Dexmedetomidine vs. midazolam-ketamine for sedation during awake fiberoptic nasal intubation in patients with difficult airway - A randomized, double-blinded, comparative trial

Abstract

Background and Aims:

Both dexmedetomidine and midazolam-ketamine are known for their minimal respiratory depressant effects. While many studies have documented the use of dexmedetomidine in providing conscious sedation during awake fiberoptic-guided nasal intubation (AFNI), the use of midazolam-ketamine combination for this procedure has not been reported. The aim of this study was to compare the efficacy of dexmedetomidine with midazolam-ketamine combination for AFNI in patients with difficult airways undergoing oromaxillofacial surgery.

Material and Methods:

This study involved 60 patients undergoing oromaxillofacial surgery. They were randomized to receive either dexmedetomidine (1 μg/kg) (group D) or a combination of midazolam (0.02 mg/kg) and ketamine (0.5 mg/kg) (group MK) for sedation during awake fiberoptic nasotracheal intubation. Both groups received topical local anesthesia during the procedure using the spray-as-you-go technique. The primary outcome measured was cough score during intubation. Secondary outcomes included overall intubation, post-intubation, and sedation scores. Incidences of side effects such as hypoxemia and bradycardia were also measured.

Results:

The cough score (group D: 1.33 ± 0.61, group MK: 1.3 ± 0.53; P = 0.822), overall intubation score, and post-intubation score were comparable between the two groups. Patients receiving dexmedetomidine were significantly more sedated than patients receiving midazolam-ketamine. Patients receiving dexmedetomidine had significantly lower heart rates and blood pressure than patients receiving midazolam-ketamine. Four patients in group D had hypoxemia, while no patient in group MK developed hypoxemia.

Conclusions:

Both dexmedetomidine and midazolam-ketamine combinations were equally effective for sedation during AFNI in patients with difficult airways scheduled for orofacial maxillary surgery. Patients receiving dexmedetomidine were more sedated, but the incidence of side effects was similar in both groups.

Introduction

The challenge with awake endotracheal intubation is balancing effective sedation with maintaining a patent airway and preserved ventilation. Various drugs have been employed for this purpose, and many new drugs and combinations are added to it regularly. The ideal sedation regimen for awake fiberoptic bronchoscopy-guided endotracheal intubation (AFI) should ensure patient cooperation, comfort, hemodynamic stability, blunting of airway reflexes, and amnesia while maintaining spontaneous ventilation and patent airway. Use of drugs and/or drug combinations that provide sedation without suppressing respiratory mechanics and control is thus extremely desirable. Another desirable feature during the procedure is a patient who is cooperative and comfortable following instructions of the proceduralist during the airway manipulation maneuvers. Conscious sedation for AFI also requires careful titration of drugs to avoid overdose, which can lead to hypoventilation and airway obstruction. Conversely, insufficient sedation can cause discomfort, anxiety, and heightened sympathetic responses. Various drugs have been used for conscious sedation during AFI, such as alfentanil, midazolam,[1,2] ketamine,[3] propofol,[4] remifentanil,[4] and dexmedetomidine.[5] Previous studies have demonstrated that dexmedetomidine exhibits better sedation than propofol[6] and fentanyl[7] during AFI.

Dexmedetomidine offers conscious sedation without causing respiratory depression, but there are episodes of over-sedation, bradycardia, hypotension, and variable amnesia. Combination of midazolam and ketamine has been utilized for conscious sedation during several procedures, including orthopedic interventions,[8] procedural sedation in pediatric hematology,[9] laceration repair,[10] and burns-associated procedures.[11]

Both midazolam and ketamine have been used as the sole anesthetic for AFI with limited success.[1,2,3] Ketamine reduces neuronal activity, leading to sedation, analgesia, hypnosis, and dissociative anesthesia. However, it lacks amnesic and anxiolytic effects. Midazolam, as sole sedation agent, is inadequate for providing sedation for AFI without causing respiratory depression.[12,13] The combination of midazolam and ketamine would, thus, be an efficacious choice for AFI by providing all the necessary effects needed for awake airway manipulation. There is, however, no documented study in research databases where this combination has been used for sedation during AFI. This study, thus, aimed to address this lacuna in the literature by comparing the effectiveness of dexmedetomidine and midazolam-ketamine combination for providing conscious sedation during AFI. We hypothesized that the efficacy of combining midazolam-ketamine would show a difference in cough score to that of dexmedetomidine during awake fiberoptic-guided nasal intubation (AFNI). The study included patients with anticipated difficult airways undergoing oromaxillofacial surgery needing AFNI. The primary outcome was the difference in cough score at the time of introduction of the endotracheal tube through the glottis between the group. Secondary outcomes were overall intubation scores, post-intubation scores, patient satisfaction scores in the postoperative period, duration of fiberoptic intubation, and Ramsay sedation scores between the groups.

