Efficacy and safety of remimazolam combined with remifentanil for sedation during awake fiberoptic intubation: a randomized controlled trial

Abstract

Introduction

There has been not study determining if remimazolam combined with remifentanil is the reasonable dosing regimen for sedation during awake fiberoptic intubation (AFOI). This prospective double-blind randomized controlled trial compare efficacy and safety of sedation using different-dose remimazolam combined with remifentanil for AFOI procedure.

Method

One hundred and fifty patients were randomly assigned to five groups receiving different interventions. The Rf group received only remifentanil infusion of 0.05 mcg/kg/min, the Rm group received only remimazolam 0.6 mg/kg/h, and the RR0.2, RR0.4 and RR0.6 groups received remifentanil 0.05 mcg/kg/min combined with remimazolam of 0.2, 0.4 and 0.6 mg/kg/h, respectively. After intravenous infusion of studied drugs for 3 min, AFOI was carried out under airway topical anesthesia. The primary outcome was the incidence of deep sedation during AFOI procedure.

Results

Deep sedation occurred only in the Rm, RR0.4 and RR0.6 groups, with incidences of 3.3%, 17.2%, and 50.0%, respectively (p < 0.001). The incidence of deep sedation in the RR0.6 group was not statistically different from that in the RR0.4 group (p > 0.05), but was higher than these in the Rf, Rm and RR0.2 groups (p < 0.05). Hypoxemia was observed only in RR0.2, RR0.4 and RR0.6 groups, with incidences of 13.3%, 44.8%, and 44.8%, respectively (p < 0.001). The recall score for AFOI procedure was significantly lower in the four remimazolam groups than in the Rf group (p < 0.005). The patients’ reaction score to AFOI procedure, cough severity, incidence of tachycardia, and willingness to repeat the procedure were lower in the RR0.6 group than in the Rf and Rm groups (p < 0.005).

Conclusions

Remimazolam 0.2–0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min are effective and feasible dosing regimens of sedation for AFOI procedure in patients with normal airway, but the regimen including remimazolam 0.2 mg/kg/h should be the better choice for balancing efficacy and safety.

Trial registration

Chinese Clinical Trial Registry, ChiCTR2100042917. Retrieved from http://www.chictr.org.cn/showproj.html?proj=65332 on January 31, 2021.

1. Introduction

Awake fiberoptic intubation (AFOI) is the recognized standard technique for managing difficult airways [1]. However, the success of AFOI highly depends on operator proficiency, effective airway topical anesthesia, and sedation management [2]. The ideal dosing regimens of sedation for AFOI procedure should be able to keep the patient quiet, comfortable, cooperative, and maintain a patent airway with spontaneous respiration [3,4]. Many agents have been used to facilitate AFOI, such as benzodiazepines, propofol, and opioids [5–7]. Nonetheless, it is usually difficult to achieve all of these requirements using a single drug [5,6]. Propofol is not advisable for facilitating AFOI in the patients with a predicted difficult airway, due to increased risks of oversedation, hypoventilation and airway obstruction, particularly when combined with opioids [8]. The Difficult Airway Society recommends that remifentanil is a suitable choice for sedation during AFOI, and midazolam can be co-administered when the combining sedative drugs are delivered and titrated by a second anesthetist [8]. However, oversedation and respiratory depression caused by the combination of midazolam and opioids remain common concerns [3,8]. There is currently no evidence suggesting that single-drug sedation protocol is superior to the others in terms of the overall rate of successful intubation and the incidence of arterial oxygen desaturation [7].

