Postoperative Nausea and Vomiting after Myringoplasty under Continuous Sedation Using Midazolam with or without Remifentanil

Postoperative nausea and vomiting (PONV), with an estimated incidence of 25-30%, are undesirable symptoms after general anesthesia and surgery.1-3 In particular, middle ear surgery is associated with a high incidence of PONV, accounting for 62-80% of patients who undergo middle ear surgery experience PONV.4-6 Many antiemetics, such as 5-HT₃ antagonists, dopamine receptor antagonists, and antihistamine drugs, have been studied for the prevention of PONV after middle ear surgery.

Midazolam, a rapid and short-acting benzodiazepine, has been widely used for the preoperative sedation or co-induction of anesthesia. The use of midazolam to prevent PONV is an off-label use of the drug. However, in several studies, midazolam has been reported to be effective for prophylaxis of PONV, when administered as a bolus before or after induction of anesthesia or as a continuous infusion postoperatively.7-10 Unlugenc, et al.10 reported that midazolam was effective for treating established PONV.

In a previous study, we indicated that midazolam-based continuous sedation for middle ear surgery had beneficial effects in overcoming pain and anxiety, and preventing PONV.11 However, that study evaluated only the development of nausea but not its severity, the incidence and frequency of vomiting, or the need for rescue drugs. Furthermore, as midazolam has no analgesic effect, some patients were sedated using remifentanil in addition to midazolam, although opioids increase the incidence of PONV.

The present study evaluated the effects of continuous sedation, by using midazolam, with or without remifentanil, on PONV in patients undergoing myringoplasty.

With approval from our institutional ethics committee and after obtaining written informed consent, 60 adult patients (ASA I or II) undergoing myringoplasty were included in this prospective randomized study. Exclusion criteria were age <20 years or >60 years, weight >100 kg, oral opening class IV Mallampati classification, history of chronic sedative use, known or suspected psychiatric disturbance, history of PONV or motion sickness, and any opioid or antiemetic medication taken within 24 hours before surgery. No premedication was used, and patients were informed before surgery that sedation would be performed for their myringoplasty. Patients were allocated to one of two groups randomized to receive remifentanil injection at loading and continuous doses of 0.5 µg kg^-1 and 0.07 µg kg^-1 min^-1, respectively, along with midazolam at loading and continuous doses of 0.04 mg kg^-1 and 0.04 mg kg^-1 h^-1, respectively, (group MR), or saline injection, instead of remifentanil, with the same doses of midazolam (group M).

PONV is one of the most common postoperative complaints and can occur after general, regional, and local anesthesia. The chemoreceptor trigger zone (CTZ) and the emetic center are associated with agonistic and antagonistic actions of various anesthetic-related agents and stimuli.12 Middle ear surgery is associated with a high risk for PONV, because the operation may stimulate the vestibular labyrinth, which is innervated by the vestibular portion of cranial nerve VIII (vestibulocochlear), which in turn activates the CTZ in the area postrema.12 CTZ manipulation leads to the activation of the parvocellular reticular formation, which is thought to be the emetic center, eventually resulting in emesis. Additionally, the sensory nerves to the auricle, external auditory meatus, tympanic membrane, middle ear cleft, and inner ear are provided by cranial nerves V (trigeminal), VII (facial), VIII (vestibulocochlear), IV (glossopharyngeal), and X (vagus). Stimulation of these parasympathetic nerves during surgical manipulations may induce PONV.13

Previous studies have shown that middle ear surgery is associated with an incidence of PONV as high as 62-80% in patients under general anesthesia.4-6 In cases of middle ear surgery under local anesthesia, Yung14 found that the most common discomforts reported were noise during surgery and anxiety, followed by dizziness and pain. In our previous study, we also showed high anxiety and pain scores in patients who underwent surgery under local anesthesia.11 Additionally, 42% of the patients suffered from PONV. However, the majority of patients who were sedated with midazolam overcame anxiety and pain satisfactorily. Furthermore, we found in the present study that continuous midazolam-based sedation was highly effective for reducing PONV after myringoplasty. Midazolam itself has an antiemetic effect, although the mechanism of the antiemetic action of midazolam has not been fully elucidated. Midazolam is thought to decrease dopamine input at the CTZ15 and decrease adenosine reuptake.16 This leads to an adenosine-mediated reduction in synthesis, release, and postsynaptic action of dopamine at the CTZ.15 Midazolam may also decrease dopaminergic neuronal activity and 5-HT₃ release by binding to gamma-aminobutyric acid receptors.17 As a result, midazolam-based continuous sedation can reduce PONV after myringoplasty.

In our previous study, we reported that the patients who were sedated using midazolam with remifentanil showed lower pain scores and higher satisfaction scores than those sedated using midazolam with saline.11 However, intraoperative use of opioids is a well-known risk factor for PONV.18, 19 Opioids stimulate the CTZ in the area postrema of the medulla, possibly through δ-receptors, thereby leading to nausea and vomiting.20 Thus, we expected that group MR would have a higher incidence of PONV than group M. However, the incidence and severity of nausea, incidence and frequency of vomiting, and need for rescue drugs were only slightly and insignificantly higher in group MR than in group M. The similar results between the groups may be attributable to an interaction between midazolam and remifentanil in the CTZ.

Another important observation in the present study was that patients felt extremely nauseous (average VAS score, 7.4) and the majority (86%) vomited eventually after PONV developed, despite the low rate of PONV (12%). Therefore, postoperative use of another antiemetic should be considered. Further studies are needed regarding the incidence and severity of nausea and incidence of vomiting after myringoplasty.

In conclusion, midazolam-based continuous sedation can reduce PONV after myringoplasty. Compared with patients receiving midazolam alone, patients receiving midazolam with remifentanil, which is a well-known risk factor for PONV, showed no difference in the incidence or severity of nausea, incidence or frequency of vomiting, or need for rescue drugs.