The effective dose of remimazolam in combination with sufentanil for sedation during transvaginal oocyte retrieval: an up-and-down sequential allocation trial

Abstract

Background

The potential adverse effects of propofol on fertilization outcomes have raised concerns among patients undergoing anesthesia for oocyte retrieval. Remimazolam is a new type of ultra-short-acting benzodiazepine. This study aimed to find the effective dose of remimazolam in combination with sufentanil for sedation in oocyte retrieval.

Methods

Patients aged 20 to 45 years with a BMI of 18.5 to 27.9 kg/m² and ASA physical status I or II, undergoing transvaginal oocyte retrieval, were recruited. All patients were administered intravenous sufentanil at a dose of 0.1 µg/kg, followed by intravenous remimazolam after 2 min. The initial remimazolam dose was set at 0.2 mg/kg and adjusted by 0.02 mg/kg according to the modified Dixon sequential method. The primary endpoint was successful sedation, defined as a Modified Observer’s Assessment of Alertness/Sedation (MOAA/S) score ≤ 1 and an Intraoperative Movement Scale (IOMS) score < 2 within 5 min. Adverse events were recorded. The estimation of ED₅₀, ED₉₀ and ED₉₅ with a 95% confidence interval (CI) was conducted by probit regression.

Results

The ED₅₀, ED₉₀ and ED₉₅ of remimazolam for successful sedation in combination with sufentanil induction were 0.205 mg/kg (95% CI 0.190–0.222), 0.232 mg/kg (95% CI 0.218–0.311) and 0.240 mg/kg (95% CI 0.223–0.339), respectively. There were three cases of hypoxemia, with no other adverse events reported, including injection pain, hypotension, bradycardia, dizziness, nausea, vomiting, or delayed emergence.

Conclusions

In summary, when combined with 0.1 µg/kg of sufentanil, the ED₅₀, ED₉₀ and ED₉₅ values of remimazolam for effective sedation during transvaginal oocyte retrieval, were 0.205 mg/kg, 0.232 mg/kg and 0.240 mg/kg, respectively, with a low occurrence of adverse events.

Trial registration

The study protocol was registered at http://www.chictr.org.cn (registration number: ChiCTR2400082576 on 01/04/2024).

Introduction

In vitro fertilization-embryo transfer (IVF-ET) is the most widely used assisted reproductive technology (ART). A transvaginal ultrasound-guided follicle punction for oocyte retrieval is an essential part of IVF-ET. Most patients experience more than moderate pain during the oocyte retrieval [1]. Thus, performing under anesthesia is advised as it enhances patient comfort and offers a favorable operating environment for the surgeon, which is conducive to improving the oocyte retrieval rate and reducing the complications [2]. The combination of propofol and opioids is the drug of choice for oocyte retrieval because it acts quickly and allows for fast recovery [3]. However, studies indicates that propofol accumulates in human follicular fluid in a manner dependent on both dosage and exposure time [4, 5] and affects oocyte maturation in vitro [6]. A retrospective cohort study demonstrated that general anesthesia with propofol during oocyte retrieval is significantly lower fertilization rates when compared to those oocyte retrieved without anesthesia [7]. The potential effects of propofol on oocyte maturation raise concerns among women undergoing anesthesia for oocyte retrieval.

Midazolam, a benzodiazepine, has no adverse effects on fertilization and embryonic development in mice [8]. Remimazolam is an ultra-short acting gamma-aminobutyric acid type A (GABA_A) receptor agonist that adds a methyl propionate side chain to midazolam and acts directly on GABA_A receptors to play a sedative role, which has the characteristics of rapid onset and metabolism, and a short duration of action. Remimazolam has been proven to be effective and safe for procedural sedation and general anesthesia [9, 10]. Benzodiazepines are considered safe for oocyte retrieval [11]. However, the effective dose of remimazolam for oocyte retrieval is not yet well known. Therefore, our study aimed to explore the ED₅₀, ED₉₀ and ED₉₅ of remimazolam when combined with sufentanil for achieving successful sedation in patients undergoing transvaginal oocyte retrieval. Successful sedation was defined as a Modified Observer’s Assessment of Alertness/Sedation (MOAA/S) score ≤ 1 and an Intraoperative Movement Scale (IOMS) score < 2 within 5 min.