Material and Methods

This prospective, simple, randomized, comparative, double-blinded study was approved by the institutional ethics committee (ref: IECPG-455/29.11.2017). Written informed consent was obtained from all the patients participating in the study. The trial was registered before patient enrollment at Clinical Trial Registry India (CTRI) (registration: CTRI/2018/01/011141, date of registration: January 4, 2018). This research was conducted in accordance with the Declaration of Helsinki, 2013, and conducted in the Department of Anesthesiology at a tertiary hospital between March 2018 and September 2019. Patients with ASA physical statuses 1 or 2, between the ages of 18 and 65 years, posted for elective oromaxillofacial surgery with anticipated difficult airways were invited for participation in the study. Exclusion criteria were BMI >35, allergy to any of the study medications, history suggestive of gastro-esophageal reflex, pregnancy, hepatic or renal disorders, history of psychotropic drug intake or drug abuse, or long-term use of tricyclic antidepressants.

Sample size calculation

The sample size was calculated using the open epi online sample size calculator. We used the scoring system and data provided in a study done by Shen et al.,[14] in which authors evaluated the cough score using dexmedetomidine in doses similar to that used in our study. The authors of this study used a categorical scoring system for cough, with the values of “3” depicting slight cough and “2” depicting moderate cough [Table 1]. Patients receiving dexmedetomidine had a mean (SD) cough score value of 2.9 (0.4) in their study. We assumed a 10% difference in mean value, which would mean an absolute change in the value by 0.29, equivalent to a 30% change in the order level of the cough score, such that the values would remain within clinically acceptable limits (slight cough). The mean (SD) cough score in group KM was calculated to be 3.2 (0.4). We considered a confidence interval of 95%, power of 80%, and ratio of sample size (groups D and KM) as 1:1, with an alpha error of 5%; the sample size obtained was 30 participants in each group (N = 60).

Table 1.

Cough score[14]

Cough score	Inference
1	None
2	Slight
3	Moderate
4	Severe

To conceal allocation, a computer-generated random number table-based series of sealed, opaque envelopes with details of drug dilution made by a statistician were used to assign patients to group D or MK randomly. Patients in group D received dexmedetomidine in the dose of 1μg/kg, and patients in group MK received midazolam 0.02 mg/kg and ketamine 0.5 mg/kg for sedation.

Patients who fulfilled the inclusion criteria were explained about the study’s nature and AFNI during the pre-anesthetic checkup in simple words. Informed written consent was taken from the patients. All patients were premedicated with 0.25 mg of oral alprazolam on the morning of the day of surgery.

Patients were shifted to operation theatre on the day of surgery, and standard monitors (SpO₂, ECG, and NIBP) were attached. Intravenous (IV) access was established, and baseline parameters were recorded. An intravenous injection of glycopyrrolate 0.2 mg bolus was given immediately after securing the IV and given to both groups. Two drops of xylometazoline 0.1% were instilled into both nostrils. An anesthesiologist who had not participated in the study chose one of the sealed envelopes and prepared the drug mentioned inside that envelope. Both the patient and the anesthetist performing AFNI were blinded to the study group. Two syringes were prepared for each patient: syringe 1 (20 mL) and syringe 2 (2 mL). In group D, syringe-1 contained dexmedetomidine 1 μg/kg of patient-weight diluted in 20 mL of normal saline, and syringe-2 had 2 mL of normal saline. In group MK, syringe 1 contained midazolam 0.02 mg/kg of patient-weight diluted in 20 mL of NS, and syringe 2 contained 0.5 mg/kg of ketamine diluted in 2 mL of normal saline [Figure 1].

Figure 1.

