Ramosetron,midazolam,and combination of ramosetron and midazolam for prevention of postoperative nausea and vomiting: A prospective,randomized,double-blind study

Abstract

Objective

To compare the efficacy of ramosetron, midazolam, and the combination of ramosetron and midazolam in the prevention of postoperative nausea and vomiting (PONV) in female patients undergoing thyroidectomy.

Methods

Patients were randomized to receive 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before the induction of anaesthesia. PONV, use of rescue antiemetics, pain severity and fentanyl consumption were assessed for 24 h after thyroidectomy.

Results

A total of 100 patients were enrolled; 94 patients completed the study. The severity of nausea was statistically significantly reduced at 0–2 h in Group RM compared with Groups R and M, and at 2–6 h in Group RM compared with Group M. The incidence of retching and vomiting was significantly lower at 0–2 h, 2–6 h, 6–12 h in Group RM than in Group M, and lower in Group R than Group M at 6–12 h. The requirement for rescue antiemetics was significantly lower at 0–2 h in group RM than in group M.

Conclusion

The combination therapy of ramosetron and midazolam provided superior antiemetic efficacy compared with midazolam single therapy, but did not show any significant additional benefits compared with ramosetron single therapy.

Introduction

Postoperative nausea and vomiting (PONV) is a frequent and distressing outcome of surgery that can adversely influence a patient’s recovery. ¹ PONV is influenced by a variety of factors that include the characteristics of the patient, the surgery and the anaesthesia used.^1,2 Moreover, female sex, nonsmoking, use of postoperative opioids and a prior history of motion sickness or PONV are well-known risk factors for PONV. ³

Thyroidectomy is one of the most common elective surgeries, and the incidence of associated PONV is relatively high, ranging between 60 and 84%.^4,5

Considering the multifactorial aetiology of PONV and the current evidence suggesting the limited efficacy of single antiemetic prophylaxis, combination therapies (involving of antiemetics with different mechanisms of actions) may offer a more effective approach in preventing PONV, in patients at moderate to high risk of PONV.^6,7

The 5-hydroxytryptamine receptor 3 (5-HT₃) antagonists are frequently used for the prevention of PONV. ⁸ Among the 5-HT₃ antagonists, ramosetron (Nasea®; Astellas Pharma Korea Inc., Seoul, Republic of Korea) is a newly developed drug with higher 5-HT₃ receptor affinity and longer duration of action than its congeners such as ondansetron and granisetron. ⁹ Midazolam, which is a rapid and short-acting benzodiazepine, has been widely used as a premedicant: several investigators have demonstrated the antiemetic properties of midazolam, when administered as a bolus before or after induction of anaesthesia, or given postoperatively as a continuous infusion.^10–12

Although several studies have reported the combination therapy of 5-HT₃ antagonists and antiemetics with different mechanisms of action,^13,14 the prophylactic effect of a combination of ramosetron and midazolam for PONV has not been assessed in patients after thyroidectomy. It was hypothesized that combination therapy with ramosetron and midazolam would be more effective than single therapy with either drug, in preventing PONV. Thus, the present study compared the antiemetic efficacy of ramosetron, midazolam, and the combination of ramosetron and midazolam, in women undergoing thyroidectomy.

Patients and methods

Study population

This prospective, randomized, double-blinded study enrolled consecutive female patients scheduled for thyroidectomy under general anaesthesia at Chung-ang University Hospital, Seoul, Republic of Korea, between January and September 2012. Eligible patients were non-smoking females, aged 20–65 years, with American Society of Anesthesiologists physical status of I or II (http://www.asahq.org/Home/For-Members/Clinical-Information/ASA-Physical-Status-Classification-System), where the need for opioids for postoperative pain management was anticipated. Patients with three or four risk factors of PONV were selected. Patients with diabetes mellitus or gastrointestinal disease, and those who were smokers, menstruating or had taken an antiemetic medication ≤72 h before surgery were excluded from the study. Decisions to enrol or exclude patients were made by an investigator at the preoperative anaesthetic evaluation who did not otherwise participate in conducting the study and data collection.

Study approval was obtained from the Ethical Committee of Chung-ang University, Seoul, Republic of Korea, and this study was carried out according to the principle of the Declaration of Helsinki (2000 version). Written consent was obtained from all patients after they were given an explanation about the purpose of the study, the side-effects that might occur, and how to use the intravenous patient-controlled analgesia (IV–PCA) device. This trial is registered with Australia New Zealand Clinical trials Registry (ANZCTR) (ACTRN12612000757819; Web address: https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=362765).

