inhalational agents and could introduce confounding factors and increase the heterogeneity of our outcome analysis [16,17]. A total of 12 articles [18-29] with 853 patients were included in our final systematic review and meta-analysis, as shown in the PRISMA flowchart (Fig. 1). The baseline characteristics of the included articles along with the type of surgery and induction and maintenance doses of remimazolam are presented in Table 1. One study included two doses of remifentanil with inhalational anesthetic, and we have included the findings for the higher dose of remifentanil with inhalational anesthetic, which was the same dose used in remimazolam group as well [18]. All of these studies were conducted in the Korea, except for the study by Hari et al. [19] performed in Japan. All the studies were published between 2022 and 2024.
Primary outcome
Our primary outcome was PONV assessed 24 h postoperatively. A total of 10 studies reported this outcome [19-28]. The incidence of PONV was significantly lower in the remimazolam group, with pooled results showing an RR of 0.51 and a 95% CI of 0.27 of 0.96. The reported heterogeneity was 43% (moderate heterogeneity) (P = 0.04; Fig. 2). A subgroup analysis was performed on the basis of the type of inhalational anesthetic used (desflurane and sevoflurane) and the type of surgery (laparoscopy or other mixed surgeries, such as orthopedic, nasal, or cardiac surgery) (Supplementary Fig. 1). The results of this analysis were similar for both groups.
1. Secondary outcomes
1) Need for rescue antiemetics
In the six studies that reported the need for rescue antiemetics in the first 24 h postoperatively, the remimazolam group required significantly lower rescue doses (RR: 0.30; 95% CI, 0.10 to 0.89; I2 = 0%; P = 0.03; Supplementary Fig. 2).
2) Pain score
Five studies reported pain using the Numerical Rating Scale (NRS), while the study by Lee et al. [18] reported pain using the Visual Analog Score from 0 to 100. To eliminate the heterogeneity of the measurement units, we used the SMD. Pooled analysis showed that the pain scores did not differ significantly between the two groups (SMD: -0.17; 95% CI, -0.38 to 0.04; I2 = 0%; P = 0.11; Supplementary Fig. 3).
3) Need for rescue analgesic
The need for rescue analgesic during the first 24 h did not differ significantly between the two groups (6 studies; RR: 0.95; 95% CI, 0.64 to 1.42; I2 = 58%; P = 0.82; Supplementary Fig. 4)
4) Postoperative recovery profile
These findings are presented in Fig. 3.
(1) Time to extubation: The time to extubation after end of surgery, as reported by seven studies, was statistically similar in both groups (SMD: 0.39; 95% CI, -0.40 to 1.18; I2 = 95%; P = 0.33)
(2) Time to reach Aldrete score ≥ 9: The time to reach an Aldrete score ≥ 9 as reported by three studies did not differ significantly between the two groups (SMD: -0.26; 95% CI, -0.59 to 0.08; I2 = 32%; P = 0.13)
(3) PACU length of stay: The length of stay in the PACU was statistically similar in the two groups (SMD: -0.29; 95% CI, -0.87 to 0.30; I2 = 93%; P = 0.34).
5) Adverse effects
The following data are shown in Supplementary Fig. 5.
(1) Intraoperative hypotension: Nine studies reported intraoperative hypotension. The pooled analysis results showed a significantly lower incidence of intraoperative hypotension in the remimazolam group (RR: 0.58; 95% CI, 0.37 to 0.89; I2 = 60%; P = 0.01)
(2) Postoperative pulmonary complications: Postoperative pulmonary complications did not differ significantly between the two groups (4 studies; RR: 1.17; 95% CI, 0.45 to 3.03; I2 = 0%; P = 0.75)
(3) Postoperative urinary retention: The incidence of postoperative urinary retention was similar in both groups (2 studies; RR: 0.93; 95% CI, 0.69 to 1.25; I2 = 0%; P = 0.63)
(4) AKI: The incidence of AKI reported in three studies was similar in both groups (RR: 1.4; 95% CI, 0.44 to 4.45; I2 = 0%; P = 0.56)
(5) Postoperative delirium: Although the incidence of postoperative delirium was reported in three studies, two studies did not report any incidence of delirium in any group.
6) QoR-15 scores
Three studies reported the QoR-15 scores at 24 h postoperatively. One study reported QoR-40 scores [19]. The pooled analysis showed no significant difference between the two groups (SMD: -0.11; 95% CI, -0.74 to 0.52; I2 = 71%; P = 0.74) (Supplementary Fig. 6).
