Comparison of remimazolam and midazolam for preventing intraoperative nausea and vomiting during cesarean section under spinal anesthesia: a randomized controlled trial
Article information
Abstract
Background
Preventing intraoperative nausea and vomiting (IONV) is crucial for maternal safety during cesarean section under spinal anesthesia. While midazolam is known to prevent IONV, we hypothesized that remimazolam would be superior due to its minimal hemodynamic effects. We compared the effects of the two drugs on IONV.
Methods
Parturients scheduled for cesarean section were randomly assigned to receive either midazolam or remimazolam. They received midazolam 2 mg or remimazolam 5 mg, with additional doses administered upon request. The primary outcome measure was the incidence of newly developed IONV during sedation. Other outcomes included overall IONV, rescue antiemetic use, shivering, hemodynamic variables, sedation scale scores, and satisfaction scores.
Results
Data from 80 participants were analyzed. Deeper sedation was induced in the remimazolam group (PGroup × Time < 0.001) despite comparable hemodynamic trends between the groups. The incidence of overall IONV was comparable between the two groups (27.5% in the midazolam group vs. 17.5% in the remimazolam group, absolute risk reduction [ARR]: 0.100, 95% CI [−0.082, 0.282], P = 0.284); however, newly developed IONV during sedation was significantly reduced in the remimazolam group (20.0% vs. 5.0%, ARR: 0.150, 95% CI [0.009, 0.291], P = 0.043). The need for rescue antiemetics was also lower in the remimazolam group (15.0% vs. 2.5%, ARR: 0.125, 95% CI [0.004, 0.246], P = 0.048).
Conclusion
Remimazolam significantly reduced the incidence and severity of newly developed IONV compared with midazolam, with minimal impact on hemodynamics, making it a useful sedative option for cesarean section.
Introduction
Spinal anesthesia is the preferred anesthetic method for cesarean section [1] as it minimizes the effects of anesthetic drugs on newborns while allowing parturients to participate in the delivery process without causing pain. However, evidence indicates a high incidence of intraoperative nausea and vomiting (IONV), reaching up to 80% during cesarean section under spinal anesthesia [2]. Preventing IONV is important not only to enhance maternal satisfaction but also to reduce the risk of gastric content aspiration [3].
The literature suggests that subhypnotic doses of midazolam effectively prevent IONV during spinal anesthesia [4,5] by reducing dopaminergic influx to the chemoreceptor trigger zone in the central nervous system [6,7]. The sedative and anxiolytic properties of midazolam offer additional advantages to parturients after mother-newborn bonding. However, the potential sedative effects of the active metabolites may lead to slow recovery or re-sedation in some patients [8].
Remimazolam, a novel ester-based benzodiazepine, is known for its rapid onset and short duration of action as it is rapidly metabolized into inactive metabolites by nonspecific tissue esterases that significantly reduces the risk of re-sedation [9,10]. In addition, a previous study suggested that remimazolam is associated with a lower incidence of hypotension than midazolam [11]. Given that IONV may be related to vagal hyperactivity and consequent hypotension induced by spinal anesthesia [3], we hypothesized that remimazolam would be superior to midazolam in preventing IONV. Therefore, this study aimed to compare the effects of remimazolam and midazolam on the IONV during cesarean section under spinal anesthesia.