Study randomization

Patients were divided into three groups based on the antiemetic treatment they were to receive: ramosetron alone (Group R); midazolam alone (Group M); ramosetron combined with midazolam (Group RM). The randomization was based on a random table, generated using the R Project for Statistical Computing (http://www.r-project.org/). Block randomization with a block size of four and equal allocation was employed to prevent imbalances in treatment assignments. The randomization sequence was generated by a statistician who was not involved with the study. Patient group allocation was revealed to the anaesthetist immediately prior to induction of anaesthesia, by means of numbered and sealed envelopes.

Antiemetic therapy and anaesthesia procedures

Patients fasted for ≥8 h for solid food and ≥2 h for clear fluids, and were hydrated appropriately prior to surgery. Patients arrived in the operating room without any premedication. After recording noninvasive blood pressure, heart rate and peripheral oxygen saturation, the randomly selected experimental medication was injected intravenously as a bolus over 30 s, immediately before anaesthesia. Patients in Group R received 0.3 mg ramosetron (Nasea®, Astellas Pharma Korea Inc., Seoul, Republic of Korea); patients in Group M received 75 µg/kg midazolam (Midazolam Injection®, Bukwang Pharm, Seoul, Republic of Korea); patients in Group RM received 0.3 mg ramosetron combined with 75 µg/kg midazolam, administered as for the single medications.

Anaesthesia was induced with 2 µg/kg fentanyl i.v. and 5 mg/kg thiopental i.v.; orotracheal intubation was facilitated with 0.8 mg/kg rocuronium bromide i.v. Inhalation anaesthesia was maintained with 1.5–2.5% sevoflurane in 50% O₂/N₂O. While maintaining anaesthesia, the tidal volume was regulated to keep the end tidal CO₂ pressure in the range of 30–35 mmHg. Lactated Ringer's solution was given at a rate of 5 ml/kg/h by i.v. infusion throughout surgery. Approximately 15 min before the end of surgery, 0.5 µg/kg fentanyl was administered i.v. for postoperative pain control and the IV–PCA device was connected. The instrument used for IV–PCA was the Basal/Bolus Infusor (Baxter Healthcare Corporation, Deerfield, IL, USA) containing 20 µg/kg fentanyl in a final solution volume of 100 ml. Continuous infusion was set at 0.2 ml/h with a bolus of 1.2 ml and a lockout time of 15 min. Muscle relaxation was antagonized by a combination of 0.4 mg glycopyrrolate i.v. and 15 mg pyridostigmine i.v. at the end of surgery. The patient was extubated when fully awake. Anaesthesia duration was defined as the period from induction until the discontinuation of nitrous oxide, and was recorded for all patients.

Assessment of efficacy

The primary endpoint of this study was the severity of PONV. The severity of nausea, incidence of retching and vomiting together, and the need for rescue antiemetics, were evaluated at 2 h in the postanaesthesia care unit (PACU) and from 2–6, 6–12, and 12–24 h in the ward. Patients were questioned by an anaesthetic nurse, who was unaware of the group identities, as to the severity of nausea and incidence of retching and vomiting. Nausea was defined as per the report of the patient when they felt nauseated, that is, ‘I feel like vomiting or a have a sensation like motion sickness’. Retching was defined as the laboured, spasmodic, rhythmic contractions of respiratory muscles without the expulsion of gastric contents. Vomiting was defined as the forceful expulsion of gastric contents. Severity of nausea was graded using an 11-point verbal numerical rating scale (VNRS), with 0 representing no nausea and 10 representing worst possible nausea. When the patient had nausea (by scoring at level 5 on the VNRS or requesting an antiemetic), the first-line rescue antiemetic was given: 10 mg metoclopramide i.v.; 5 mg dexamethasone i.v. was the second-line antiemetic treatment.

Postoperative pain intensity was assessed using a visual analogue scale (VAS) that ranged between 0 mm (no pain) and 100 mm (worst pain imaginable). Patients received 30 mg of ketorolac i.v. if they complained of pain (>40 mm on the VAS), and the total amount of infused fentanyl from the IV–PCA device over 24 h was recorded.

Assessment of safety

The incidence of the most frequently reported side-effects of the 5-HT₃ antagonists (including headache, dizziness, and drowsiness) were also assessed. ¹⁵ No other side-effects were recorded. These variables were assessed by an anaesthetic nurse, who was unaware of the group identities.