Methods

Ethics and trial registration

The study was approved by the Ethics Committee of Zhejiang Provincial People’s Hospital (approval No.: KY2024018) and registered in the Chinese Clinical Trial Registry (registration No.: ChiCTR2400082576). Written informed consent was obtained from all patients.

Inclusion and exclusion criteria

This study was conducted in the Reproductive Center of Zhejiang Provincial People’s Hospital from April 2024 to May 2024. Patients aged 20–45 years old with BMI18.5–27.9 kg/m² and ASA grade I-II, undergoing transvaginal oocyte retrieval were enrolled. Exclusion criteria included the following: (1) allergic to benzodiazepines and opioids or had a history of severe allergy; (2) severe dysfunction of major organs; (3) a history of mental illness or taking psychotropic drugs(including benzodiazepine oral patients); (4) myasthenia gravis; (5) Those who do not willing to participate.

Study protocol

Before the procedure, patients routinely fasted for 8 h and avoided drinking for 2 h, and received no premedication. A peripheral venous catheter was inserted and 0.9% sodium chloride (NaCl) infusion was infused in the preparation room. After entering the operating room, adoption of lithotomy position, vital signs including blood pressure (BP), heart rate (HR) and blood oxygen saturation (SPO₂) were monitored, and received supplemental oxygen via nasal cannula at a flow of 4 L/min throughout the procedure until fully awake.

Patients received intravenous injection of Sufentanil (lot number: 31A101612, Yichang Renfu Pharmaceutical Co., LTD, China) 0.1 µg/kg. 2 min later, remimazolam (lot number: 30T10351, Yichang Renfu Pharmaceutical Co., LTD, China.) was slowly administered intravenously within 1 min. The modified Dixon sequential method was used to determine the dose of remimazolam. The dose of remimazolam administered in the next patient was determined by the reaction of the previous patient, with a dose gradient of 0.02 mg/kg. According to the preliminary results and reference to the relevant literature, the initial dosage of remimazolam was 0.2 mg/kg [12]. The dosage was sequentially decreased by 0.02 mg/kg if the previous patient had a negative reaction. Negative reaction was defined as the modified observer’s assessment of alertness/sedation scale (MOAA/S) was ≤ 1 point [13] and intraoperative movement scale (IOMS) was less than 2 points within 5 min (Table 1) [14]. Conversely, in the case of positive reaction, the subsequent patient received a 0.02 mg/kg increment of remimazolam. Patients were then administered additional 2.5 mg doses of remimazolam for rescue sedation, with a maximum of five doses within 15 min. The study was terminated after eight crossovers were achieved. Flumazenil is readily available as a reversal agent for remimazolam in case of excessive sedation or delayed emergence.

Table 1.

Intraoperative movements scale during oocyte retrieval (IOMs)

Grade	Response
0	No movement
1	Ankles Movements (feet dorso-flexion) Non procedure interferent
2	Knees Movements (legs flexo-extension) (with/without feet movements) Non procedure interference (Aspiration could stop)
3	Pelvis/hips Movements (with/without legs/thighs movements) Aspiration must be stopped
4	Rude movements of the pelvis, chest and/or arm Aspiration must be stopped

Outcomes

The primary outcomes of the study were the ED₅₀, ED₉₀ and ED₉₅ of remimazolam in combination with sufentanil for successful sedation during transvaginal oocyte retrieval.

The secondary outcomes included: procedure time (defined as the time span from the insertion to the removal of the vaginal ultrasound probe). Record some adverse events including injection pain, bradycardia (defined as heart rate < 50 beats/min), hypoxemia ( defined as SPO₂ < 90%), hypotension (defined as mean arterial pressure drop greater than 20% of the base value), intraoperative awareness, nausea and vomiting, dizziness, and delayed emergence.

Sample size and statistical analysis

According to modified Dixon’s up-and-down method, at least six crossover points and including at least 20–40 patients were required to calculate the ED₅₀ [15]. Our study was ended when eight crossover pairs in the same direction occurred. A total of 29 patients were enrolled.

SPSS 23.0 statistical software (SPSS, Inc., Chicago, IL, USA) was used for data analysis. Normal distribution examination was assessed with Shapiro-Wilk test. Normally distributed continuous variables were expressed as mean ± standard deviation. Non-normal distribution indicators were described as median (M) and interquartile range (IQR). The ED_50, ED₉₀ and ED₉₅ of remimazolam and their corresponding 95% confidence intervals (CI) were calculated using the probit regression analysis. Graphpad Prism8 software was used to draw the sequential graphs of remimazolam and dose-effect fitting curve. P value < 0.05 was considered statistically significant.