Materials and Methods
Study population
This prospective randomized controlled study included pregnant women scheduled for elective cesarean section between May 2023 and January 2024, following Consolidated Standards of Reporting Trials (CONSORT) guidelines. The study was conducted in accordance with the Ethical Principles for Medical Research Involving Human Subjects outlined in the Declaration of Helsinki of 1975 (revised in 2013) and was approved by the Institutional Review Board (IRB) of Severance Hospital, Yonsei University Health System (Seoul, Republic of Korea; IRB No. 4-2022-1528) in January 2023 and registered at Clinicaltrials.gov in February 2023 (NCT05736341). Informed consent was obtained from all participants involved in the study in accordance with the relevant guidelines and regulations. The inclusion criteria were as follows: (1) American Society of Anesthesiologists (ASA) physical status classification II-III; (2) gestational age of 36 weeks or more; (3) elective cesarean section under spinal anesthesia; and (4) voluntary request for sedation after mother-newborn bonding. The exclusion criteria were as follows: (1) morbid obesity with a body mass index of 40 kg/m2 or higher; (2) multiple gestations; (3) incomplete spinal anesthesia; and (4) massive intraoperative bleeding exceeding 1,500 ml.
Study design and outcome measurements
All enrolled parturients were allocated to the study groups according to a randomized sequence provided in Excel® (Microsoft). Parturients were randomly assigned to either the remimazolam or midazolam group and blinded to the group allocation.
Routine monitoring, including three-lead electrocardiography, pulse oximetry, and non-invasive blood pressure measurements, was initiated upon arrival in the operating room. Tympanic temperature was assessed at 15-minute intervals. Oxygen was administered via a nasal cannula at a rate of 2 L/min, and ondansetron 4 mg was injected intravenously. Subsequently, the parturients were positioned in the left lateral decubitus position, and 12 mg of bupivacaine was injected into the subarachnoid space of the lower lumbar region to achieve the T4 level of sensory blockade.
After delivery, oxytocin infusion was initiated, and additional uterotonic agents such as carbetocin or methylergonovine were administered at the surgeon’s request. After mother-newborn bonding, parturients received an intravenous injection of 5 mg remimazolam (Byfavo, Hana Pharm Co., Ltd.) over 1 min [12] or 2 mg of midazolam, depending on the assigned group. After the sedative injection, the Modified Observer’s Assessment of Alertness/Sedation (MOAA/S) scale was evaluated at 5-minute intervals [13]. Parturients were also asked if they required additional sedatives every 5 min and additional doses of remimazolam (2.5 mg) [12] or midazolam (1 mg) were administered upon request. Heart rate, oxygen saturation, and blood pressure were assessed at 3-minute intervals from the induction of spinal anesthesia to delivery and at 5-minute intervals for the remaining part of the surgery. In the post-anesthesia care unit (PACU), these variables were assessed at 10-minute intervals. The duration of stay in the PACU, maternal satisfaction score (0 = unsatisfied; 1 = neutral; 2 = satisfied), and duration of maternal hospitalization after surgery were recorded. Maternal symptoms, such as nausea, retching, vomiting, and shivering, were recorded throughout the study period. When retching or vomiting occurred, 10 mg of metoclopramide diluted in 50 ml of saline was administered.
Hypotension was defined as a mean blood pressure < 60 mmHg or a decrease of > 20% from baseline. Phenylephrine infusion at 50 μg/min was initiated post-bupivacaine injection, with dose adjustments to accordingly maintain normotension. Additional vasopressors, including ephedrine and norepinephrine, were considered at the discretion of the attending anesthesiologist. Atropine injection was considered in parturients with bradycardia (< 45 beats/min). When desaturation defined as oxygen saturation < 95% occurred, airway interventions such as jaw thrust or positive pressure ventilation were considered. A Bair HuggerTM temperature management system (3M) was used for warming after induction of spinal anesthesia. If the parturient requested further warming, an additional warmed blanket was applied, and a pethidine injection was considered in cases of shivering.
The primary outcome was the incidence of newly developed IONV during sedation. Other outcomes included overall IONV, additional sedative use, retching or vomiting requiring rescue antiemetics, MOAA/S score, mean blood pressure, heart rate, oxygen saturation, incidence of hypotension, hypoxia, shivering, newly developed shivering during sedation, total dose of phenylephrine, and use of other cardiovascular drugs such as atropine, ephedrine, or norepinephrine. Delivery time defined as the time from the induction of spinal anesthesia to delivery, surgical duration, fluid intake, blood transfusion, urine output, and blood loss were assessed.