Statistical analyses

In a pilot study involving 10 subjects who underwent thyroidectomy with ramosetron, the SD of the VNRS was 1.2 (unpublished data). Hypothesizing that the SD would be equal for the three groups (midazolam, ramosetron, and ramosetron combined with midazolam), the results using the calculated value with an α-error of 5% and a power of test of 80% to distinguish a difference in the VNRS of 1, showed that 30 subjects were required for each group of patients. Considering patient refusal rate and compliance rate as 5% and 90%, respectively, 100 patients were required for the present study.

For intergroup comparisons, the distribution of the data was first evaluated for normality using the Shapiro–Wilk test. Normally distributed data were presented as the mean ± SD or mean ± SE. Groups were compared using analysis of variance and a post hoc Tukey test. The non-normally distributed data were expressed as medians (quartile 1–3) and were analysed using the Kruskal–Wallis test with Bonferroni’s correction.

Descriptive variables were presented as n (%) and analysed using χ²-test or Fisher’s exact test, as appropriate. A P-value < 0.05 was considered statistically significant. All analyses were performed using SPSS® statistical software, version 18.0 (SPSS Inc., Chicago, IL, USA) for Windows®.

Results

A total of 100 patients were enrolled in the study; four patients were excluded because of incomplete data collection and two patients were excluded because of withdrawal of consent. Thirty patients were randomized to Group M and 32 patients each to Groups R and RM. Patient characteristics and operative data were similar in the three groups (Table 1).

Table 1.

Baseline characteristics of female patients who underwent thyroidectomy, given antiemetic therapy with either 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before induction of anaesthesia.

Characteristic	Group R n = 30	Group M n = 32	Group RM n = 32
Age, years	47.5 (38.5–56.5)	47.0 (39.3–47.8)	48.5 (37.3–54.0)
Height, cm	159.3 ± 8.4	160.9 ± 9.8	160.5 ± 8.2
Weight, kg	60.7 ± 8.5	60.1 ± 9.0	59.3 ± 9.2
Anaesthesia time, min	138.5 (130.0–156.3)	150 (126.3–173.8)	145.0 (130.8–164.8)

Data presented as mean ± SD or median (quartile 1–3).

Normally distributed data were analysed using analysis of variance and a post hoc Tukey test; non-normally distributed data were analysed using the Kruskal–Wallis test.

No statistically significant between-group differences (P ≥ 0.05).

Severity of nausea, as measured on the VNRS, was statistically significantly reduced at 0–2 h in Group RM compared with Groups R and M, and also at 2–6 h in Group RM compared with Group M (P < 0.05 for between-group differences) (Figure 1). The incidence of retching and vomiting was statistically significantly lower at 0–2 h, 2–6 h, 6–12 h in Group RM than in Group M, and was also lower in Group R than in Group M at 6–12 h (P < 0.05 for between-group differences) (Figure 2). The requirement for rescue antiemetic medication was statistically significantly lower only at 0–2 h in Group RM compared with Group M (P < 0.05 (Table 2).

Figure 1.

Severity of nausea, graded using an 11-point verbal numerical rating scale (VNRS), over 24 h postsurgery in female patients who underwent thyroidectomy. Patients received either 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before induction of anaesthesia. Values expressed as mean ± SE. *P < 0.05 Group RM compared with Group M; ^†P < 0.05 Group RM compared with Group R (analysis of variance).

Figure 2.

Incidence of retching and vomiting over 24 h postsurgery in female patients who underwent thyroidectomy. Patients received either 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before induction of anaesthesia. Values expressed as mean ± SE. *P < 0.05 compared with Group M (analysis of variance).

Table 2.

Requirements for rescue antiemetics over the first 24 h postoperation in female patients who underwent thyroidectomy and received either 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before induction of anaesthesia.

Time postsurgery	Group R n = 30	Group M n = 32	Group RM n = 32	Statistical significance a
0–2 h	7 (23.3)	13 (40.6)	4 (12.5)	P < 0.05*
2–6 h	4 (13.3)	8 (25.0)	3 (9.4)	NS
6–12 h	0 (0)	2 (6.3)	0 (0)	NS
12–24 h	0 (0)	0 (0)	0 (0)	NA

Data presented as n (%).

a

Categorical data were analysed using Fisher’s exact test.