Drug syringes and infusion pump used for blinding: Syringe 1 (20 mL) and syringe 2 (2 mL). In group D, syringe 1 contained dexmedetomidine 1 µg/kg (e.g. 50 kg patient 50 µg) diluted to 20 mL with normal saline, and syringe 2 had 2 mL of normal saline. In group MK, syringe 1 contained midazolam 0.02 mg/kg (e.g., 50 kg patient 1 mg) diluted to 20 mL of normal saline, and syringe 2 contained 0.5 mg/kg of ketamine (e.g. 50 kg patient 25 mg) diluted to 2 mL of normal saline so that both the group will have equal amount of volume and syringe 1 in both the group connected to infusion pump at rate of 120 mL/h so that whole 20 mL will be given in 10 minutes. And at the end of 10 minutes of infusion, syringe 2 will be given as bolus in both groups. An anesthesiologist who had not participated in the study chose one of the sealed envelopes and prepared the drug mentioned inside that envelope.

Infusion in syringe 1 was started at 120 mL/h for 10 minutes after baseline monitoring was instituted.

The anesthesiologist, unaware of the study drug, noted the vital parameters and Ramsay sedation scores (RSS)[15] at 5 and 10 minutes of infusion. Airway topicalization was started 5 minutes after the infusion of syringe 1 had begun. Initially, 2–3 sprays of 10% lignocaine were instilled into the nostril selected. This was followed by spraying on the posterior pharyngeal wall. At the end of the infusion of drug in syringe 1 (10 minutes), drug/normal saline in syringe 2 was injected.

All AFNIs were performed by anesthesiologists with an experience of a minimum of 10 successful AFIs in difficult airway patients. The procedure of AFNI was started only after the intravenous injection of solution in syringe 2. In either of the groups, if the patients had RSS < 2 even after syringe 2 administration, midazolam was given intravenously in aliquots of 0.5 mg until RSS achieved was ≥ 2.

The fiberoptic bronchoscope (FOB) of size 5.7-mm OD, loaded with Ring, Adair, and Elwyn (RAE) cuffed nasotracheal tube of 7–7.5-mm inner diameter for male patients and 6.5–7-mm inner diameter for female patients, was inserted into the selected nostril. Vitals and RSS were noted while inserting the FOB. After orientation and localization of the laryngo-epiglottic region, 2 mL of lidocaine 2% was sprayed on the supraglottic area with the help of a multi-orifice epidural catheter threaded through the working channel of the FOB. The epidural catheter was introduced into the glottis, and 2 mL of 2% lidocaine was sprayed on the vocal cords and trachea.

The epidural catheter was removed after the vocal cord and tracheal topicalization. FOB was introduced into the glottis till the carina was visualized. The well-lubricated endotracheal tube (ETT) was slid over FOB into the glottis, and its tip was positioned 2–3 cm above the carina. The FOB was then withdrawn, keeping the ETT in situ. The cough score [Table 1] and the overall intubation score [Table 2] were noted during insertion of the ETT into the glottis. Any episode of desaturation (SpO₂ <90%) was noted. The bronchoscope was withdrawn, and patients were oxygenated during all such episodes.

Table 2.

Overall intubation score

Parameter	Inference	Score
Grimacing on insertion of ETT through the nose	None	1
	Mild	2
	Moderate	3
	Severe	4
Vocal cord movement during introduction of bronchoscope	Open	1
	Moving	2
	Closing	3
	Closed	4
Limb movement during introduction of ETT into the trachea	None	1
	Mild	2
	Moderate	3
	Severe	4
Overall intubation score		3-12

ETT - Endotracheal tube

The cuff of the ETT was inflated with the appropriate amount of air. EtCO₂ curve trace was obtained on the monitor, and time was noted, which was the end of successful intubation. The total duration of the procedure was noted.

Post-intubation scores (1 - cooperative, 2 - restless/minimal resistance, 3 - severe resistance/GA required immediately) were noted after successful intubation. The anesthesiologist who noted all the peri-procedural data and was unaware of the group allocation also made postoperative visits the next day to know the patient satisfaction score (1 - excellent, 2 - good, 3 - acceptable, 4 - poor) during AFNI.

During nasotracheal intubation, hypertension was defined as MAP > 20% of the baseline level. Labetalol in boluses of 5 mg was used intravenously to treat hypertension. Hypotension was defined as MAP ≤ 20% of the baseline level, and it was treated with an intravenous bolus of ephedrine 6 mg. Tachycardia was defined as HR > 100 beats/min and treated with intravenous esmolol of 25 mg. Bradycardia was defined as HR < 50 beats/min and treated with intravenous atropine 0.6 mg.

Statistics

Data entry was done in Excel, and STATA software version 13.0 was used for analysis. Continuous variables were presented with mean and standard deviation for normally distributed data, and the median was used for non-normally distributed data. Student t-test and Wilcoxon rank-sum test were applied for comparisons. Chi-square and Fisher’s exact test assessed categorical variables. Intention-to-treat analysis was used, with statistical significance set at P < 0.05.