Statistical analysis
The sample size for this trial was calculated using the G*Power software (version 3.1.9.2). A previous study reported a 58.6% incidence of nausea and vomiting after midazolam injection during cesarean delivery [14]. In contrast, the incidence following remimazolam administration in this clinical setting has not yet been adequately studied. However, since the incidence of acute postoperative nausea and vomiting with remimazolam in gynecological surgeries has been reported to be as high as 27% [15], we assumed that this rate might also be relevant to cesarean sections under spinal anesthesia. Consequently, each group required 38 subjects to achieve a two-tailed significance level of 5% (α = 0.05) and 80% power (β = 0.05). Accounting for a 5% dropout rate, a total of 80 participants (40 per group) were required for this study.
All statistical analyses were performed using R software version 4.3.1 (R Foundation for Statistical Computing). The unpaired Student’s t-test was used to analyze continuous variables, and the Mann–Whitney U test was used to analyze variables that did not meet normality. Chi-square or Fisher’s exact tests were used to compare categorical variables between the groups. Repeated variables were analyzed using repeated-measures ANOVA. Post-hoc analysis with Bonferroni correction for within-group comparisons was performed for multiple comparisons. The effect size was calculated as follows: absolute risk reduction (ARR) for categorical variables, Cohen’s d for variables that met normality assumptions, and Cliff’s delta for variables that did not meet normality assumptions. Statistical significance was set at P < 0.05.
Results
Eighty pregnant women who underwent a cesarean section between May 2023 and January 2024 were enrolled in this study. All the participants successfully completed the study protocol without meeting the exclusion criteria, so data from 80 participants were analyzed (Fig. 1).
Parturient enrollment.
Baseline parturient characteristics and anesthetic data prior to sedation are shown in Tables 1 and 2, respectively. Age, height, weight, body mass index, gestational age, ASA classification, and history of hypertension and diabetes mellitus were comparable between the groups. After induction of spinal anesthesia, shivering occurred more frequently in the remimazolam group. Other variables, including the incidence of IONV or hypotension, total dose of phenylephrine, atropine use, delivery time, and administration of carbetocin or methylergonovine, were similar between the groups.
Parturient Characteristics
Anesthetic Data Prior to Sedation
The data assessed after the sedative injections are presented in Table 3 and Fig. 2. Additional sedative requirements upon maternal request were similar between the groups; however, deeper sedation was induced in the remimazolam group during the first 10 min (MOAA/S score 4.5 ± 0.9 in the midazolam group vs. 3.3 ± 1.5 in the remimazolam group at 5 min after sedative injection, Cohen’s d: 1.001, 95% CI [0.527, 1.471]; 4.6 ± 0.7 vs. 4.1 ± 1.1 at 10 min after sedative injection, Cohen’s d: 0.569, 95% CI [0.113, 1.021], PGroup × Time < 0.001). Heart rate, oxygen saturation, and mean blood pressure trends were similar between the groups. The incidence of overall IONV was comparable between the two groups (27.5% vs. 17.5%, ARR: 0.100, 95% CI [−0.082, 0.282], P = 0.284); however, the incidence of newly developed IONV during sedation was significantly lower in the remimazolam group (20.0% vs. 5.0%, ARR: 0.150, 95% CI [0.009, 0.291], P = 0.043). The need for rescue antiemetics to relieve retching or vomiting was also lower in the remimazolam group (15.0% vs. 2.5%, ARR: 0.125, 95% CI [0.004, 0.246], P = 0.048). Three desaturation events were recorded after remimazolam injection: desaturation of one parturient spontaneously resolved before any intervention, and desaturation of the other two parturients immediately recovered with the jaw-thrust maneuver without further complications. Positive-pressure ventilation was not required in either group. Overall shivering incidence was higher in the remimazolam group (17.5% vs. 50.0%, ARR: −0.325, 95% CI [−0.520, −0.130], P = 0.002); however, the incidence of newly developed shivering after delivery was similar between the groups (12.5% vs. 25.0%, ARR: −0.125, 95% CI [−0.294, 0.044], P = 0.152). Hypotension was observed in one parturient who received remimazolam, but the total doses of phenylephrine and atropine were similar between the groups. No additional vasopressors (norepinephrine or ephedrine) were required to maintain normotension. Operative time, intraoperative fluid injection, urine output, blood loss, recovery time in the PACU, hospital stay, and parturient satisfaction scores were comparable between the groups.