Remimazolam is a new benzodiazepine drug with good sedative efficacy and safety [9]. It can be rapidly metabolized into inactive metabolites by ubiquitous tissue esterases and may be completely reversed by specific antagonist flumazenil [10]. It is thought to have a negligible impact on airway tone, potentially reducing the risk of airway obstruction during sedation management [7]. A recent meta-analysis of remimazolam for procedural sedation demonstrates that compared with midazolam, remimazolam improves the success rate of procedure, reduces the need for rescue medication, and alleviates the recall of endoscopy [11]. As a single agent used for sedation of AFOI procedure, remimazolam is suggested at a maximum load dose of 0.1 mg/kg followed by continuous infusion of 0.5–0.75 mg/kg/h [12]. Recently, both a randomized controlled trial [13] and a retrospective analysis [14] have assessed the efficacy and safety of sedation using remimazolam for AFOI and have demonstrated that remimazolam compared with dexmedetomidine can provide a shorter intubation time, a higher incidence of anterograde amnesia and more stable hemodynamics when sufentanil is administered with a target-controlled infusion or a bolus dose. However, remimazolam 0.093 mg/kg over 10 min combined with a target-controlled infusion of sufentanil 0.2 ng/mL results in a deeper sedation level with the Modified Observer’s Assessment of Alertness/Sedation Scale (MOAA/S) score of less than 2 in more than a quarter of patients, and a high incidence of hypoventilation (23.3%) [13], which may contribute to the effects of long action duration and potential respiratory inhibition of sufentanil [15]. Remifentanil, an ultra-short-acting opioid with a plasma half-life of 10 min, is metabolized by non-specific esterases in blood and tissues and is easy to titrate. It avoids accumulation effects and allows a rapid recovery [15,16]. It can also be reversed with naloxone in the occurrence of overdose event. Remifentanil has been used as a single agent to provide moderate sedation for AFOI, but it is associated with a high postoperative recall of the procedure owing to the lack of anterograde amnesia [6]. Theoretically, as an alternative sedation regimen for AFOI, the combination of remimazolam and remifentanil, has the potential advantages of rapid onset and recovery compared with midazolam or dexmedetomidine, and of lower risks of oversedation and airway collapse compared with propofol. However, there has been not study determining if remimazolam combined with remifentanil is a reasonable dosing regimen of sedation for AFOI procedure. This prospective, double-blind, randomized controlled trial aimed to compare the sedative effect, intubation conditions, intubation recall, and adverse events of different-dose remimazolam combined with a low fixed dose of remifentanil for sedation of AFOI, providing a reference base for further researches about the optimal sedatives for AFOI.

2. Methods

2.1. Trial design

The protocols of this clinical trial were approved by the Ethics Committee of Beijing Friendship Hospital (No.2020-P2-292-02, Date: 21/01/2021) and registered with the Chinese Clinical Trial Registry (ChiCTR2100042917, Date: 31/01/2021). The trial was carried out from August 2021 to February 2023 in the operating room of Beijing Friendship Hospital according to the Declaration of Helsinki. This study was reported based on the CONSORT guidelines (Supplementary material 1-CONSORT) [17].

2.2. Patients

Patients who were age of 18–60 years and American Society of Anesthesiologists (ASA) physical status classifications 1–2, underwent elective surgery under general anesthesia and consented to receive the AFOI were recruited. All patients received preoperative assessment according to our routine practice. Exclusion criteria included age < 18 years or >60 years, pregnant or lactating women, body mass index (BMI) >30 kg/m², acute respiratory tract diseases, known or anticipated difficult airways, severe cardiopulmonary dysfunction, systolic blood pressure (SBP) <90 mmHg, history of drug or alcohol abuse, allergy to the studied drugs, neuropsychiatric disorders, inability to communicate effectively, and patient refusal.

The included patients were informed with the details of the research protocols before operation and were given the option to opt out if desired. Before the study, the written informed consents were acquired from all patients or their legally authorized representatives.

2.3. Randomization and blinding

According to a 1:1 allocation ratio, the included patients were randomly assigned to one of five groups using a computer-generated randomization schedule: Rf, Rm, RR0.2, RR0.4, and RR0.6. The Rf group was only given an intravenous infusion of remifentanil 0.05 mcg/kg/min (Yichang Renfu Pharmaceuticals Co., Ltd., China, lot number: 30A02111). The Rm group was only given an intravenous infusion of remimazolam 0.6 mg/kg/min (Yichang Renfu Pharmaceuticals Co., Ltd., lot number: 10T0511). Other than an intravenous infusion of remifentanil 0.05 mcg/kg/min, the RR0.2, RR0.4 and RR0.6 groups also received an intravenous infusion of remimazolam 0.2, 0.4 and 0.6 mg/kg/h, respectively. Two experienced anesthesiologists (CWL and FSX) were responsible for AFOI procedure, and two independent anesthesiologists (XRZ and YC) performed observation and data collection during AFOI procedure. Using 50-ml syringes, the studied drugs were prepared as 10, 15, and 30 mg of remimazolam in 25 ml of 0.9% normal saline and 0.75 mg of remifentanil in 25 ml of 0.9% normal saline. According to the group assignment and patient weight in kilograms, both remimazolam and remifentanil were administered by anesthesiologists who did not participate in the trial implementation and data collection. The anesthesiologists were responsible for trial implementation, and patients were blinded to group assignment.

