Combined Use of Remimazolam and Propofol in Total Intravenous Anesthesia: A Retrospective Analysis

Shinju Obara^*, Yuki Watanabe Hoshino, Satoki Inoue

Department of Anesthesiology, Fukushima Medical University School of Medicine, Hikarigaoka, Fukushima, Japan

---

Abstract

Remimazolam is a newly introduced benzodiazepine anesthetic. Despite being classified as ultrashort-acting, it may cause delayed emergence owing to its not-so-short 80% context-sensitive decrement time. This study retrospectively analyzed 21 patients (12 men and 9 women; median age, 65 years; range, 17–82 years) classified as ASA I or II who underwent general anesthesia with a combination of remimazolam and propofol for anesthesia maintenance during the study period (May 2022 to May 2023). It aimed to determine the optimal effect-site concentrations of remimazolam, propofol, and remifentanil during surgery and recovery times. The population median values of individual median effect-site concentrations for remimazolam, propofol, and remifentanil during adequate sedation (patient state index 25–50) were 0.34 µg/mL, 1.00 µg/mL, and 5.00 ng/mL, respectively. At the recovery of responsiveness, the median values of individual median effect-site concentrations for remimazolam, propofol, and remifentanil were 0.17 µg/mL, 0.53 µg/mL, and 0.88 ng/mL, respectively. There was a significant negative correlation between the individual median propofol effect-site concentrations and age during patient state index range of 25–50. However, no correlation was observed between age and the other effect-site concentrations mentioned above. The patients regained responsiveness at a median of 13 min after termination of anesthetics, except for 2 who required flumazenil before tracheal extubation. Flumazenil was administered to facilitate emergence from sedation in two patients before extubation and in another two after extubation. No adverse events occurred, including resedation or intraoperative awareness. This study demonstrates the potential of combining remimazolam and propofol for effective anesthesia and rapid recovery. To optimize dosing strategies for the combination of remimazolam and propofol, larger prospective pharmacokinetic and pharmacodynamic studies are warranted.

---

Introduction

Remimazolam is a recently introduced intravenous benzodiazepine anesthetic that is classified as ultra-short-acting. From the perspective of pharmacokinetics, the context-sensitive half-time (i.e., the time required for plasma drug concentrations to decrease by 50% after discontinuation of continuous administration) of remimazolam is 4.9 min, even after 180 min of continuous administration, when simulated using pharmacokinetic (PK) model published by Schüttler et al.¹, which may seemingly result in quick recovery from anesthesia. However, the effect-site concentration (CE), not the plasma concentration, better reflects the clinical effects of intravenous anesthetics. Context-sensitive decrement time (CSDT) represents the time required for CEs to decrease by a certain percentage after the termination of infusion. Clinically, patients may experience delayed emergence from remimazolam-based anesthesia owing to its not-so-short 80% CSDT (i.e., 44 min after 180 min remimazolam administration^2,3. Although the residual effects of remimazolam can be antagonized by flumazenil, this carries the risk of resedation^4,5. Due to the pharmacodynamic (PD) interactions in the sedation effect between benzodiazepines and propofol^6,7, the combination of remimazolam and propofol is expected to reduce the required doses of each drug that may provide adequate sedation and rapid recovery of responsiveness. At present, there are extremely limited case reports involving the simultaneous use of remimazolam and propofol. Our study group reported three cases where anesthesia was maintained with remimazolam 0.2–0.3 mg/kg/h and propofol CE of 1 µg/mL⁸. However, the optimal combination of doses required for general anesthesia remains unknown. This study aimed to describe the optimal dose (i.e., effect-site concentrations) ranges of remimazolam and propofol during anesthesia maintenance and at the recovery of responsiveness after the termination of administrations when simultaneously administered. The time until recovery from anesthesia was recorded.

