Does Propofol-based or Remimazolam-based Anesthesia Impact the Incidence of Postoperative Nausea and Vomiting When Used In Combination With Prophylactic Dexamethasone and Ondansetron?

Postoperative nausea and vomiting (PONV) after general anesthesia is a frequent complication. The incidence of PONV is generally reported to be 20% to 30%¹; however, this rate increases to approximately 80% as the number of patient risk factors increase.² Oral surgery is considered to have a higher risk for PONV than other surgeries due to bleeding from the surgical field being swallowed.³ Orthognathic surgery, in particular, involves intermaxillary fixation which restricts mouth opening and may lead to aspiration and airway obstruction following emesis⁴; as such, prevention and treatment are important. At our institution, we administer a combination of dexamethasone, effective for preventing edema and PONV, and ondansetron, the gold standard for PONV prevention, in orthognathic surgery. Combined administration of dexamethasone and ondansetron has been reported to significantly reduce postoperative nausea after orthognathic surgery in the early phase (≤2 hours after the end of anesthesia).⁵

To date, propofol was the only intravenous (IV) anesthetic used for maintenance of general anesthesia; however, in August 2020, remimazolam was launched in Japan. Currently, there are only 2 IV anesthetics that can be used for total intravenous anesthesia (TIVA) in Japan: propofol and remimazolam. Therefore, we retrospectively compared the incidence of PONV in patients undergoing orthognathic surgery under TIVA using propofol or remimazolam. The incidence of PONV in the early postoperative period (≤2 hours after completion of anesthesia) was significantly higher in the remimazolam group than in the propofol group when no prophylactic antiemetic was administered.⁶ Therefore, we conducted this study to evaluate the significance of prophylactic administration of dexamethasone and ondansetron to reduce the incidence of PONV during remimazolam anesthesia.

The primary objective of this study was to compare the incidence of PONV between propofol and remimazolam anesthesia in combination with prophylactic dexamethasone and ondansetron. The secondary objective was to investigate the severity of PONV and the quality of postoperative recovery respectively.

METHODS

This randomized clinical trial was conducted at Tokyo Dental College Suidobashi Hospital (Tokyo, Japan). The trial was approved by the Institutional Ethics Review Board of Tokyo Dental College (Approval No., 1182; Approval date, June 23, 2023) and registered in the University Hospital Medical Information Network (Clinical trial registration number, UMIN000051512; Registration date, July 3, 2023). Written informed consent was obtained from all patients participating in the study. The study was conducted in accordance with the principles of the Declaration of Helsinki and in compliance with the CONSORT guidelines.

Subjects were 20 to 50 years of age and preoperative American Society of Anesthesiologists physical status (ASA-PS) classification I or II scheduled for bilateral sagittal split ramus osteotomy or genioplasty under general anesthesia. Patients were excluded if unable or unwilling to provide consent for the study or if they had emergency surgery, contraindications to any of the study medications, a body mass index (BMI) greater than or equal to 30 kg/m², taken antiemetic medications within 24 hours before surgery, or were regular users of benzodiazepines.

Using a computer-generated random numbers table, patients were randomly assigned to 1 of 2 groups: propofol (group P) or remimazolam (group R). Group assignment was performed by a dental anesthesiologist not participating in the study.

All patients fasted for at least 8 hours before surgery and received no premedication. General anesthesia was performed according to the following protocol. After the patient arrived in the operating room (OR), standard anesthetic monitors (noninvasive blood pressure cuff, pulse oximeter, electrocardiogram, capnography, and bispectral index [BIS]) and a muscle relaxation monitor (electromyography, AF-201P, NIHON KOHDEN) were attached, peripheral IV access was secured, and a remifentanil infusion (0.5 µg/kg/min) was initiated. General anesthesia was induced using a target-controlled infusion (TCI) of propofol (4 µg/mL) or a continuous infusion of remimazolam (12 mg/kg/h), followed by nasotracheal intubation after administration of rocuronium (0.6 mg/kg). After intubation, a pharyngeal pack was inserted, and IV dexamethasone (6.6 mg) was administered. Infiltration and alveolar nerve blocks using 2% lidocaine with epinephrine 1:80,000 were administered before surgery commenced. General anesthesia was maintained with propofol or remimazolam along with oxygen (1 L/min), air (2 L/min), and remifentanil. For propofol and remimazolam, infusion rates were adjusted accordingly to achieve BIS values between 40 and 60. Remifentanil was lowered to a rate of 0.2 µg/kg/min and increased or decreased by 0.05 µg/kg/min if the systolic blood pressure increased or decreased by 20% or more compared with 5 minutes earlier. A rocuronium infusion (7 µg/kg/min) was started, the degree of neuromuscular blockade was measured every 5 minutes using a train-of-four (TOF) monitoring, and a single dose of rocuronium (0.1–0.2 mg/kg) was administered when the TOF count was 4.