Results

In the final analysis, data from procedures done on 60 patients were included. Figure 2 illustrates the flow of patients. Demographic characteristics (age, height, weight, BMI, ASA-PS) were comparable between the two groups [Table 3]. The primary outcome, coughing during ETT insertion through the glottis, was not significantly different between groups [Table 4]. The secondary outcomes were similar in both groups [Table 4]. Hypoxic episodes occurred only in four patients in the dexmedetomidine group. There were no episodes of bradycardia and no need for atropine in either of the groups. There was statistically significant increased sedation during fiberoptic scope insertion through the nose in group D when compared with group MK (P = 0.038). Patients receiving dexmedetomidine were significantly more sedated than patients receiving midazolam-ketamine [Table 5]. In both groups, no extra-intravenous midazolam aliquots were required to achieve RSS ≥ 2. The heart rate and blood pressure of patients in the dexmedetomidine group were significantly lower as compared to those in the midazolam-ketamine group. None of the patients, however, needed pharmacological treatment for low heart rate or blood pressure. The total amount of lignocaine in 10% spray and 2% sprayed through the epidural catheter for anesthetizing the airway was comparable (D: 275.6 ± 41.5 mg vs. KM: 282.6 ± 64.2 mg)

Figure 2.

Patient recruitment as per consolidated standards of reporting trials (CONSORT)

Table 3.

Demographic data of patients included in the study

Parameter	Group D		Group MK
	Mean	SD	Mean	SD
Gender; M: F	27:3		27:3
Age in years	43.9	9.9	47.3	10.6
Weight in kg	62.2	10.9	58.9	13.2
Height in cm	164.7	8.09	161.4	7.3
BMI in kg/m2	22.9	3.7	22.4	3.8
ASA status - 1/2/3	17/13/0		12/18/0

Values are presented as mean±SD, number of patients. M: F - Male: Female, ASA - American Society of Anesthesiologists, BMI - Body Mass Index, D - Dexmedetomidine, MK - Midazolam-Ketamine

Table 4.

Comparison of outcomes between group D and group MK

Parameter	Group D	Group MK	P	Mean difference	CI
Cough score (primary outcome)	1.33±0.61	1.3±0.53	0.822	-	-
Overall intubation score 3-12	4.5±0.77	4.4±0.89	0.539	0.133	−0.299-0.566
Post-intubation score 1/2/3	28/2/0	30/0/0	0.492	-	-
Duration of intubation (min)	4.5±1.4	4.3±1.4	0.655	-	-
Postoperative satisfaction score 1/2/3/4	21/7/2/0	24/6/0/0	0.477	-	-

Values are presented as mean±SD, number of patients. Cough score: 1 - none, 2 - mild, 3 - moderate, 4 - severe, Post-intubation score: 1 - Cooperative, 2 - Restless/minimal resistance, 3 - severe resistance/GA required immediately. Postoperative satisfaction score: 1 - Excellent, 2 - Good, 3 - Acceptable, 4 - Poor. D - Dexmedetomidine, MK - Midazolam-Ketamine

Table 5.

Comparison of Ramsay sedation score between group D and group MK

RSS	Median (min-max)		P
	Group D	Group MK
5 min	2 (1, 3)	2 (1, 3)	0.999
10 min	2.2 (2, 4)	2.1 (2, 3)	0.159
FOB insertion	2.9 (2, 4)	2.6 (2, 3)	0.038*
ETT insertion in glottis	3.1 (1, 5)	3.0 (2, 5)	0.235
Before GA	3.6 (2, 5)	3.5 (3, 5)	0.338

Values are presented as median (1Q, 3Q). D - Dexmedetomidine, MK - Midazolam-Ketamine, *P<0.05 - Significant, FOB - Fiberoptic bronchoscopy insertion, ETT - Endotracheal tube insertion in the glottis, GA - General Anesthesia

Discussion

We compared dexmedetomidine 1 μg/kg infusion with a combination of midazolam 0.02 mg/kg and ketamine 0.5 mg/kg bolus for AFNI in patients with anticipated difficult airways. We found that the primary outcome, cough score during ETT introduction through vocal cords, was comparable between both groups. It showed against our study’s hypothesis that a combination of midazolam and ketamine would show a difference in cough score to that of dexmedetomidine during AFNI. However, both groups showed a clinically acceptable from a no to mild cough score only.

