Introduction

Awake fiberoptic nasotracheal intubation is used in patients with an anticipated difficult airway. Adequate sedation during the procedure is important to maintain a patent airway and minimize discomfort.

Many agents are used for sedation during awake fiberoptic intubation, including benzodiazepines, opioids, ketamine, propofol, and dexmedetomidine [1–4]. Remifentanil is of interest because it has analgesic and antitussive properties during awake intubation, while allowing communication with the patient to be maintained [5]. Dexmedetomidine is an attractive alternative as a sedative for managing a difficult airway because of its anxiolytic and analgesic properties [6–10]. It has been used to achieve conscious sedation in patients undergoing awake fiberoptic nasal intubation [11–13].

Previous studies have demonstrated that both dexmedetomidine and remifentanil are better than propofol target-controlled infusion (TCI) in conscious fiberoptic intubation [14–16]. However, to our knowledge, no previous studies have compared dexmedetomidine to remifentanil. The purpose of this study was to compare the effectiveness of dexmedetomidine and remifentanil in providing conscious sedation during awake fiberoptic nasotracheal intubation in patients undergoing oromaxillofacial surgery.

Methods

This study was approved by the Institutional Review Board of the Ethics Committee of Shanghai Ninth People’s Hospital affiliated with Shanghai Jiao Tong University School of Medicine, and written informed consent was obtained from each patient. It was also registered as a clinical trial (http://www.clinicaltrials.gov, identifier: NCT01474213).

Forty-two adult patients with American Society of Anesthesiologists (ASA) classifications of I–III were recruited for elective awake fiberoptic nasotracheal intubation following diagnosis of maxillofacial cancer or fracture with limited mouth opening. Exclusion criteria were pregnant or lactating women, heart rate (HR) <50 beats per min, systolic blood pressure <90 mmHg, atrioventricular block on ECG, high-dose opiate medication, use of an α2-adrenoceptor agonist or antagonist within the previous 14 days, liver cirrhosis, contraindications to nasal intubation, and refusal to be involved in the study. One patient declined consent, and one operation was cancelled. Forty patients were randomized 1:1 to receive dexmedetomidine or remifentanil using a covariate adaptive randomization algorithm. The randomization schedule was stratified based on both the ASA classification (I–II vs. III) and Mallampati classification (III vs. IV).

Two consultant anesthetists were present during all procedures. One was responsible for performing the awake nasal fiberoptic intubation and scoring sedation, endoscopy, intubation, and post-intubation conditions; the other administered the study drugs. An anesthetist nurse collected anesthetic data and postoperative records. The intubating anesthetist, the patients, and the nurse anesthetist were all blinded to the study drugs used.

In the operating room, intravenous access was established, and standard monitoring parameters were recorded every minute (electrocardiogram, non-invasive blood pressure, pulse oximetry, and respiratory rate). Nasal oxygen (3 l/min) was administered, and no premedication was given to patients. The study drugs were administered with a 50-ml syringe as 200 μg (2 ml) of dexmedetomidine in 48 ml of 0.9 % saline or 1 mg remifentanil in 50 ml of 0.9 % saline. Patients in the dexmedetomidine group received a loading dose (1.5 μg/kg) infused over 10 min followed by a continuous infusion of 0.7 μg/kg/h. Patients in the remifentanil group received remifentanil via an Orchestra Base Primea (Fresenius Vial) infusion system using a Minto pharmacokinetic model. The initial target was 3.0 ng/ml, and the TCI was adjusted by 0.5 ng/ml after the target concentration at the effect site had equilibrated with the plasma concentration until the desired level of sedation was achieved. During drug infusion, two drops of ephedrine hydrochloride and nitrofurazone (containing approximately 1 mg of ephedrine) were instilled into both nasal cavities, and the tongue and hypopharynx were sprayed with 7 % lidocaine (45 mg). The level of patient sedation was assessed using the Ramsay Sedation Scale (RSS; Table 1) once per minute. After 15 min, any patient with a RSS = 1 was given an infusion of rescue midazolam in 0.5-mg bolus doses until a RSS of 2 was achieved.

Table 1 Ramsay sedation scale
Full size table

Fiberoptic intubation was started once the RSS reached 2. A fiberoptic scope (Pentax FI-7BS, 6.0 mm; Japan) was loaded with a 7.0-mm tracheal tube for male patients or a 6.5-mm tube for females [17]. Two milliliters of 2 % lidocaine was sprayed onto the glottis and below the vocal cords via the channel of the fiberoptic scope.

Our primary outcome measures were endoscopy score, intubation score, and post-intubation scores, which were obtained using Rai’s standard scoring system [14]. Endoscopy and intubation were graded from 0 to 5 according to the positive data encountered during the procedure (Appendix 1), with lower scores indicating better conditions. Similarly, post-intubation was scored from 1 to 3, with higher scores indicating worse conditions. The times taken to achieve sedation (RSS = 2), endoscopy (from inserting the fiberoptic scope to visualizing the carina), and tracheal intubation (from inserting the tracheal tube into the nostril to ensuring intubation with capnography) were recorded.