Methods

This study retrospectively analyzed patients who underwent general anesthesia using a combination of remimazolam and propofol. It was approved by the institutional ethics committee (No. REC2023-031). The requirement for written informed consent was waived via the opt-out method. This trial was registered with the University Hospital Medical Information Network Clinical Trial Registry: UMIN000052431. Cases in which general anesthesia was performed during the study period (May 2022–May 2023) were extracted from medical records using the following method: To induce anesthesia, remimazolam 12 mg/kg/h was administered. After confirming appropriate sedation through patient response and raw electroencephalogram pattern, muscle relaxant and opioid were administered. Subsequently, remimazolam was reduced to a maintenance dose of 1 mg/kg/h and then gradually tapered. Then, propofol was administered using the Diprifusor target-controlled infusion system (TERUFUSION Syringe Pump Type SS 3 TCI, Terumo Corporation, Tokyo, Japan). The anesthetics were titrated at the discretion of the anesthesiologist using the patient state index (PSI) value and clinical signs. During anesthesia emergence, tracheal extubation was performed as per institutional routine after confirming the resolution of muscle relaxation effects, adequate spontaneous breathing, and purposeful response to verbal commands. As part of the selection process, cases in which both propofol and remimazolam were utilized were extracted from the electronic medical records, and those that did not deviate from the aforementioned methods were selected.

The CEs of remimazolam (RZCE) and remifentanil (RFCE) were simulated based on published PK models^3,9, with the dosing history obtained from electronic anesthesia records using PKPD Tools (Minto and Schnider, URL; http://pkpdtools.com/excel/, accessed August 20, 2024). The PSI values were obtained using the SedLine system (Masimo Corporation, Irvine, CA, USA). During the maintenance period of anesthesia, combinations of remimazolam dose, RZCE, the CEs of propofol (PCE), and RFCE were extracted when the PSI value indicated an appropriate sedation depth (i.e., 25–50). For these endpoints, the median was calculated for each case, and the population median values of the aforementioned individual median values were determined. This method was used to reduce the effect of variations in data count owing to differences in the duration of each case. The time from the discontinuation of remimazolam and propofol administration to recovery of responsiveness was recorded, excluding those in patients who received flumazenil. In addition, the combinations of RZCE, PCE, and RFCE at the time of recovery of responsiveness after the discontinuation of remimazolam and propofol was extracted.

To investigate the effect of aging on PD of each anesthetic, the Pearson or Spearman correlation coefficients between the median values of remimazolam dose, RZCE, PCE, and RFCE during PSI 25–50 for each patient and age were examined. Furthermore, the correlations between the values of RZCE, PCE, and RFCE at the time of responsiveness recovery for each patient and age were examined. In addition, the correlation between time to responsiveness recovery and age was investigated. The Pearson correlation was employed if both variables were normally distributed; otherwise, the Spearman correlation was used. The Shapiro–Wilk test was employed to evaluate normality. The frequency of flumazenil use and the incidence of complications (such as intraoperative awareness, resedation after exiting the operating room, or anesthesia-related mortality) were investigated. The R software version 3.6.2 (the R foundation for Statistical Computing) was used for statistical analyses. Categorical data were expressed as n (%), whereas continuous data were expressed as median [range]. P values <0.05 were considered as statistically significant.

Result

This study analyzed 21 patients out of 4482 patients who received general anesthesia during the study period, with a median age of 65 (range: 17–82) years, a median height of 162.8 (range: 131.7–180.9) cm, and a median weight of 57.6 (range: 40.5–88.1) kg. The median body mass index was 23.4 (range: 17.5–27.4) kg/m². The cohort comprised 12 men and 9 women. According to the ASA physical status classification, 12 and 9 patients were classified as ASA I and II, respectively. Regarding comorbidities, 14.3% had hypertension (n = 3), 9.5% had diabetes (n = 2), and 4.8% had heart disease (n = 1). No patients had chronic kidney disease or cerebral vascular disease. Regional anesthesia was utilized in 47.7% of the cases (n = 10). The surgeries performed were gynecological (n = 5, 23.8%), urological (n = 4, 19.0%), oral (n = 3, 14.3%), ophthalmic (n = 1, 4.8%), otorhinolaryngological (n = 4, 19.0%), orthopedic (n = 2, 9.5%), and gastrointestinal endoscopy (n = 2, 9.5%).