*

Group RM compared with Group M.

NS, no statistically significant between-group differences (P ≥ 0.05); NA, not applicable.

There were no significant differences in the severity of postoperative pain between the groups (data not shown). Additionally, there were no significant between-group differences in cumulative fentanyl consumption from the IV–PCA device, or rescue analgesic requirement within the first 24 h postsurgery (Table 3).

Table 3.

Quantity of intravenous patient-controlled analgesia (IV–PCA), and requirement for rescue analgesia, over the first 24 h postoperation in female patients who underwent thyroidectomy, given either 0.3 mg ramosetron (Group R), 75 µg/kg midazolam (Group M) or 0.3 mg ramosetron combined with 75 µg/kg midazolam (Group RM) before induction of anaesthesia.

Time postsurgery	Group R n = 30	Group M n = 32	Group RM n = 32
0–2 h
PCA volume, ml	0.9 ± 1.1	1.0 ± 1.0	0.8 ± 1.0
Rescue analgesia	2 (6.7)	3 (9.4)	3 (9.4)
2–6 h
PCA volume, ml	6.1 ± 4.4	5.4 ± 3.2	6.5 ± 3.9
Rescue analgesia	0 (0)	0 (0)	1 (3.1)
6–12 h
PCA volume, ml	9.1 ± 5.8	8.1 ± 4.2	8.9 ± 5.3
Rescue analgesia	0 (0)	1 (3.1)	0 (0)
12–24 h
PCA volume, ml	12.5 ± 7.8	10. ± 9.2	13.5 ± 7.6
Rescue analgesia	0 (0)	0 (0)	0 (0)

Data presented as mean ± SD or n (%) patients.

Continuous data analysed using an analysis of variance; categorical data analysed using Fisher’s exact test.

No statistically significant between-group differences (P ≥ 0.05).

Discussion

In the present study, undertaken in female patients undergoing thyroidectomy, the combined administration of ramosetron and midazolam provided superior antiemetic efficacy than midazolam alone in terms of severity of nausea, incidence of retching and vomiting, and use of rescue antiemetics, especially in the early postoperative period in the PACU. However, our findings did not show any significant additional benefits compared with ramosetron alone, except for a reduction in the severity of nausea at 0–2 h.

To the best of our knowledge, there are no available reports evaluating the combination therapy of ramosetron and midazolam as a bolus for adult patients for the prophylaxis of PONV. In paediatric patients (aged 4–12 years), who underwent strabismus surgery, 6 µg/kg ramosetron or 6 µg/kg ramosetron and 0.1 mg/kg midazolam (administered prior to induction of anaesthesia) exhibited a similar incidence of nausea, retching and vomiting during the first and second 24-h periods, ¹⁴ which is consistent with findings of the present study.

Furthermore, studies of the combination therapy of ramosetron and midazolam as a continuous infusion have not been reported, but it has been documented that a continuous infusion of ondansetron and midazolam was more effective than ondansetron continuous-infusion monotherapy. ¹⁶ In this study, the incidence of PONV in the group receiving ondansetron and midazolam added to a IV–PCA device containing fentanyl was significantly lower than that observed in the group receiving ondansetron added to the IV–PCA at PACU, 24 h after recovery. ¹⁶ These differences between bolus and continuous infusion are probably due to the short elimination half-life (1.7–2.6 h) of midazolam. The present study results also show that the combination therapy of ramosetron and midazolam was effective in reducing the severity of nausea only in the early postoperative period. These drugs differ, however, in potency and efficacy with regard to the pharmacodynamic actions, and the antiemetic effect of midazolam last longer than the effects of sedation. ¹⁷ Thus, additional studies are needed to resolve the above differences.

Comparative studies of monotherapy with either ramosetron or midazolam for preventing PONV have not been reported, but publications have compared the antiemetic effects of ondansetron and midazolam. For example, treatment using ondansetron for antiemetic prophylaxis did not provide a superior benefit compared with midazolam in minor gynaecological or urological procedures, ¹⁸ and midazolam used in subhypnotic doses was as effective as ondansetron in patients undergoing abdominal or gynaecological surgery without untoward sedative or cardiovascular effects. ¹⁷ These findings^17,18 correspond to the results presented here, in that the incidence of nausea between ramosetron alone and midazolam alone was not significantly different. Ramosetron was, however, more effective in reducing the incidence of vomiting than midazolam, in the present study.