Publication bias was assessed for the primary outcome, that is, PONV, and was depicted by a funnel plot that showed slight asymmetry to indicate small study effects (Supplementary Fig. 7). Egger’s regression test was performed to quantify the asymmetry, and it did not reveal any significant asymmetry (P = 0.46). Sensitivity analysis confirmed that the overall pooled estimate remained stable even after excluding studies with smaller sample sizes (Supplementary Fig. 8).
2. GRADE assessment
The certainty of evidence for the incidence of PONV was adjudged to be high as per the GRADE methodology because of the low risk of bias in most included studies and consistent findings across the sensitivity and subgroup analyses. Details of the GRADE assessment of other parameters are provided in Table 2.
3. Risk-of-bias assessment
The risk-of-bias was low for all but two studies. The study by Lee et al. [27] was rated as showing “some concerns” regarding domain D1 (bias from the randomization process), since details regarding randomization and allocation concealment were not reported. The study by Cho et al. [26] did not mention whether the outcome assessors were blinded, leading to concerns regarding domain D2 (bias due to deviation from the intended interventions) (Fig. 4).
4. TSA
To assess the adequacy of the sample size for the assessment of PONV, we performed TSA for the primary outcome. At a type I error of 5% and 80% power, the low risk of bias-based RR reduction and incidence in the control arm was 30%, and after model variance-based heterogeneity correction, the required sample size was found to be adequate (Supplementary Fig. 9). The cumulative Z-curve crossed the TSA boundary and reached the conventional boundary, demonstrating that the RIS was reached. Thus, the current evidence was considered to be robust and additional trials were unlikely to change the direction or significance of the effect estimate for the incidence of PONV.
DISCUSSION
Our study is the first systematic review and meta-analysis to include all RCTs comparing remimazolam and inhalational anesthesia for postoperative recovery. The findings from our meta-analysis revealed that the use of remimazolam led to a significant reduction in PONV and a decreased need for rescue antiemetics within the first 24 h post-surgery. However, remimazolam and inhalational anesthetics showed no notable differences in terms of other outcomes, such as pain scores, requirement for rescue analgesics, time to extubation, length of stay in the PACU, incidence of postoperative complications, including pulmonary complications, AKI, urinary retention, and delirium, and QoR-15 scores. Remimazolam use was associated with a significantly lower incidence of intraoperative hypotension than that associated with the use of inhalational anesthesia.
Postoperative recovery involves a broad spectrum of physiological, physical, and psychological factors. Postoperative recovery is generally defined as the process of returning to a preoperative or pre-illness baseline state [30]. Among the key aspects of postoperative recovery, nausea and vomiting particularly influence patient satisfaction, since they have been consistently ranked as some of the most undesirable postoperative outcomes in patient surveys [31]. We included PONV as a primary outcome because it has been found to be a significant predictor of postoperative recovery [32]. Studies have shown that the use of inhalational anesthetics is a risk factor for PONV [5]. In a recent meta-analysis of 317 RCTs, TIVA with propofol was associated with a significant decrease in PONV, emergence delirium, and overall quality of recovery in comparison with volatile anesthetics [33]. Remimazolam, an imidazobenzodiazepine like midazolam, has been shown to reduce the incidence of PONV, probably due to its effect on GABA-related inhibitory pathways [19]. A systematic review and meta-analysis comparing propofol and remimazolam TIVA in patients undergoing gastroscopy found a similar incidence of PONV between the two groups [34]. This could be because propofol has been demonstrated to be effective against PONV, with a shorter duration and lesser amount of propofol infusion required during gastroscopy. The significantly lower incidence of PONV in the remimazolam group in our meta-analysis demonstrates the antiemetic properties of benzodiazepine-like drugs and the excitatory effects of inhalational anesthetics on the vestibular system and vagus nerve [28]. Moreover, subgroup analyses based on the type of inhalational agent and surgery showed consistent findings, confirming the robustness of the observed reduction in PONV with remimazolam use. These results suggest that the benefits of remimazolam in reducing PONV are not limited to a specific inhalational agent or surgical population, but can be observed across varied clinical settings.
The NRS scores and rescue analgesic consumption were similar in both groups. This lack of difference could be partly attributed to the predominance of laparoscopic surgeries within the study, which generally cause lower levels of postoperative pain than open procedures. Similarly, previous studies examining postoperative pain outcomes with other anesthetics, such as sevoflurane, desflurane, and propofol, found no major differences in analgesic requirements [35]. The effects of remimazolam on GABA receptors may play a role in attenuating pain perception post-surgery, contributing to the comparable analgesic outcomes between remimazolam and inhalational anesthetics.