The summary of remimazolam doses and the combinations of RZCE, PCE, and RFCE during PSI 25-50 are shown in Table 1. During PSI 25–50, no correlation was found between age and median of the remimazolam dose, RZCE, and RFCE, although there was a significant negative correlation between age and median PCEs (PCE = 1.41–0.005 * age, P = 0.049, r = −0.45). RZCE, PCE, and RFCE at the time of recovery of responsiveness are also summarized in the table. There was no correlation between age and RZCEs, PCEs, and RFCEs at the time of responsiveness recovery.

Table 1: Main results

Flumazenil was administered to two patients (9.5%) before tracheal extubation to facilitate emergence from anesthesia: a 61-year-old ASA I female patient with no comorbidities (Pt A) and a 74-year-old ASA I male patient with no comorbidities (Pt B). The durations from discontinuation of remimazolam and propofol to flumazenil 0.2 mg administration (RZCE, PCE, and RZCE at that time) in Pt A and Pt B were 18 min (RZCE 0.11 µg/mL, PCE 0.25 µg/mL, and RFCE 0.37 ng/mL) and 14 min (RZCE 0.19 µg/mL, PCE 0.28 µg/mL, and RFCE 0.26 ng/mL), respectively. Both patients were discharged from the operating room without any problems and did not experience resedation. Furthermore, flumazenil was administered to two patients (9.5%) after tracheal extubation to facilitate emergence from anesthesia: a 68-year-old ASA II male patient with controlled chronic obstructive pulmonary disease (Pt C) and a 73-year-old ASA II male patient with controlled hypertension (Pt D). The durations from discontinuation of remimazolam and propofol to responsiveness recovery in Pt C and Pt D were 21 and 12 min, respectively. Both patients were extubated after confirming that they were awake (Pt C: RZCE 0.10 µg/mL, PCE 0.6 µg/mL, RFCE 0.85 ng/mL; Pt D: RZCE 0.14 µg/mL, PCE 0.5 µg/mL, RFCE 0.90 ng/mL). However, both subsequently fell into moderate sedation and became responsive only to stimulation. Although the discharge criteria from the operating room were met, to promote full awakening, flumazenil was administered at 2 and 6 min postextubation, after which both patients fully awoke and were transferred back to their rooms without resedation.

In the 19 cases without flumazenil administration before the extubation, the median duration of anesthesia was 244 (range: 118–653) min. The patients in these cases regained responsiveness and were extubated within a median of 13 (range: 7–20) min after termination of both anesthetics. No correlation was observed between time to recovery of responsiveness and age. No anesthesia-related complications were observed in any cases.

Discussion

To date, reports on the combination of remimazolam and propofol are scarce⁸. Considering the PD interactions between propofol and benzodiazepine-based anesthetics, it is theoretically possible to achieve anesthesia with relatively small doses of each^6,7 that may enable rapid emergence from anesthesia.