Blood pressure, HR, and peripheral oxygen saturation (SpO2) were recorded before starting drug infusion and every minute until tracheal intubation. Atropine (0.5 mg) was injected if the HR was <40 beats/min, and 10 mg ephedrine was given when the blood pressure was <90 mmHg. Hemodynamic data (mean arterial pressure, HR, and SpO2) were compared between the two groups at three time points: baseline (during pre-anesthetic preparation), endoscopy (immediately after fiberoptic endoscopy), and intubation (immediately after intubation). Hypoxic episodes (SpO2 <90 %) and the use of ephedrine or atropine for hemodynamic support were also recorded.

A postoperative interview was conducted 24 h after the procedure to evaluate patients’ recall (memory of endoscopy and intubation), side effects (sore throat and hoarseness), and satisfaction score (1, excellent; 2, good; 3, acceptable; 4, poor).

The sample size calculation was based on a pilot study. Sixteen patients in each group were required to detect a difference of “1 over 5” in the endoscopy or intubation scores with a power of 0.8 and a type I error of 0.05. To compensate for dropped cases and deviation from normality, 42 patients were studied. After assessing normality, continuous data were compared using unpaired t tests, and the Mann-Whitney test was used to compare non-continuous data and non-normally distributed data. Chi-squared (χ2) tests or Fisher’s exact test were used to compare categorical data between the two groups. A p value of <0.05 was considered significant. SPSS 17.0 statistical software was used for all analyses.

Results

Forty patients completed this study. The demographic data were comparable between groups (Table 2). Data acquired during the fiberoptic procedures are shown in Tables 3 and 4. Different distributions of endoscopy, intubation, and post-intubation scores in the two groups are illustrated in Table 5.

Table 2 Patient characteristics
Full size table
Table 3 Measurements made during fiberoptic intubation
Full size table
Table 4 Adverse events and satisfaction data during awake intubation
Full size table
Table 5 Number of patients exhibiting different levels of endoscopy, intubation, and post-intubation scores in the two groups
Full size table

All patients had successful fiberoptic intubation at the first trial. No patients needed a rescue infusion of midazolam to achieve adequate sedation. The patients in the dexmedetomidine group had better endoscopy scores than those in the remifentanil group. During endoscopy insertion, equivalent numbers of patients in the two groups appeared to cough upon lidocaine treatment via scope, while fewer patients seemed to have limb movements in the dexmedetomidine group. There were no significant differences in intubation or post-intubation scores between the two groups. During tracheal intubation, no patients in either group exhibited localization of two limbs under the management of two sedatives. Fewer patients in the remifentanil group coughed on entering the trachea compared to those in the dexmedetomidine group (p = 0.018). Only one patient in the remifentanil group exhibited obvious discomfort after nasotracheal intubation. The time to achieve sedation was shorter in the remifentanil group than in the dexmedetomidine group.

As shown in Figs. 1, 2 and 3, the HR was less in the dexmedetomidine group at the end of endoscopy and intubation (p < 0.001 and p = 0.004, respectively). On the contrary, peripheral oxygen saturation at the end of endoscopy was less in the remifentanil group (p = 0.003). There were no differences in mean arterial pressure between the two groups.

Fig. 1
figure 1

Changes in mean arterial blood pressure (MAP) in patients receiving dexmedetomidine (closed circle) or remifentanil (closed triangle) during intubation. Hemodynamic parameters at three time points were analyzed: (1) baseline: pre-anesthetic preparation; (2) endoscopy: immediately after fiberoptic endoscopy; (3) intubation: immediately after tracheal intubation. There were no significant differences in MAP between the two groups at any time point

Full size image
Fig. 2
figure 2

Changes in heart rate (HR) in patients receiving dexmedetomidine (closed circle) or remifentanil (closed triangle) during intubation. Hemodynamic parameters at three time points were analyzed: (1) baseline: pre-anesthetic preparation; (2) endoscopy: immediately after fiberoptic endoscopy; (3) intubation: immediately after tracheal intubation. There were significant differences in HR at the end of endoscopy and intubation (p < 0.001 and p = 0.004, respectively) between the two groups. # p < 0.001; *p < 0.05

Full size image
Fig. 3
figure 3

Changes in peripheral oxygen saturation (SpO2) in patients receiving dexmedetomidine (closed circle) or remifentanil (closed triangle) during intubation. Hemodynamic parameters at three time points were analyzed: (1) baseline: pre-anesthetic preparation; (2) endoscopy: immediately after fiberoptic endoscopy; (3) intubation: immediately after tracheal intubation. There were significant differences in SpO2 at the end of endoscopy (p = 0.003) between the two groups. *p < 0.05

Full size image

Two patients in the remifentanil group exhibited desaturation; their SpO2 values decreased to 85 and 88 %, respectively, and their respiratory rates decreased to seven and eight times per min, respectively. Loud auditory stimuli and high-flow nasal oxygen (5 l/min) were required to resolve transient hypoxia. Two patients in the dexmedetomidine group who developed bradycardia received temporary support with atropine (Table 4).