2.4. Interventions

All patients fasted for 8 h preoperatively. Following patient entered into the operating room, standard monitoring, such as noninvasive blood pressure, electrocardiography, pulse oxygen saturation (SpO₂) and bispectral index (BIS), was established. After intravenous access was established, scopolamine 0.3 mg was intravenously administered as an anti-salivary drug, and intravenous infusion of the studied drugs was initiated via pressure-driven syringe pumps at the same rates, that is, remimazolam 0.5 ml/kg/h and remifentanil 0.1 ml/kg/h. During the study, lactated Ringer’s fluid was intravenously administered at a rate of 5–10 ml/kg/h.

After intravenous infusion of the studied drugs for 3 min, airway topical anesthesia was initiated using 2% lidocaine via the working channel of a fiberscope (EB-220R, Zhejiang UE Medical Corp., China). A modified ‘spray-as-you-go’ technique was used for airway topical anesthesia, as described in our previous work [3,18]. That is, 2 ml of lidocaine at 30-sec interval was sprayed into the airway until patient allowed to advance the fiberscope to the next airway structure. During topical anesthesia of the airway, oxygen was delivered at a rate of 5 L/min via a nasal tube. Once adequate airway topical anesthesia was achieved, a fiberscope was inserted into the reinforced tracheal tube with a size of 6.5-mm for female or a size of 7.0-mm for male. According to the techniques previously described [3,18], the patient’ jaw was lifted by an assistant, and AFOI was performed. After the fiberscope was passed between the vocal cords into the mid-trachea, the tracheal tube was advanced downwards along the fiberscope until the tube’ tip achieved 3–4 cm above the carina. During the intubation procedure, additional airway topical anesthesia was allowed with a single lidocaine spray through the fiberscope for patient comfort, as needed. After the trachea was successfully intubated, the fiberscope was withdrawn from the airway, the tube cuff was slowly inflated, and the tracheal tube was connected to the ventilation system of anesthesia machine. Then anesthesia was induced using propofol 2 mg/kg and rocuronium 0.6 mg/kg, and maintained with continuous intravenous infusion of propofol and remifentanil, which were adjusted to keep the BIS values between 40 and 60. If needed, additional doses of rocuronium or cisatracurium were administered intravenously to maintain satisfactory muscle relaxation during the surgery.

2.5. Outcome assessment

The independent anesthesiologist observed sedation level, BIS values, hemodynamic variables, and SpO₂ at 1 min (T1) and 3 min (T2) after infusion of the studied drugs, immediately before the fiberscope was introduced into the glottis (T3), and immediately after intubation (T4). The sedation level was determined by using the classic 6-point MOAA/S score (Supplementary scoring criteria) [19,20]. A MOAA/S score of 2 or less was defined as deep sedation, according to previous literature [21].

Hypotension was defined if systolic blood pressure (SBP) was less than 80 mmHg or diastolic blood pressure (DBP) was less than 50 mmHg, and was treated with intravenous phenylephrine (40 mcg) or ephedrine (6 mg). Hypertension was defined if SBP was more than 180 mmHg or DBP was more than 100 mmHg, and treated with intravenous urapidil 10–15 mg. Bradycardia was defined if heart rate (HR) was less than 50 beats/min and was treated with intravenous atropine 0.2–0.4 mg. Tachycardia was defined if HR was more than 120 beats/min and was treated with intravenous esmolol 25–50 mg [3,22]. Hypoxemia was defined as SpO₂<92%, which reflects clinically significant oxygenation impairment [23]. When the MOAA/S score was less than 3, with the conditions that severe oxygen desaturation (SpO₂ < 75% at any time or < 90% for > 60 s [24]) occurred following the conventional interventions including verbal and tactile stimuli, jaw thrust and nasal oxygen supplement, bronchoscope was removed from the airway and face mask ventilation was carried out as required, intravenous infusion of the studied drugs was stopped and flumazenil 0.2 mg or naloxone (50 mcg) was intravenously administered. If patients received intravenous naloxone or flumazenil, they were excluded from the final analysis.

Intubation conditions were evaluated during AFOI procedure using the variables described in previous studies [3,18, 25,26], including patient’ reaction to fiberscopy and tracheal intubation assessed using a 5-point scale, cough severity evaluated by a 4-point scale, and patient’s tolerance to the tracheal tube assessed using a 3-point scale. The details of these scoring systems used for assessments of intubation conditions were provided in supplementary scoring criteria.