This retrospective study demonstrates the potential of maintaining anesthesia without complications and achieving rapid responsiveness recovery in most cases using the aforementioned combination. In the present study, during anesthesia maintenance under the conditions of RFCEs tolerant to nociceptive stimulation, the population median values of the individual median dose of remimazolam (0.30 mg/kg/h) and PCe (1.00 µg/mL) were much smaller than their respective standard doses when administered separately as specified in the package inserts (i.e., 1 mg/kg/h and 3 µg/mL, respectively) or in the studies by Doi et al.¹⁰ (remimazolam 0.99 ± 0.39 mg/kg/h) and Ogawa et al.¹¹ (3 µg/mL), which might have been due to the PD interaction between remimazolam and propofol. As regards the emergence time from anesthesia, Doi et al. reported that after anesthesia induction with remimazolam at 12 mg/kg/h and adjustment of the initial infusion dose of 1 mg/kg/h at the anesthesiologist’s discretion, the time to extubation was 19.2 ± 10.8 min, which seemed longer than our results. From a PK perspective, the 80% CSDT for each anesthetic is not substantially short. For example, in a 40-year-old man, after 244 min of remimazolam and propofol administration, the 80% CSDTs are 61 and 71 min, respectively, according to simulations using published PK models^3,12. However, if the maintenance concentrations are low, the time it takes to drop from administration discontinuation to the awakening range is faster than when the maintenance concentrations are “standard,” and the effects of enhanced interaction between the two drugs are likely to be reduced. This rapid recovery would be significant for patient turnover and reduction of the overall time in the operating room. As regards the PD during recovery from anesthesia, Kim et al.¹³ reported in a retrospective simulation study using the Schüttler model that the median RZCE for responsiveness recovery from TIVA with remimazolam and remifentanil was 0.3 µg/mL. Furthermore, Im et al.¹⁴ reported in an RCT that the PCE at eye opening in patients aged 40–54 years in TIVA with propofol and remifentanil was 0.9 ± 0.2 µg/mL. Our results indicate that the patients recovered from anesthesia at a lower RZCE (although caution is needed in interpretation due to the different PK model used in Kim et al.’s study) and a lower PCE than those of these studies, which may also be due to the PD interaction between remimazolam and propofol.

With respect to age and the required amount of anesthetic during maintenance within the PSI range of 25–50, it is well established that age affects the PD of propofol, allowing for dose reduction with increasing age¹⁵, consistent with our finding that the median of the PCE during anesthesia maintenance is negatively correlated with age. As regards remimazolam, Masui et al.³ have demonstrated that age does not have much effect on the PK of remimazolam. Therefore, if age affects the dosage requirements of remimazolam, it is likely due to changes in PD. In our study, unlike with propofol, age did not affect RZCE during maintenance. One possible reason is that the impact of aging on the PD of remimazolam might be small. However, this contradicts the previous findings indicating that the dose and RZCE required for sedation are lower in older patients^16,17 and the evidence that the PD of midazolam—a benzodiazepine anesthetic similar to remimazolam, assessed via EEG—decreases with age, reducing the plasma concentration needed for sedation¹⁸. At present, there are no published studies that thoroughly examined the association between age, CEs predicted using the Masui model, and clinical effects of remimazolam, indicating the need for future PD studies. Another factor could be the use of PSI as a PD index and the treatment of PSI values between 25 and 50 uniformly as indicative of appropriate sedation depth. PSI reportedly increases with age during appropriate general anesthesia with sevoflurane, desflurane, or propofol¹⁹. Although the influence of age on PSI during remimazolam anesthesia remains unknown, if a similar effect exists in remimazolam, practitioners may misinterpret a PSI value above 50 in elderly patients—despite it indicating appropriate depth—as a reason to increase the dose, potentially leading to overdosage. Similarly, if deep sedation is actually achieved but PSI remains above 25, they may avoid reducing the dose, which also results in overdosage. Thus, the effect of age on PSI may have prevented a lower RZCE from being observed in elderly patients. The association between PSI and RZCE during remimazolam use as well as the effect of age on this association should be further investigated. In addition, the unknown effects on EEG of the combined use of remimazolam with propofol and remifentanil as well as the small sample-size possibly limiting the detection of age-related effects on remimazolam PD could explain our findings. In our study, the lack of an age effect on RZCE during recovery from anesthesia contradicts the findings of Kim et al.¹³, which indicated that age was significantly correlated with the simulated RZCE at responsiveness recovery. Notably, the PK model used by Kim et al.¹³, developed by Schüttler et al., was based on a younger population and does not cover elderly patients. However, Masui et al.³ reported that age does not have a substantial effect on the PK of remimazolam, which may lend some credibility to the findings of Kim et al. In our study, the small sample-size may have also influenced the results.