Data Assessed after Sedative Injection
Hemodynamic trend. (A) MOAA/S, (B) heart rate, (C) SpO2, and (D) mean blood pressure. MOAA/S: The Modified Observer’s Assessment of Alertness/Sedation, OP: operation, PACU: post-anesthesia care unit, SpO2: peripheral oxygen saturation. *P < 0.001 vs. midazolam group.
Discussion
Remimazolam is gaining popularity as a useful option for procedural sedation [16], but evidence regarding its use in cesarean sections is scarce. To our knowledge, this is the first study comparing the effects of remimazolam and midazolam during a cesarean section under spinal anesthesia. We demonstrated that remimazolam is superior to midazolam in preventing newly developed IONV during sedation.
Alleviating IONV is a top concern for parturients in relation to enhanced recovery after cesarean section [17]. Our results indicated that in the remimazolam group, only two of seven parturients had newly developed IONV, whereas eight of 11 parturients experienced newly developed IONV after midazolam injection. In addition to the lower incidence of newly developed IONV, the severity of IONV requiring rescue antiemetics was significantly lower in the remimazolam group. Although evidence indicates a favorable effect of a subhypnotic dose of midazolam in preventing IONV [5,18], it appears likely that the antiemetic effect of midazolam was inadequate to mitigate the effects of emetogenic factors in our study. In fact, the presence of various emetogenic factors around the time of sedative administration may have interfered with the interpretation of the results. Specifically, parturients may experience severe abdominal discomfort during the fetal extraction maneuver, even under an adequate level of sensory blockade. In addition, various uterotonic agents that may aggravate IONV are frequently used [19]. Oxytocin was routinely infused in all parturients, and supplemental uterotonics, including carbetocin and methylergonovine, were frequently administered at the surgeon’s discretion. Our results suggest that remimazolam effectively suppressed the occurrence and severity of newly developed IONV during sedation.
An important factor that may influence the incidence or severity of IONV with spinal anesthesia is sympathectomy-related hypotension [3] that occurs in up to 75% of women receiving spinal anesthesia for cesarean delivery [20]. Intraoperative hypotension results in brain stem ischemia and consequent activation of the vomiting center [3]. As evidence indicates that remimazolam induces less hypotension than midazolam during bronchoscopy [11], we assumed that remimazolam would exhibit similar effects during cesarean section, thereby reducing IONV. However, we found no differences in the incidence of hypotensive events or hemodynamic trends between the two groups, most likely because of the precise adjustment of phenylephrine infusion to maintain normotension. Additionally, the dose of phenylephrine infused to support blood pressure did not differ between the groups, indicating that the severity of hypotension was similar between the groups. However, it should be considered that the cardiovascular effects evoked by benzodiazepines are proportional to sedation depth [21] and that remimazolam induced more profound sedation than midazolam in our study. Hence, our results imply that remimazolam has a similar impact on hemodynamics compared to a lesser potent dose of midazolam, which is consistent with the findings of a previous study [11].
