Postoperative nausea and vomiting (PONV) is one of the most common complications reported in patients following surgery and anesthesia, with an incidence of 30-80% [19]. PONV may decrease the quality of recovery and cause aspiration, intracranial hypertension, dehydration, acid-base disturbance, electrolyte imbalance, and neurological deterioration [5-7,18], which leads to a prolonged hospital stay and increased medical costs [12]. Therefore, PONV prophylaxis has become a critical issue for the enhancement of the quality of recovery and surgical outcomes.

The pathophysiology of PONV involves stimulation of various afferent pathways and emetic reflex activation [22]. Afferent pathways associated with PONV include the chemoreceptor trigger zone (CTZ), vagal mucosal pathway of the gastrointestinal system, neuronal pathway of the vestibular system, reflex afferent pathway of the cerebral cortex, or midbrain afferents [22]. Several pharmacological interventions, including anticholinergics, antihistamines, corticosteroids, dopamine receptor antagonists, and serotonin receptor antagonists, for treating and preventing PONV have been assessed [2,3].

Microvascular decompression (MVD), a treatment of choice for hemifacial spasm (HFS) or trigeminal neuralgia (TN) is associated with an increased risk of PONV, which occurs in >70% patients who undergo MVD [13]. HFS or TN is usually caused by the abnormal compression of a cranial nerve [16], and surgical procedures require a close approach to neural structures by the CTZ or vestibular systems to relieve irritated nerves [17]. Thus, surgical manipulation may predispose patients to PONV after MVD [21]. Several studies have been conducted to prevent PONV in patients undergoing MVD [23,25].

Ramosetron, a selective serotonin receptor antagonist, can effectively reduce the incidence of PONV after various types of surgeries, including neurosurgery [11,20]. However, whether ramosetron administration reduces the incidence of PONV after MVD remains unclear. We hypothesized that prophylactic use of ramosetron reduces the incidence of PONV after MVD. Therefore, this study aimed to determine the efficacy of prophylactic ramosetron administration in reducing the incidence and severity of PONV in patients undergoing MVD and to elucidate its safety.

The protocol of this prospective, double-blind, single-center, randomized controlled trial was approved by the Institutional Review Board of Seoul National University Bundang Hospital (protocol code B-2003/600-005 and date of approval April 7, 2020) and was registered within the UMIN Clinical Trials Registry (UMIN 000040178). Written informed consent was obtained from all eligible patients before surgery. The study was conducted according to the guidelines of the Declaration of Helsinki. Adult patients (aged >19 years) with an American Society of Anesthesiologists physical status I/II scheduled to undergo MVD under general anesthesia were enrolled in this trial. We excluded the following patients : those with a body mass index of <18.5 or >35.0 kg/m2; those with a history of craniotomy or anticancer chemotherapy, severe renal or hepatic dysfunction, or QTc prolongation; and those who used preoperative antiemetics within the 24-hour period before surgery, used opioids for >2 weeks, or were pregnant.

Patients were randomly allocated to one of two groups (control or ramosetron group) in a 1 : 1 ratio using computer-generated random sequences with a block size of 4 (random allocation software, version 2.0; Isfahan University of Medical Sciences, Isfahan, Iran). Each group was treated at the end of surgery, as follows : the control group, 2 mL saline; and the ramosetron group, 0.3 mg ramosetron (Nasea^®, Daiichi-Sankyo Korea, Seoul, Korea). Identical 2-mL syringes containing either saline or 0.3-mg ramosetron were prepared and administered intravenously during dura closure. Random sequences were sealed in an opaque envelope and opened by an anesthesiologist who prepared the drug but was not further involved with the study. All patients, neurosurgeons, anesthesiologists, and outcome assessors were blinded to the group assignments.