The dose of lidocaine for airway topical anesthesia, intubation time (from insertion of fiberscope into the mouth to confirmation of successful intubation with capnography), the number of attempts for successful intubation (withdrawal of the fiberscope from the airway after an unsuccessful intubation indicating the requirement of an additional attempt), and the partial pressure of end-tidal carbon dioxide (P_ETCO₂) that appeared firstly after successful intubation were also recorded.

At 24 h after surgery, two independent anesthesiologists (JM and KS) who were blind to group assignment conducted the follow-up and evaluated patient’s recall of AFOI procedure using a 3-point scale [3,27]: 1, none; 2, partial; and 3, full. The patient’s willingness to repeat the procedure was assessed using a 3-point scale [28]: 1, accepted; 2, likely accepted; and 3, refusal.

2.6. Sample size estimation

The primary outcome of this study was the incidence of deep sedation during AFOI procedure. This study included a five-arm comparison and all pairwise comparisons were of interest. The sample size of this study was evaluated on the basis of the between-group difference in primary outcome. In our pilot trial, deep sedation occurred in 0 of 10 patients in the RR0.2 group and in 4 of 10 patients in the RR0.6 group. Thus, at least 23 patients per group were required to demonstrate a 25% difference in the incidence of deep sedation between the RR0.2 and RR0.4 groups, and between the RR0.4 and RR0.6 groups, with a power of 0.8 and a significance level of 0.005 with Bonferroni correction for multiple comparisons. Considering possible dropouts and other variables of interest, 30 patients in each group were included.

2.7. Statistical analysis

All data analysis were completed with the SPSS (Version 26.0, IBM SPSS Inc., Chicago, IL, USA). Continuous data were assessed for the normality of distribution using a Shapiro-Wilk test and the homogeneity of variance using a Levene median test. If continuous data were normally distributed and had homogeneous variance, they were present as mean ± standard deviation (SD) and their inter-group comparisons were performed with repeated analysis of variance (ANOVA), with a Bonferroni correction for multiple comparisons. The Kruskal-Wallis H test was applied for the between-group comparisons of non-normally distributed continuous or ordinal data. The Pearson’s chi-squared or Fisher’s exact tests were applied for the between-group comparisons of categorical data. Accounting for multiple comparisons, p < 0.005 (α = 0.05/10) was considered statistically significant for pair-wise post-hoc analysis with Bonferroni correction. Statistical significance was set at a level of p < 0.05.

3. Results

3.1. Patients’ characteristics

One hundred and eighty-eight patients were assessed for eligibility from August 16, 2021, to February 10, 2023, in this study. Of these eligible patients, 38 were excluded from the study because they participated in the pilot trial (n = 20), did not meet the inclusion criteria (n = 3), or refused to participate (n = 15). The remaining 150 patients were randomized into five groups with 30 patients in each group. After randomization, two patients were withdrawn from the study because the AFOI was not completed (one in the RR0.4 group due to severe nausea caused by laryngeal reflex and one in the RR0.6 group due to severe cough). No patient was excluded from the study because of missing data. Finally, 148 patients were included in the data analysis (Figure 1). There were no statistically significant differences among the five groups with respect to sex, age, height, weight, BMI, ASA physical status classifications, Mallampati grade, and comorbidities (p > 0.05, Table 1).

Figure 1.

The CONSORT flow diagram of the study population. Rf group, patients receiving alone remifentanil 0.05 mcg/kg/min; Rm group, patients receiving alone remimazolam 0.6 mg/kg/h; RR0.2, RR0.4 and RR0.6 groups, patients receiving remimazolam 0.2, 0.4 and 0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min, respectively.

Table 1.

Characteristics of patients.