The chemoreceptor trigger zone and the emetic centre are associated with agonistic and antagonistic actions of various anaesthetic-related agents and stimuli. ¹ Head and neck surgeries (including thyroid surgery) are associated with high rates of nausea and vomiting; these symptoms are probably related to vagal stimulation during surgical manipulation and the substantial levels of inflammation and oedema that are found in neck tissues.^19,20 As well as causing discomfort to patients undergoing thyroid surgery, PONV increases the risk of other adverse consequences. For example, vomiting can increase the risk of postoperative bleeding, resulting in haematomas in the neck, and can also increase the risk of airway obstruction. ²¹

Ramosetron is a newly developed 5-HT₃ receptor antagonist with a more potent and longer receptor antagonising effect compared with previous-generation 5-HT₃ receptor antagonists. ²² It is a carbazalone derivative that is structurally related to serotonin and possesses specific serotonin receptor antagonist properties without altering dopamine, histamine, adrenergic or cholinergic receptor activity. Metabolism of inactive metabolites occurs predominantly in the liver and the elimination half-life of ramosetron is 4–5 h. ²³ In addition, the elimination half-life of ramosetron (9 h) is longer than that of ondansetron (3.5 h) or granisetron (4.9 h).^24,25 The dose of ramosetron used in the present study was based on previous findings, that a minimum of 0.3 mg of ramosetron is required to prevent PONV during the first 48 h after thyroidectomy.^26,27

The antiemetic effects of midazolam (which is used as a premedicant or as an auxiliary drug for the management of sedation, amnesia or situations with a requirement for an anxiolytic effect) [has been established. A midazolam dose of 50–75 µg/kg has been recommended for prophylactic antiemetic use; ¹⁰ another study reported that 75 µg/kg of midazolam injected into patients who had undergone thyroidectomy was as effective as 4 mg of ondansetron in preventing PONV, without creating any delay in the recovery time. ²⁸ Thus, it was decided that a dose of 75 µg/kg of midazolam should be used in the present study. The antiemetic mechanism of midazolam has not been fully elucidated, but postulated mechanisms include: i) glycine mimetic inhibitory effects; ii) enhancement of the inhibitory effects of gamma amino butyric acid; iii) augmentation of adenosinergic effects; iv) inhibition of dopamine release; v) augmentation of adenosine-mediated inhibition of dopamine in the chemoreceptor trigger zone. ¹⁰ Clinicians may hesitate to use hypnotic agents for antiemetic purposes because of the dangers of sedation. Doses used in the present study were, however, lower than those recommended for sedation, and the antiemetic effect of midazolam has already been shown to last longer than the effects of sedation. ¹⁷ The incidence of drowsiness was observed to be similar among the groups, and no significant difference with respect to recovery time was found between the different groups in this study.

Since midazolam has been found to act synergistically with fentanyl for the induction of anaesthesia, the dose of fentanyl can be reduced, compared with the level used for fentanyl monotherapy. ²⁹ Thus, this study investigated not only the effectiveness of midazolam for PONV but also its effectiveness for pain reduction after surgery. Our findings showed, however, show that VAS scores and the amount of infused fentanyl among the groups were not significantly different within the first 24 h after surgery.

There were some limitations to the present study. First, there was no control group because of anticipated ethical problems of including a control group who would not receive any antiemetic therapy postsurgery. Secondly, there are certain inconsistencies between the duration of action of ramosetron and the time-period of postoperative fentanyl use. The duration of action of a single bolus of ramosetron is ∼24 h, but IV–PCA was infused for 48 h in this study. Although a low incidence of PONV was expected >24 h postsurgery, patients remained under close observation during this period, and received antiemetic rescue medications immediately if they began vomiting or requested rescue medication >24 h postoperatively. Finally, the study population included patients with three or more risk factors for PONV, and could therefore be considered a biased, selective, population. Thus, our findings should be interpreted cautiously and cannot be applied to patients with a potentially lower risk of PONV without further investigation.

In conclusion, in female patients undergoing thyroidectomy, combination therapy involving ramosetron and midazolam provided superior antiemetic efficacy compared with midazolam monotherapy, in terms of reducing the incidence of nausea and vomiting, severity of nausea and use of rescue antiemetics. These findings were especially evident in the early postoperative period. No significant additional benefits of the combination therapy regimen were observed when it was compared with ramosetron single therapy, except for the reduction in the severity of nausea at 0–2 h.