A larger, prospective PK/PD study is warranted to further investigate the appropriate dosages of these two agents. Comparing prop/remifentanil with prop/remifentanil/remimidazolam by examining the sedative effects across wide ranges of effect-site concentration combinations would be of interest, using a sophisticated approach such as a criss-cross study design²⁰, across a broad population.

In this study, flumazenil was utilized in 9.5% of the cases before tracheal extubation. Phase IIb/III trial examining the efficacy and safety of remimazolam in ASA I or II patients, 8.7%–9.3% of the patients took more than 30 min to awaken from remimazolam anesthesia and required flumazenil administration¹⁰. Although this administration rate was similar to our findings, in our study, flumazenil was administered at the discretion of the anesthesiologist in less than 30 min, indicating that natural awakening might have occurred if more time had been allowed. This represents a limitation of the current study. Furthermore, in two cases from our study, flumazenil was administered after the patients had spontaneously awakened and were in a sedated state postextubation. We hypothesized that the presence of the tracheal tube served as a stimulus for awakening, and once the tube was removed, the absence of this stimulus made the residual sedative effects apparent. Therefore, while the combination of propofol and remimazolam enables rapid awakening, it is crucial to closely monitor the postextubation state and, if necessary, administer an antagonist. In our study, no specific characteristics in patient background or CE of anesthetics were identified in cases requiring flumazenil. Because complete awakening was achieved with flumazenil, individual variability in the PK and/or PD of remimazolam was presumed to contribute to these cases.

This study has several limitations. First, as a preliminary descriptive study intended to inform future prospective research, and due to the lack of similar previous studies, sample-size estimation was not performed. Second, because cases were selected from a specific period and followed a somewhat standardized methodology, the sample size was small, and there was no comparison group, which led us to conduct a descriptive study.

Conclusion

The study demonstrates the potential of using a combination of remimazolam and propofol to effectively maintain anesthesia while ensuring rapid recovery times. This anesthetic method may be beneficial, especially in neurosurgical procedures, particularly for patients requiring intravenous anesthesia to avoid the effects of inhalation agents on intraoperative neuromonitoring and for whom rapid and reliable emergence is crucial to enable early detection of any postoperative neurological deficits. However, the findings should be interpreted with caution due to the study limitations. Future research should focus on larger, prospective studies with detailed PK and PD analyses to optimize dosing strategies and enhance patient outcomes in clinical anesthesia practice. 

List of Abbreviations

PK: pharmacokinetics

CE: effect-site concentration

CSDT: context-sensitive decrement time

PD: pharmacodynamics

UMIN: University Hospital Medical Information Network Clinical Trial Registry

PSI: Patient state index

RZCE: effect-site concentrations of remimazolam

RFCE: effect-site concentrations of remifentanil

PCE: effect-site concentrations of propofol

Declarations

Ethics Approval and Consent to Participate

This study was approved by the institutional ethics committee (No. REC2023-031). This trial is registered with UMIN Clinical Trial Registry: UMIN000052431.

Consent for Publication

The requirement for written informed consent was waived via the opt-out method.

Availability of Data and Material

Not applicable

Competing Interests

All authors report no conflicts of interest.

Funding

The authors declare no funding for this report.

Credit Author Statement

Shinju Obara: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft.

Yuki Watanabe Hoshino: Investigation, Writing – review & editing.

Satoki Inoue: Supervision, Writing – review & editing.

Acknowledgements

We would like to thank the Enago (https://www.enago.jp/) for editing this manuscript.

Presentation

This study was presented in abstract form at the 18th World Congress of Anaesthesiologists (March 4, 2024, Singapore).