Many authors have described the advantages of dexmedetomidine in patients undergoing AFI.[1,6,16,17,18] The effectiveness of dexmedetomidine infusion for AFI in the same dose as was used in our study has been compared with fentanyl and propofol in previous studies.[6,7] Dexmedetomidine was found to be superior to both fentanyl (1 μg/kg) and target-controlled infusion of propofol. Some studies have also described the use of dexmedetomidine bolus followed by an infusion.[1,14,17]

Use of dexmedetomidine for AFI and AFNI has been compared with midazolam in various studies (there are at least 3 studies).[1,2,19] All these found dexmedetomidine in various doses to be more effective than midazolam in providing sedation and patient comfort during the procedure. The main disadvantage of using midazolam as a sole anesthetic agent for awake airway procedures is that in doses used for conscious sedation, it does not suppress airway reflexes adequately as it has no analgesic property. The proceduralists thus prefer to use it in combination with opioids.[19] Using midazolam with opioids, however, has the propensity to cause respiratory depression, which is undesirable in patients with difficult airways. The combination of midazolam and ketamine would thus be an ideal agent for this procedure. Ketamine will provide the necessary analgesic effect without suppressing respiration. The dose of ketamine used in our study (0.5 mg/kg) is unlikely to cause many side effects related to the use of ketamine, such as delirium and agitation. The increased risk of oral secretions can be easily taken care of by the use of glycopyrrolate, which is a norm in patients undergoing airway procedures.

The most common side effect of dexmedetomidine is hypotension and bradycardia. On the contrary, patients receiving midazolam-ketamine would understandably have a better hemodynamic profile.[8,9]

The combination of midazolam-ketamine is a popular technique for many minor procedures where sedo-algesia is desired. These include endoscopic procedures, minor gynecological procedures, incision of burn contractures, etc. Despite a widely documented safety margin, we could not find any reports of its use for AFIs.

The advantage of ketamine in AFI is that it preserves reflexes. It also does not cause any respiratory depression. Theoretically, dexmedetomidine also does not cause any respiratory depression. In our study, four patients receiving dexmedetomidine had hypoxia. The sample size was inadequate to calculate statistical significance for the incidence of side effects. Other authors have also reported hypoxia in patients undergoing AFI. Yousuf et al.[19] observed hypoxia in 5/30 patients in the dexmedetomidine group.

Similarly, another previous study showed hypoxic episodes in 4/30 patients in the dexmedetomidine group.[20] Shen et al.[14] also observed hypoxia in patients receiving dexmedetomidine. The suspected mechanism of hypoxia in the dexmedetomidine group is over-sedation and upper airway obstruction. Our study showed increased sedation levels in patients receiving dexmedetomidine during FOB insertion compared to the midazolam-ketamine group.

Despite multiple studies, the dose of dexmedetomidine for AFI is not standardized. Some authors have used 1.5 μg/kg[17,21] as a bolus; others have used 1 μg/kg.[6,7,14] Many authors have also described the follow-up infusion in the dose of 0.5–0.7 μg/kg/h.[1,14,17] Our study’s overall duration of AFI was 4.5 ± 1.4 min in the dexmedetomidine group and 4.3 ± 1.4 min in the midazolam-ketamine group. An infusion for 5 minutes is expected not to provide many benefits. Procedures expected to take longer, for example, due to procedural difficulty or operator inexperience, may, however, need continuation of an infusion of dexmedetomidine.

One advantage of using midazolam-ketamine could be the low cost of both these drugs compared with dexmedetomidine. As the intubating conditions and post-intubation satisfaction were equivalent in both groups, midazolam-ketamine may be a cheaper alternative to dexmedetomidine. Further studies, however, may be required to find the exact cost-effectiveness of both drugs.

Our study had many limitations. The small sample size for adverse effects interpretation. Single-center design. A single patient cohort for orofacial surgeries limits the generalization of our results. The use of entropy/BIS measurements for sedation assessment would have provided more objective data. The lack of a control group hindered the objective evaluation of the effects of local anesthesia in successful sedation. Anesthesiologists’ subjective assessments (Ramsay sedation score, cough score, and post-intubation score) could affect standardization.

In conclusion, patients with anticipated difficult airways undergoing AFNI for orofacial-maxillary surgery, dexmedetomidine, and the midazolam-ketamine combination showed similar outcomes for cough scores, overall intubation scores, and postoperative satisfaction. Patients receiving dexmedetomidine showed higher sedation levels and had a lower heart rate and blood pressure compared to patients receiving midazolam-ketamine.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.