An important factor in choosing anesthetics for surgery is the potential for adverse postoperative outcomes. Inhalational anesthetics are associated with delayed emergence from anesthesia, delayed discharge to the PACU, postoperative pulmonary complications, and increased environmental impact due to carbon emissions [36]. Interestingly, our meta-analysis did not find any significant differences between remimazolam and inhalational anesthetics in terms of the time to extubation, length of stay in the PACU, or adverse effects, including postoperative pulmonary complications, AKI, and delirium. The type and duration of surgeries included in the analysis may have influenced these outcomes. Notably, the number of RCTs reporting these specific postoperative outcomes was limited, highlighting the need for further research to elucidate the impact of remimazolam on these variables.
We also included intraoperative hypotension as a secondary outcome because of its potential influence on postoperative complications such as delayed emergence, PONV, and overall recovery quality. Remimazolam use was associated with significantly lower intraoperative hypotension than inhalation anesthesia. Studies have shown that a bolus followed by continuous infusion of remimazolam is associated with better cardiac output, likely contributing to its stable hemodynamic properties [37]. This attribute makes remimazolam particularly suitable for patients at a higher risk of hemodynamic instability during surgery, since stable intraoperative blood pressure is critical for maintaining tissue perfusion and reducing the risk of perioperative complications.
In our analysis, the QoR-15 score, a validated measure of patient-reported postoperative recovery, was comparable between the remimazolam and inhalational anesthesia groups. Thus, despite the pharmacological differences between the agents, the overall quality of recovery, as perceived by the patients, remained similar. However, since only a limited number of studies have reported this outcome, these findings should be interpreted with caution and warrant further investigation in larger, high-quality trials. A recent systematic review by Park et al. primarily focused on the safety and hemodynamic effects of remimazolam versus volatile agent-based anesthesia [38]. Our analysis offers a more comprehensive evaluation of postoperative recovery, encompassing not only PONV but also pain scores, analgesic use, postoperative delirium, and functional recovery indicators, such as QoR-15 scores and postoperative complications, which were not addressed in detail in the earlier study. Additionally, we strengthened the robustness of our findings by incorporating TSA to ensure that our conclusions were statistically conclusive. While earlier studies primarily focused on safety and hemodynamic parameters, our work provides broader, clinically actionable insights into the overall postoperative experience and recovery quality.
One strength of this study was the use of TSA, which provided sufficient evidence to conclude that remimazolam significantly reduced the incidence of PONV in comparison with inhalational anesthetics. The addition of this statistical method enhanced the rigor and reliability of our findings. Moreover, we conducted clinically relevant subgroup analyses to bolster the robustness of our results. The sensitivity analysis also indicated that the primary outcome was not influenced by the presence of smaller studies. The quality of evidence for most outcomes was moderate-to-high, strengthening the validity of our conclusions. Despite these strengths, our study had some limitations. A key limitation was the limited generalizability of the findings to broader populations, since the included studies were predominantly conducted in the Korea. Additional RCTs in diverse settings, as well as investigations into the role of remimazolam in major surgeries with extensive fluid shifts and longer durations, would be beneficial. Further studies are needed to assess the applicability of remimazolam in pediatric and geriatric populations, because these groups may have different physiological responses to anesthesia. In addition, a few outcomes such as time to extubation and PACU length of stay showed considerable heterogeneity. This variability likely reflects differences in institutional protocols, anesthetic practices, and the criteria for PACU discharge, among other factors. Such heterogeneity is often an unavoidable challenge in evidence synthesis, arising from real-world clinical diversity rather than methodological flaws [39]. Although such heterogeneity does not compromise the overall validity of the study, caution should be exercised when interpreting the outcomes.
In summary, this meta-analysis highlights the notable advantages of remimazolam in the postoperative setting. Its use is associated with a markedly lower incidence of PONV and more stable intraoperative hemodynamics than inhalational anesthetics. TSA confirmed that the current evidence was statistically conclusive and unlikely to change with further studies, reinforcing the strength of our findings. Remimazolam may be particularly beneficial in patients who are at increased risk of PONV or require stable hemodynamic control during surgery. Although additional research is necessary to fully understand the applications of remimazolam in broader and more diverse populations, our results support its potential role as an effective alternative to volatile anesthetics in modern perioperative care.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