Ventilatory settings were set at a tidal volume of 6 to 10 mL/kg, a rate of 8 to 12 breaths/min, a mean end-expiratory carbon dioxide partial pressure of 35 ± 5 mm Hg, and an end-expiratory positive pressure of 5 cmH₂O, and an IV infusion (acetate Ringer’s solution) was run at 5 to 10 mL/kg/h. Fentanyl (2 µg/kg) was administered approximately 20 minutes before the end of surgery, and ondansetron (4 mg) was administered approximately 15 minutes before the end of surgery.

After completion of surgery, the degree of neuromuscular blockade recovery was assessed using a muscle relaxation monitor, and sugammadex (2–4 mg/kg) administered as needed. Flumazenil (0.2 mg) was administered in group R, and, if the desired state of arousal was not achieved within 4 minutes after administration, a dose of 0.1 mg was repeated at every 1 minute (maximum total dose, 1 mg). Patients were extubated after confirming adequate recovery of spontaneous ventilation and responsiveness to verbal commands.

The severity of PONV was assessed using a numerical rating scale (NRS; 0, no nausea; 10, worst nausea imaginable) at 3 times: immediately after the end of anesthesia (right before leaving the OR), 2 hours later, and 24 hours later. If the patient complained of nausea or vomiting, the timing was recorded and the NRS was reassessed. If the NRS score was more than 4, IV metoclopramide (10 mg) was administered for rescue. Metoclopramide was administered up to 2 times per day as needed and the timing recorded if administered. Twenty-four hours after the end of anesthesia, the quality of postoperative recovery was assessed using the Quality recovery 40 (QoR-40) tool.⁷

Patient demographics (age, sex, weight, smoking history, motion sickness, and history of PONV) and surgical/anesthetic factors (surgery type, operation time, anesthesia time, intraoperative opioid use [remifentanil or fentanyl], blood loss during surgery, total infusion volume, and postoperative fentanyl use) were compared between the 2 groups. The incidence and timing of PONV and the maximum NRS scores were compared between the groups at the following times: overall (extubation to 24 hours later), immediate (extubation to end of anesthesia), early (end of anesthesia to 2 hours later), and late (2 to 24 hours later). The groups were also compared with respect to metoclopramide administration and total QoR-40 scores along with each of its 5 categories (physical comfort, physical independence, patient support, pain, and emotional state).

RESULTS

This study recruited a total of 129 patients who fulfilled the inclusion criteria between July 2023 and May 2024. Of these, 6 could not give consent, 3 had a BMI greater than or equal to 30 kg/m², 1 used an antiemetic within 24 hours before surgery, 1 regularly used benzodiazepines, and 2 had contraindications to study medications and were excluded. In addition, 16 patients were excluded due to study protocol deviations. As such, data from a total of 100 patients were ultimately included in the study’s analysis (Figure).

View Figure

• Download as Powerpoint
• Download Figure

There were no significant differences in patient demographics between the 2 groups. The mean total remifentanil use was significantly higher in group R (P = .03). However, there were no significant differences in the other surgical/anesthetic factors (Table 1).

Table 1.Patient Demographics and Surgical/Anesthetic Factors

View Fullscreen

The overall incidence of PONV within 24 hours after surgery was higher in the remimazolam group. A total of 3 (6%) patients in group P and 9 (18%) patients in group R had PONV overall, but this difference lacked any statistical significance (P = .12). The incidence of PONV during each period (immediate, early, and late) was higher in group R but was not statistically significant. Similarly, nausea overall and vomiting overall both had higher incidence rates in group R vs group P, but neither difference was statistically significant. Aside from the immediate period, both also had increased incidence rates in the early and late periods; however, there were no significant differences between the groups at any of the time periods (Table 2).

Table 2.PONV Incident Rates

View Fullscreen

Although group R had higher maximum NRS scores than group P overall and at each postoperative period, none of those differences were statistically significant (Table 3).

Table 3.Postoperative NRS Scores

View Fullscreen

More total patients in group R than in group P received metoclopramide rescue; however, there were no statistically significant differences found between the groups (Table 4).

Table 4.Metoclopramide Use

View Fullscreen

Similarly, the total mean QoR-40 scores and the mean scores for its 5 categories lacked any statistical significance between the groups (Table 5).

Table 5.QoR-40 Scores

View Fullscreen

DISCUSSION

We compared the incidence of PONV between propofol-based vs remimazolam-based TIVA when used in combination with prophylactic dexamethasone and ondansetron. The overall incidence of PONV was 6% in group P and 18% in group R, which lacked significance. Although the incidence of PONV is reportedly higher with remimazolam than with propofol when used for TIVA during orthognathic surgery,⁶ the results of this study revealed no significant difference between the two groups. As such, the prophylactic administration of dexamethasone and ondansetron combined when used during remimazolam-based TIVA may reduce the incidence of PONV to a similar level as propofol-based TIVA.