A 0.02 mg/kg midazolam bolus was administered intravenously pre-surgery in the reception area. In the operating room, patients were monitored using electrocardiography, noninvasive blood pressure measurement, pulse oximetry, and bispectral index assessment (Medtronic, Minneapolis, MN, USA). After preoxygenation, general anesthesia was induced using propofol (4 mcg/mL) and remifentanil (3 ng/mL) with a target-controlled infusion pump (Fresenius Vial, Brezins, France). After loss of consciousness, 0.6 mg/kg rocuronium was administered as a neuromuscular blockade; subsequently, tracheal intubation was performed. Plain endotracheal tubes (Covidien, Mansfield, MA, USA) with inner diameters of 7.5 and 7.0 mm were used for male and female patients, respectively. Patients were ventilated with an inspired oxygen fraction of 0.5, a tidal volume of 6-8 mL/kg of ideal body weight, and a positive end-expiratory pressure of 5 cmH2O. The respiratory rate of each patient was controlled to maintain an end-tidal carbon dioxide level of 30-35 mmHg. Anesthesia was maintained at a bispectral index of 40-60 during surgery by adjusting the target concentration of either propofol (3.5-5.0 mcg/mL) or remifentanil (2.0-4.0 ng/mL). No additional rocuronium was administered for intraoperative neurophysiological monitoring. Hypotension, defined in this study as a mean blood pressure <60 mmHg, was treated with either 5 mg ephedrine or 20 mcg phenylephrine, as appropriate. At the end of surgery, 50 mcg/kg neostigmine and 10 mcg/kg glycopyrrolate were used to reverse effects of residual neuromuscular blockade, and a patient-controlled analgesia (PCA) device delivering 180 mg ketorolac (total volume, 100 mL) was connected intravenously. The PCA device (AutoMed 3200; Ace Medical, Seoul, Korea) was programmed to deliver a bolus dose of 2 mL, with a continuous 2 mL/h infusion rate, and a 15-minute lockout time. After patients could breathe continuously, they were extubated and transferred to the post-anesthesia care unit (PACU). Patients were discharged from the PACU when they had a modified Aldrete score ≥9 [24].

One neurosurgeon performed all MVDs, as described previously [9]. Briefly, with the patient in the supine position, the head was rotated approximately 20-30° away from the affected side without the use of head fixation. Brain retractors were not used during the surgical procedure. A 4-5-cm curvilinear skin incision, three quarters below the mastoid notch in patients with HFS and half above the mastoid notch in those with TN, was performed along the hairline. Following the identification of the digastric groove, a 2-2.5-cm craniectomy was performed below the digastric groove for HFS and above the digastric groove for TN. An incision of the dura mater was made along the inferoposterior margin of the sigmoid sinus. Subsequently, the arachnoid membrane/trabeculations along the involved cranial nerves were dissected carefully. The offending vessel(s) were decompressed from the cranial nerves after exploration of the entire intracranial portion of each of these involved nerves, including the root exit/entry zone and cisternal segment. A watertight dural closure was subsequently achieved. Finally, the deep and superficial muscles and the skin were approximated.

After surgery, a blinded outcome assessor evaluated patient outcomes at the following predefined time points : 1 hour, 24 hours, and 48 hours postoperatively. The primary outcome was the incidence of PONV throughout the 48-hour postoperative period. Based on a previous study [15], nausea was defined as an unpleasant feeling due to which the patient had the urge to vomit, and vomiting was defined as forceful expulsion of gastric contents from the mouth.

Secondary outcomes were PONV severity throughout the 48-hour postoperative period, rescue antiemetic requirements during the 48-hour postoperative period, PONV satisfaction score at postoperative day 2, and duration of hospital stay. The severity of PONV was evaluated using a 4-point scale, which was described in a previous study, as follows [23] : 0, no symptoms; 1, few mild symptoms but no requirement for treatment; 2, moderate symptoms that require treatment; and 3, severe, persistent symptoms after treatment. Intravenous metoclopramide (10 mg) was administered as a rescue antiemetic for patients who wanted treatment or experienced more than one episode of vomiting. On postoperative day 2, all patients were asked to provide satisfaction scores using an 11-point numerical rating scale, as follows : 0, very dissatisfied; 10, very satisfied. Adverse events such as constipation, or hiccups were also recorded.

A total of 142 patients were considered for enrollment in the study from May 2020 to March 2021, and 42 were excluded (Fig. 1). Therefore, 100 patients were randomly placed in each study group, and data obtained from 97 were included in final analyses. The intervention was discontinued in one patient of the ramosetron group owing to a PCA (use of fentanyl instead of ketorolac), and two (one from each group) were lost to follow-up due to postoperative cognitive dysfunction or postoperative delirium (Fig. 1). Table 1 includes characteristics of patients, surgery, and anesthesia; no significant differences between groups were noted.

The incidence of PONV was significantly lower in the ramosetron group than in the control group (p=0.028) during the study period (Table 2). Effects of ramosetron on PONV reduction were significant on postoperative day 1 (1-24 hours) (p=0.040, Table 2). Additionally, the severity of PONV assessed using a 4-point scale was determined to be lower in the ramosetron group than in the control group (p=0.041, Table 2). The number of patients who received a rescue antiemetic drug was lower in the ramosetron group than in the control group, but the difference was not statistically significant (p=0.068, Table 2). Patient satisfaction scores in both groups were comparable (p=0.827, Table 2).