Post-anesthetic interview parameters are shown in Table 4. Recall of intubation was higher in the remifentanil group (65 %) than in the dexmedetomidine group (30 %) (p = 0.027). Patient satisfaction (median [interquartile range]) was greater in the dexmedetomidine group than in the remifentanil group (2 [1–2] and 2 [2–3], respectively, p = 0.022). Although there was no statistically significant difference for recall of endoscopy between the two groups (70 vs. 40 % with p = 0.057), the result appears to be clinically worthwhile. Other post-anesthetic parameters did not differ between the two groups.

Discussion

This study showed that both dexmedetomidine and remifentanil were effective in patients undergoing conscious fiberoptic nasotracheal intubation. Although endoscopy was better tolerated with dexmedetomidine, the intubation scores were similar in both groups. The ability of both drugs to achieve the desired level of sedation and the absence of differences in sedation scores in both groups during the procedure may suggest that the difference in endoscopy scores reflects inherent properties of dexmedetomidine other than sedation.

Dexmedetomidine induces conscious sedation by activating the endogenous sleep-promoting pathway. Compared with midazolam, patients sedated with dexmedetomidine are aroused more easily when stimulated; in addition, they experience little respiratory depression [18, 19]. Xerostomia has been reported as an effect of dexmedetomidine. Therefore, this drug may provide a better field for fiberoptic endoscopy [20, 21]. Moreover, the analgesic effects may play a role in improving patient comfort during fiberoptic intubation, giving better endoscopy scores. On the other hand, the antitussive action of remifentanil lowered the incidence of coughing during tracheal intubation. This property may have made intubation scores comparable between the two groups, despite higher endoscopy scores in the remifentanil group. During awake intubation, it is crucial that the patient is relaxed and cooperative. Most patients in both groups could obey commands immediately after nasotracheal intubation at the doses used in our study.

Some trials have investigated the dose of dexmedetomidine required for conscious sedation. Bergese [10] found that a loading dose of 1 μg/kg followed by a 0.7 μg/kg/h maintenance infusion of dexmedetomidine achieved adequate sedation in patients undergoing awake fiberoptic intubation with or without small supplemental doses of midazolam. Chu [11] reported that a loading dose of dexmedetomidine (1 μg/kg) provided conscious sedation with spontaneous ventilation for fiberoptic intubation. In our pilot study, a loading dose of dexmedetomidine (1.5 μg/kg) infused over 10 min followed by a continuous infusion of 0.7 μg/kg/h enabled a smooth intubation procedure with no predisposition to airway obstruction or respiratory depression.

Use of remifentanil has been reported to be useful for awake intubation with or without the application of local anesthetic to the larynx and trachea [22–25]. Few studies have reported the effective target concentration of remifentanil required for sedation by TCI in awake intubation. Rai and colleagues [14] found that an effect-site concentration of remifentanil of 3.2 (2.8–3.5) ng/ml achieved adequate sedation for awake fiberoptic intubation. In their study, the median endoscopy score was better than ours with a similar concentration of remifentanil [1 (0–3) vs. 3 (1–3)]. This discrepancy may be due to the use of midazolam as a premedication in their patients and the application of more local anesthetic during fiberoptic insertion. Moreover, Vennila and colleagues [26] reported that intubation was achieved with a mean (standard deviation) effect-site concentration of 8.06 (3.52) ng/ml during tracheal intubation in the absence of local anesthesia and benzodiazepine. In our study, the mean (range) remifentanil TCI dose was 3.9 (3.5–4.1) ng/ml, which provided adequate comfort relief of pain and anxiety with smaller doses of local anesthetics.

Dexmedetomidine has been reported to prevent hemodynamic responses by decreasing noradrenaline release and centrally mediated sympathetic tone [27]. However, it may cause side effects including hypotension, bradycardia, atrial arrhythmia, and hypoxia [18]. In our study, two patients developed bradycardia, which was easily managed with atropine. In contrast, previous studies indicate that doses of remifentanil as low as 0.25 μg/kg/min can lead to significant dose-dependent respiratory depression. In this study, two patients experienced transient hypoxia. Although loud auditory stimuli and high-flow nasal oxygen (5 l/min) rapidly resolved the situation, caution is needed when using remifentanil in patients with anticipated mask ventilation.

Without midazolam premedication, greater recall of intubation in the remifentanil group may indicate that remifentanil has less of an amnesic effect than dexmedetomidine at the same level of sedation. The pharmacokinetic properties of remifentanil and the mode of administration in the study may be the reasons for faster achievement of the desired level of sedation. Amnesia and lower endoscopy scores, on the other hand, may explain the greater patient satisfaction with dexmedetomidine sedation.

Conclusion

Both dexmedetomidine and remifentanil were effective in patients undergoing awake fiberoptic nasotracheal intubation. Compared with remifentanil, dexmedetomidine offers better endoscopy scores, lower recall of intubation, and greater patient satisfaction, with minor hemodynamic side effects.