References

1. Schüttler J, Eisenried A, Lerch M, et al. Pharmacokinetics and Pharmacodynamics of Remimazolam (CNS 7056) after Continuous Infusion in Healthy Male Volunteers: Part I. Pharmacokinetics and Clinical Pharmacodynamics. Anesthesiology. 2020; 132(4): 636-51.
2. Obara S. Simulation of residual sedation effect of remimazolam: pharmacokinetic-pharmacodynamic simulation can be an additional standard anesthesia monitoring method. J Anesth. 2022; 36(2): 167-70.
3. Masui K, Stöhr T, Pesic M, et al. A population pharmacokinetic model of remimazolam for general anesthesia and consideration of remimazolam dose in clinical practice. J Anesth. 2022; 36(4): 493-505.
4. Masui K. Caution!! Reappearance of remimazolam effect after a flumazenil bolus: a larger bolus of flumazenil and a lower total remimazolam clearance are higher risks. J Anesth. 2022.
5. Obara S, Kamata K, Nakao M, et al. Recommendation for the practice of total intravenous anesthesia. J Anesth. 2024.
6. Hendrickx JF, Eger EI 2nd, Sonner JM, et al. Is synergy the rule? A review of anesthetic interactions producing hypnosis and immobility. Anesth Analg. 2008; 107(2): 494-506.
7. Short TG, Plummer JL, Chui PT. Hypnotic and anaesthetic interactions between midazolam, propofol and alfentanil. Br J Anaesth. 1992; 69(2): 162-7.
8. Watanabe Y, Obara S, Inoue S. Total intravenous anesthesia management with simultaneous use of remimazolam and propofol: A case series of three patients. SAGE open medical case reports. 2023; 11: 2050313x231204574.
9. Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development. Anesthesiology. 1997; 86(1): 10-23.
10. Doi M, Morita K, Takeda J, et al. Efficacy and safety of remimazolam versus propofol for general anesthesia: a multicenter, single-blind, randomized, parallel-group, phase IIb/III trial. J Anesth. 2020; 34(4): 543-53.
11. Ogawa T, Obara S, Akino M, et al. The predictive performance of propofol target-controlled infusion during robotic-assisted laparoscopic prostatectomy with CO(2) pneumoperitoneum in the head-down position. J Anesth. 2020; 34(3): 397-403.
12. Marsh B, White M, Morton N, et al. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth. 1991; 67(1): 41-8.
13. Kim KM, Bang JY, Lee JM, et al. Effect-site concentration of remimazolam at loss and recovery of responsiveness during general anesthesia: a simulation study. Anesth Pain Med. 2022; 17(3): 262-70.
14. Im KS, Kim JB, Lee JM, et al. The optimal concentrations of propofol at eye opening and orientation after propofol-remifentanil TCI in Koreans. Anesth Pain Med. 2011; 6(2): 131-37.
15. Schnider TW, Minto CF, Shafer SL, et al. The influence of age on propofol pharmacodynamics. Anesthesiology. 1999; 90(6): 1502-16.
16. Chae D, Kim H-C, Song Y, et al. Pharmacodynamic analysis of intravenous bolus remimazolam for loss of consciousness in patients undergoing general anaesthesia: a randomised, prospective, double-blind study. Br J Anaesth. 2022; 129(1): 49-57.
17. Oh J, Park SY, Lee SY, et al. Determination of the 95% effective dose of remimazolam to achieve loss of consciousness during anesthesia induction in different age groups. Korean J Anesthesiol. 2022; 75(6): 510-7.
18. Albrecht S, Ihmsen H, Hering W, et al. The effect of age on the pharmacokinetics and pharmacodynamics of midazolam. Clin Pharmacol Ther. 1999; 65(6): 630-9.
19. Obert DP, Schneider F, Schneider G, et al. Performance of the SEDLine Monitor: Age Dependency and Time Delay. Anesth Analg. 2023; 137(4): 887-95.
20. Short TG, Ho TY, Minto CF, et al. Efficient trial design for eliciting a pharmacokinetic-pharmacodynamic model-based response surface describing the interaction between two intravenous anesthetic drugs. Anesthesiology. 2002; 96(2): 400-8.