Variables	Rf n = 30	Rm n = 30	RR0.2 n = 30	RR0.4 n = 29	RR0.6 n = 29	Statistical values	p values
Sex: male	16(53.3)	14(46.7)	13(43.3)	17(58.6)	10(34.5)	4.035*	0.401
Age, years	43.4 ± 11.3	43.7 ± 11.1	45.9 ± 10.4	45.2 ± 11.4	45.4 ± 10.0	0.324†	0.862
Weight, kg	70.0 ± 10.8	71.4 ± 13.2	67.3 ± 13.3	70.0 ± 12.3	67.6 ± 14.4	0.566†	0.688
Height, cm	166.3 ± 8.5	167.1 ± 7.8	167.5 ± 8.2	167.9 ± 7.9	166.9 ± 7.4	0.184†	0.946
BMI, kg/m2	25.3 ± 3.0	25.4 ± 3.2	23.8 ± 3.2	24.8 ± 3.4	24.1 ± 3.9	1.304†	0.271
ASA physical status, I/II	3(10.0)/27(90.0)	2(6.7)/28(93.3)	8(26.7)/22(73.3)	5(17.2)/24(82.8)	6(20.7)/23(79.3)	5.728*	0.220
Mallampati grade	1(1–1[1–2])	1(1–1[1–2])	1(1–1[1–1])	1(1–1[1–2])	1(1–1[1–3])	6.375‡	0.173
Comorbidities
Hypertension	10(33.3)	4(13.3)	7(23.3)	8(27.6)	6(20.7)	3.718*	0.446
Diabetes mellitus	0(0)	0(0)	2(6.7)	2(6.9)	4(13.8)	6.648*	0.087

Data are n (%), mean ± SD or median (IQR [range]). BMI, Body mass index; ASA, American Society of Anesthesiologists.

^*χ² values.

^†F values.

^‡H values.

3.2. Primary outcome analysis

As shown in Figure 2, during AFOI procedure, deep sedation occurred only in the Rm, RR0.4 and RR0.6 groups, with incidences of 3.3%, 17.2%, and 50.0%, respectively. The incidences of deep sedation were significantly different among five groups (p < 0.001). The incidence of deep sedation during AFOI procedure was not statistically different among the Rf, Rm, RR0.2 and RR0.4 groups (p > 0.05), but was significantly higher in the RR0.6 group than in the Rf, Rm and RR0.2 groups (p < 0.05). The incidence of deep sedation was comparable between the RR0.4 and RR0.6 groups (p > 0.05).

Figure 2.

The incidence of deep sedation during AFOI procedure in each group. Rf group, patients receiving alone remifentanil 0.05 mcg/kg/min; Rm group, patients receiving alone remimazolam 0.6 mg/kg/h; RR0.2, RR0.4 and RR0.6 groups, patients receiving remimazolam 0.2, 0.4 and 0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min, respectively. AFOI, awake fiberoptic intubation.

3.3. Sedation levels and BIS values during AFOI procedure

The BIS values and MOAA/S scores during AFOI procedure are shown in the Table 2. The MOAA/S scores and BIS values at T1 were not statistically different among the five groups (p > 0.05). The MOAA/S scores at T2 and T3 were significantly lower in the RR0.6 group than in the other four groups (p < 0.005). The MOAA/S scores at T3 and T4 were significantly reduced in the Rm and RR0.4 groups compared with the Rf group (p < 0.005), but they were not statistically different among the Rm, RR0.2 and RR0.4 groups (p > 0.005). The MOAA/S score at T4 in the RR0.6 group was similar to that in the RR0.4 group (p > 0.005), but it was significantly lower than those in the Rm and RR0.2 groups (p < 0.005).

Table 2.

MOAA/S scores and BIS values during AFOI procedure.

	Rf n = 30	Rm n = 30	RR0.2 n = 30	RR0.4 n = 29	RR0.6 n = 29	Statistical values	p values
MOAA/S: 0 (no response) to 5 (alert)
T1	5(5–5[5–5])	5(5–5[5–5])	5(5–5[5–5])	5(5–5[5–5])	5(5–5[5–5])	0.000‡	1.000
T2	5(5–5[5–5])	5(5–5[4–5])	5(5–5[4–5])	5(5–5[4–5])	4(4–4[3–5])★*#§	89.801‡	<0.001
T3	5(5–5[5–5])	4(4–5[3–5])★	4.5(4–5[4–5])	4(4–5[2–5])★	3(3–3[2–4])★*#§	85.342‡	<0.001
T4	5(4–5[4–5])	3.5(3–4[2–4])★	4(4–4[3–4])	3(3–4[2–4])★	3(2–3[2–3])★*#	92.101‡	<0.001
BIS
T1	97(96–99[93–99])	97(96–98[95–99])	97(96–98[92–99])	97(96–98[92–99])	97(97–98[94–99])	3.428‡	0.517
T2	96.5(95–98[87–99])	96(95–97[93–98])	96(94–97[84–99])	96(94–97[82–98])	95(93–96[79–98])★	15.332‡	0.004
T3	96.5(95–98[88–99])	95.5(95–96[94–98])	96(94–97[89–98])	92(87.5–96[77–98])★	89(85–92.5[72–96])★*#	56.063‡	<0.001
T4	96.5(95–98[90–99])	94.5(93–96[87–98])	96(95–97[88–98])	93(86.5–96[81–98])★	84(82–88[73–91])★*#§	75.711‡	<0.001

Data are medians (IQR [range]). MOAA/S: modified Observers’ Assessment of Alertness/Sedation; BIS: Bispectral index.