Results

A total of thirty-two patients evaluated for eligibility, one was excluded for not meeting the inclusion criteria, and two declined participation in the trial. Finally, twenty-nine subjects completed the study (Fig. 1). The baseline characteristics of patients are presented in Table 2. The sequential doses of remimazolam combined with 0.1 µg/kg of sufentanil for effective sedation in patients undergoing transvaginal oocyte retrieval using the up-and‐down method are shown in Fig. 2. From the probit regression analysis, ED₅₀, ED₉₀ and ED₉₅ were 0.205 mg/kg (95% CI 0.190–0.222), 0.232 mg/kg (95% CI 0.218–0.311) and 0.240 mg/kg (95% CI 0.223–0.339), respectively. Dose-effect curve of remimazolam is show in Fig. 3. Table 3 includes hemodynamic changes at different time points. While both blood pressure and heart rate exhibited a degree of reduction following the administration of remimazolam, no patients developed hypotension or bradycardia. Transient hypoxemia occurred in three patients (10.3%) during the procedure and was successfully treated by performing jaw elevation to achieve tongue lift and/or temporarily increasing the oxygen flow. No other adverse events, such as injection pain, dizziness, nausea, vomiting, or delayed emergence, were observed during or after the procedure. No patient required the use of flumazenil for reversal.

Fig. 1.

Study flow diagram

Table 2.

Baseline characteristics of patients

Variable	n = 29
Age (years), mean (SD)	32.6 ± 4.3
Height (cm), mean (SD)	161.1 ± 4.2
Weight (kg), median (IQR)	56 [50–62.5.5]
BMI (kg/m2), median (IQR)	21.5[19.7–23.8]
ASA I, n(%)	21(72.4)
ASA II, n(%)	8(27.6)
Procedure time (min)	10[8–12]

SD Standard deviation, IQR Interquartile range, BMI Body mass index, ASA American Society of Anesthesiologists

Fig. 2.

The up-and-down diagram

Fig. 3.

Dose-effect relationship of Remimazolam in combination with sufentanil for oocyte retrieval

Table 3.

Hemodynamic parameters at different time points

	T0	T1	T2	T3
MAP(mmHg)	87.66 ± 9.26	77.76 ± 7.62	75.69 ± 7.86	79.14 ± 8.25
HR(beats/min)	79.00 ± 8.70	72.83 ± 8.47	70.97 ± 8.37	73.93 ± 7.68

MAP Mean arterial pressure, HR Heart rate, T0 before anesthesia induction, T1 2 min after anesthesia induction, T2 5 min after anesthesia induction, T3 At the end of procedure

Discussion

The dose-response study demonstrated that remimazolam in combination with 0.1 µg/kg of sufentanil for sedation during transvaginal oocyte retrieval were 0.205 mg/kg (95% CI 0.190–0.222), 0.232 mg/kg (95% CI 0.218–0.311) and 0.240 mg/kg (95% CI 0.223–0.339), respectively. A study showed that, for patients aged between 20 and 40 years, the optimal dose to achieve loss of consciousness (LoC) is 0.25 to 0.33 mg/kg ([16], whereas Juyeon oh et al. suggested a dose of 0.367 mg/kg [17], which were notably higher than that in our study. This may be related to the absence of analgesics, as opioids enhance the anesthetic effect of sedatives ([18].

Transvaginal oocyte retrieval was usually performed in patients by intravenous anesthesia with maintained spontaneous breathing [11]. The combination of propofol and opioids is favored in sedation due to its shorter recovery time and greater patient satisfaction. However, Previous studies have indicated that propofol might negatively impact reproductive outcomes [4–7]. Moreover, when compared with remimazolam, propofol exhibits a higher incidence of adverse events, including hypotension, hypoxemia, bradycardia, and injection pain [19, 20]. Several studies have shown that remimazolam is effective for procedural sedation, with a success rate comparable to that of propofol [9]. All participants completed this study under remimazolam anesthesia.

The Dixon’s up-and-down method is commonly employed to calculate the ED_50, which has the advantages of equiring a smaller sample size and a shorter study period. In this study, the experimental drug dose was determined by an equal difference increase or decrease method, which made it easier to calculate the dose of remimazolam and reduce the difficulty of dosing. This method usually recommended at least six crossing points. To improve the accuracy of the results, our experiment ended when eight crossing points were obtained.