Nonetheless, the differences in sedation levels achieved between the two groups might be a possible reason for the inferior outcomes in the midazolam group regarding IONV. We initiated sedation with the minimum effective dose of each drug [12,22] and did not target a specific level of sedation depth afterward; instead, we assessed the participants’ satisfaction with their sedation status every 5 min and administered additional sedatives only when they requested them. Consequently, the sedation satisfaction score surveyed in the PACU was comparable between the groups despite the differences in sedation depths between the two groups. Our results also imply that a subhypnotic dose of midazolam is as effective as a larger dose of remimazolam in terms of anxiolysis; however, further research is needed to confirm this hypothesis.
Spinal anesthesia impairs thermoregulatory vasoconstriction below the level of the blockade and results in shivering [23,24] that is not only distressing but also leads to potentially detrimental effects such as increases in physiological stress, oxygen consumption, or even myocardial infarction in morbid patients [25]. Although benzodiazepine reduces shivering via the GABA pathway [26–28], the overall incidence of shivering increased after sedative injection in both groups (10.0% and 17.5% in the midazolam group and 30.0% and 50.0% in the remimazolam group, respectively). Although the overall incidence of shivering was higher in the remimazolam group, there was no significant difference in the incidence of newly developed shivering during sedation. Additionally, it should be considered that the incidence of shivering prior to sedation was already higher in the remimazolam group compared to the midazolam group. We presume that the surgical environment of the cesarean section has offset the favorable effects of benzodiazepines on intraoperative shivering. During surgical preparation, most of the abdomen is inevitably exposed to cooled air in the operating room, although other regions are covered with warmed blankets. In addition, cesarean section often involves massive irrigation and significant volume loss, necessitating rapid infusion of crystalloids that are generally not warmed due to high cost. Although recent studies have mainly focused on pharmacological interventions to prevent shivering in cesarean section [29,30], our results imply that non-pharmacological interventions, such as prewarming, warmed intravenous crystalloid solutions, proactive use of warming devices, and maintaining a meticulous operating room temperature, should not be neglected [31].
Although remimazolam is well known for its hemodynamic stability, its safety profile during breastfeeding remains unknown. In contrast, the safety of midazolam during breastfeeding has been well documented [32,33]. Given that midazolam is metabolized into active metabolites whereas remimazolam is metabolized into inactive metabolites, we presume that the risk of adverse effects in breastfed infants would also be lower with remimazolam than with midazolam. However, further studies are required to confirm the safety of remimazolam during breastfeeding.
The limitations of this study were as follows. First, due to the difference in administration methods between remimazolam and midazolam, blinding the participating anesthesiologist was not feasible. Second, differences in sedative potency between benzodiazepines may have influenced our results. Discrepancies in sedation levels arose because sedatives were administered only at the maternal request. A more robust comparison between the two drugs would have been possible if a specific MOAA/S target had been set in advance, but clinically there was a reduced need to sedate the parturient further while she was already satisfied with the current level of sedation. Finally, if sedation had started immediately after the induction of spinal anesthesia, it would have allowed for a more detailed assessment of the IONV-preventive effect of the sedatives. However, to avoid disrupting mother-newborn bonding, the administration of sedatives had to be delayed until after the mother and newborn were bonded, that, in part, complicated the assessment of sedative effects on IONV.
In conclusion, compared with midazolam, remimazolam significantly reduced the incidence and severity of newly developed IONV during sedation with minimal impact on hemodynamics, making it a useful sedative option for cesarean section.
Notes
Funding
This research was supported by Hana Pharm Co., Ltd., Seoul, Republic of Korea.
Conflicts of Interest
No potential conflict of interest relevant to this article was reported.
Data Availability
The data analyzed during the current study are available from the corresponding author on reasonable request.
Author Contributions
Kyuho Lee (Investigation; Validation; Writing – original draft)
Seung Ho Choi (Methodology; Supervision)
Sangil Kim (Data curation; Methodology)
Hae Dong Kim (Data curation; Investigation)
Hyejin Oh (Software; Visualization)
Seung Hyun Kim (Conceptualization; Funding acquisition; Supervision; Writing – review & editing)