Duration of hospital stay did not significantly differ between the groups (p=0.731, Table 3). There was no significant difference between groups in the incidence of adverse events (Table 3). Six patients in the ramosetron group and two in the control group reported constipation (p=0.159). In the ramosetron group, two patients complained of hiccups (p=0.242). All symptoms were resolved by the day of discharge.

Although the control group has more female patients compared to the ramosetron group (57% vs. 50%), multivariate analysis revealed that the use of ramosetron could independently prevent the development of PONV (p=0.039, Table 4).

This is the first randomized clinical trial to assess the efficacy and safety of ramosetron, a 5HT3 antagonist, against PONV in patients undergoing MVD. Findings of the study suggest that prophylactic administration of ramosetron reduces the incidence and severity of PONV after MVD without causing any adverse events. However, the administration of ramosetron at the end of surgery did not affect rescue antiemetic use, or duration of hospital stay.

It is worth noting that ramosetron reduced both the incidence and severity of PONV after neurosurgery. In addition to causing postoperative pain, the occurrence of PONV is highly concerning after surgeries performed under anesthesia. Especially in patients undergoing neurosurgery, PONV may increase arterial and intracranial pressure, which can be confused with neurologic complications. The etiology of PONV is associated with patient traits, the anesthetic used, and surgical factors [8], although MVD has not been determined to be a surgical risk factor for PONV. However, many studies have suggested that MVD is a type of neurosurgery associated with a high risk of PONV. This is likely because the cranial nerve dissected during MVD is close to the vestibular nerve and nucleus of the brainstem [13,17].

The incidence of PONV after MVD has been reported to be approximately 70%, and in this study, its incidence rate in the control and ramosetron groups was approximately 51% and 29%, respectively. The relatively low incidence of PONV in the control group may be explained by the fact that the PCA regimen used in this study included non-steroidal anti-inflammatory drugs (ketorolac) instead of opioids. A previous univariate analysis that included PONV patients revealed that the use of a ketorolac-based PCA was not significantly associated with PONV [26]. Prophylactic antiemetic treatment with ramosetron decreased the incidence of PONV by >20% relative to that by control treatment. Few studies on the effects of other 5HT3 antagonists on PONV in patients who underwent MVD have been conducted thus far [10,23]. Thongrong et al. [23] conducted a prospective, randomized, controlled trial to assess the use of ondansetron with dexamethasone in patients who underwent MVD and found that intraoperative administration of the drug combination decreased the incidence of PONV 24 hours post-surgery by 15% versus control treatment (66.7% vs. 81.5%, respectively). Another retrospective observational study that assessed palonosetron use revealed that decreases in the incidence of PONV were higher when a combination of prophylactic palonosetron and sugammadex (reverses neuromuscular blockade) was administered versus control treatment (incidence, 19.3% vs. 37.2%, respectively) [10]. The incidence of PONV in this study was significantly lower than that previously reported in patients following MVD. Underreporting due to the retrospective nature of the study is possible. Lee et al. [13] studied the effect of transdermal scopolamine on PONV after MVD and found that treatment decreased the severity of PONV and rescue antiemetics use throughout the 48-hour postoperative period without affecting PONV incidence, which was high (approximately 70%).

Although PONV may prolong the duration of the hospital stay [12], the hospital stay duration did not significantly differ between the two groups assessed in this study. A possible explanation for this may be that only three patients (two in the control group and one in the ramosetron group) had severe and refractory PONV symptoms on postoperative day 2.

This study has some limitations. First, we did not record headaches, drowsiness, and dizziness, which are common adverse events related to ramosetron treatment [14,20]. However, it is difficult to distinguish between the adverse events associated with ramosetron and post-neurosurgical complications. Second, the severity of PONV was assessed using a 4-point scale, although an 11-point numerical rating scale might have provided more detailed information regarding the severity of PONV in the individuals included in this study. Third, the prophylactic effect of ramosetron seems to be significant up to postoperative day 1 (24 hours after surgery) although ramosetron is known to be effective up to 48 hours after surgery [4]. This waning effect of ramosetron after one day may explain the non-superiority in satisfaction scores of the ramosetron group. Additionally, it was difficult to validate the effect of ramosetron between the two groups after 24 hours due to low incidence of PONV during this period.

In conclusion, prophylactic administration of ramosetron reduced the incidence and severity of PONV in patients with MVD without causing serious adverse events. Thus, ramosetron administration may improve the quality of recovery in patients undergoing MVD. Further studies using a multimodal antiemetic approach are needed in patients with MVD to assess the enhancement in their quality of recovery.