While Bispectral Index (BIS) is commonly utilized for assessing anesthesia depth for some agents, it may not accurately reflect remimazolam sedation levels. A study indicated that remimazolam maintained a higher BIS level compared to propofol [21]. Despite administering the maximum dose of 0.4 mg/kg remimazolam, 21.6% of patients had an intraoperative BIS values exceeding 60 [22]. Furthermore, the interpretation of BIS values can be complex and potentially confounded by various factors, including patient-dependent pathophysiological perturbations [23]. Therefore, clinical scales like the MOAA/S are often used to monitor sedation levels during remimazolam administration [24, 25], as was done in our study. The degree of sedation is assessed through the observation of subjects’ responses to straightforward commands, thereby offering real-time feedback regarding the efficacy of sedation to clinical practitioners.

Transient hypoxemia occurred in three patients (10.3%) and was rapidly resolved with simple jaw-thrust maneuvers and/or a temporary increase in oxygen flow. The underlying mechanism is attributed to the fact that remimazolam, despite causing less respiratory depression than propofol, retains a central respiratory inhibitory effect [26, 27]. This effect is significantly enhanced by opioids like sufentanil, leading to an increased incidence of hypoxemia, with individual variations in drug response playing a role even at low doses. The 10.3% incidence observed here is highly consistent with the results of published large randomized controlled trials utilizing remimazolam-alfentanil combinations for procedural sedation. For instance, Dong et al. [26] reported a hypoxemia incidence of 9.6% in the remimazolam group for ERCP sedation in their trial. Similarly, Li et al. [27] found a hypoxemia rate of 10.8% in the remimazolam group during gastrointestinal endoscopy sedation. Furthermore, our study specifically recruited a low-risk cohort, consisting of non-obese patients while excluding those on long-term psychotropic medications (including benzodiazepines), indicating that this hypoxemia is an inherent pharmacodynamic effect. Thus, these findings collectively reinforce that vigilant respiratory monitoring is imperative during remimazolam-opioid co-administration, notwithstanding its overall favorable safety profile. The pain associated with drug injection induces both physical discomfort and anxiety in patients, subsequently diminishing their overall satisfaction. In our study, we did not observe any instances of injection pain, consistent with the findings reported by Guo et al. [28]. Conversely, Huang et al. identified an incidence rate of 6.67% [25], which may be due to the fact that our study compounded sufentanil, as opioids exert a certain degree of inhibition on injection pain [29].

This study has some limitations. First, the Modified Dixon sequential method is a classic method to calculate ED₅₀, but it is not accurate enough to predict ED₉₀ and ED₉₅. The Fixed dose grouping method or the biased coin design (BCD) up-and-down method involving larger sample sizes could have provided a more precise estimation of ED₉₀ and ED₉₅. Further research is needed to explore ED₉₀ and ED₉₅ with greater clinical relevance. Second, because a fixed dose of sufentanil was used, our findings cannot be extrapolated to situations where remimazolam is combined with different concentrations of sufentanil or with other types of opioids. Third, all participants were young women aged 20 to 45 who were classified as ASA I or II. Since both sex and age are known to correlate with the remimazolam dose required for loss of consciousness [16, 17, 30, 31], this limited demographic may restrict the generalizability of our findings to a broader population.

Conclusion

In summary, when combined with 0.1 µg/kg of sufentanil, the ED₅₀, ED₉₀, and ED₉₅ values of remimazolam for effective sedation during transvaginal oocyte retrieval were 0.205, 0.232, and 0.240 mg/kg, respectively, with a low incidence of adverse events.

Authors’ contributions

Study design and implementation: LYF, LFQ and LJH. Data collection: LYF and FJF. Data analysis/interpretation: WB and DGC. Manuscript drafting and editing: LYF, DGC and LJH. All authors read and approved the final manuscript.

Funding

This work was supported by the Department of Medical and Health Science and Technology of Zhejiang Province(No.2021KY706, No.2025KY664, No.2025KY662).

Data availability

All data in this study are available upon reasonable request via email to the corresponding author.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Zhejiang Provincial People’s Hospital (approval No.: KY2024018) and registered in the Chinese Clinical Trial Registry (registration No.: ChiCTR2400082576). Written informed consent was obtained from all patients.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.