The incidence of PONV without antiemetics during remimazolam anesthesia in orthognathic surgery has been reported to be 51.2%,⁶ while the incidence of PONV in group R in this study was 18%. Therefore, it is possible that the combined administration of dexamethasone and ondansetron during remimazolam anesthesia could significantly reduce the incidence of PONV. Because PONV after orthognathic surgery is a complication that should be avoided,⁴ administration of dexamethasone and ondansetron during remimazolam anesthesia for orthognathic surgery was considered desirable. In contrast, the incidence of PONV in oral surgery procedures with low PONV risk (eg, a tooth extraction or cystectomy) was reported to be 11.7% when ramosetron (a 5-HT3 antagonist) was administered,⁹ which was lower than the incidence found in this study. Because it is recommended to administer multiple antiemetics that act on different receptors for surgeries with high PONV risk,¹⁰ 2 different antiemetics were administered in combination in this study. However, for surgeries with low PONV risk, a single antiemetic agent may be sufficient to reduce the incidence of PONV.

In this study, flumazenil was administered to all patients in group R to ensure rapid awakening. The context-sensitive half-time (CSHT) of remimazolam has been reported to be 16.4 minutes (when administered to people in their 30s for 4 hours),¹¹ and it would take longer to wake up without the use of flumazenil. As such, we decided to administer flumazenil to group R for the purposes of this study.

In a study comparing PONV incidence in flumazenil-treated and nontreated groups for gynecological day surgery performed under remimazolam anesthesia, the incidence of PONV was significantly higher in the flumazenil group.¹² Because midazolam has been reported to have PONV inhibitory activity comparable with that of ondansetron,¹³ it is possible that remimazolam also has PONV inhibitory activity similar to that of midazolam. Therefore, it is possible that the administration of flumazenil antagonized this PONV inhibitory effect as well as the sedative effect. Accordingly, it is conceivable that the incidence of PONV in group R could have been less than 18% in this study if the flumazenil was not administered.

In this study, the total intraoperative dose of remifentanil was significantly higher in the remimazolam vs the propofol group, although the incidence of PONV was not significantly different. Intraoperative blood pressure was significantly higher during remimazolam anesthesia than during propofol anesthesia,¹⁴ therefore, the dosage of remifentanil may have been increased to maintain blood pressure in the optimal range. Because opioids are considered to be a risk factor for PONV, it is possible that the difference in the total dose of remifentanil affected the incidence of PONV. In this study, there was no significant difference in the incidence of PONV between the two groups; however, the incidence of PONV was approximately 3 times higher in group R vs group P (18% vs 6%, respectively), suggesting that remifentanil could have been associated with the increased incidence of PONV in group R.

The maximum NRS scores were not significantly different between the 2 groups overall or for any of the study periods. The combined administration of dexamethasone and ondansetron did not change the severity of PONV in the propofol and remimazolam groups during any of the periods based on the NRS data.

Similarly, the QoR-40 data revealed no significant differences between the 2 groups in total score or in the scores of the 5 categories, indicating no differences in the quality of postoperative recovery. One of the 5 categories, “pain”, has been reported to be correlated with PONV, with more pain associated with a higher incidence of PONV.¹⁵ However, in this study, there was no significant difference between the 2 groups, suggesting that the quality of postoperative recovery, including pain, had no effect on PONV.

This study had several limitations, the first being its small sample size which was calculated based on the results of a previous study. Nevertheless, our results may need to be verified using a larger number of subjects in future studies. Second, subjects included in this study were relatively young and healthy (20–50 years of age; ASA-PS I or II). Therefore, it is unclear whether similar results could be obtained for older patients and/or those with health problems. However, in this study, we wanted to verify the incidence of PONV in orthognathic surgery, so we ended up with young, healthy patients as they are typically who undergo such surgeries. Thirdly, flumazenil was administered only to the remimazolam group in the hope of rapid awakening. Flumazenil being given only to the remimazolam group may have affected the incidence of PONV. Therefore, it is possible that the incidence of PONV increased with the administration of flumazenil. Future studies should consider not using flumazenil.

CONCLUSION

The incidence of PONV 24 hours after orthognathic surgery administered under remimazolam anesthesia with prophylactic dexamethasone and ondansetron was not significantly different from that administered under propofol anesthesia with prophylactic dexamethasone and ondansetron. There were also no significant differences in the severity of PONV nor the quality of recovery between the 2 groups for any of the study periods.