^‡ H values.

T1, 1 min after infusion of studied drugs; T2, 3 min after infusion of studied drugs; T3, immediately before fiberscope introduced into the glottis; and T4, immediately after intubation.

^★p < 0.005; Compared with Rf group.

^* p < 0.005; Compared with Rm group.

^# p < 0.005; Compared with RR0.2 group.

^§ p < 0.005; Compared with RR0.4 group.

The BIS values at T2 were significantly lower in the RR0.6 group than in the other four groups (p < 0.005), but it was not statistically different among the other four groups (p > 0.005). The BIS values at T3 and T4 were significantly lower in the RR0.4 and RR0.6 groups compared with the Rf group, and were significantly lower in the RR0.6 group compared with the Rm group (p < 0.005). Additionally, the BIS value at T4 was significantly decreased in the RR0.6 group than those in the RR0.2 and RR0.4 groups (p < 0.005). However, the BIS values at all four time points were not statistically different among the Rm, RR0.2 and RR0.4 groups (p > 0.005).

3.4. Outcomes related to AFOI procedure

The patient’ reaction score to AFOI procedure was not statistically different among the Rf, Rm, RR0.2 and RR0.4 groups (p > 0.005), but it was significantly lower in the RR0.6 group than in the Rf and Rm groups (p < 0.005). Cough severity was not statistically different among the Rf, Rm, RR0.2 and RR0.4 groups (p > 0.005), but it was significantly reduced in the R0.6 group compared with the Rf and Rm groups (p < 0.005). There were no statistically significant differences in the tolerance score to tracheal tube among the Rf, Rm, RR0.2 and RR0.4 groups (p > 0.005). Compared with the Rf group, the tolerance score to tracheal tube was significantly decreased in the RR0.6 group (p < 0.005).

The recall score for AFOI procedure was not significantly different among the four remimazolam groups (p > 0.005), but it was significantly lower in the four remimazolam groups than in the Rf group (p < 0.005). The willingness score of patient to repeat an AFOI was not statistically different among the Rf, Rm, RR0.4 and RR0.6 groups (p > 0.005), but it was significantly lower in the RR0.6 group than in the Rf, Rm and RR0.2 groups (p < 0.005). The five groups were comparable in terms of lidocaine consumption and number of attempts for successful intubation (p > 0.05). There were no statistically significant differences in the intubation time among the Rf, Rm and RR0.4 groups (p > 0.005). The intubation time was shorter in the RR0.2 and RR0.6 groups than in the Rf, Rm and RR0.4 groups (p < 0.005), but the difference was not statistically significant between the RR0.2 and RR0.6 groups (p > 0.005, Table 3).

Table 3.

Data related to AFOI procedure.

Variables	Rf n = 30	Rm n = 30	RR0.2 n = 30	RR0.4 n = 29	RR0.6 n = 29	Statistical values	p values
Reaction score to AFOI	2(2–3[1–3])	2(2–3[1–3])	2(2–2[1–3])	2(2–3[1–3])	2(1–2[1–3])★*	22.205‡	<0.001
Cough severity	3(2–3[2–4])	3(2–3[2–4)	2(2–3[1–3])	2(2–3[1–3])	2(2–2[1–3])★*	27.192‡	<0.001
Tolerance to tracheal tube	2(2–2[1–3])	2(2–2[1–3])	2(1–2[1–3])	2(1–2[1–3])	1(1–2[1–3])★	24.398‡	<0.001
Intubation time, min	7.5 ± 1.0	7.8 ± 1.0	6.1 ± 0.8★*	7.0 ± 1.5*#	6.1 ± 1.0★*§	15.326†	<0.001
Numbers of attempts	2(2–3[1–5])	2(2–3[1–5])	2(1–3[1–4])	2(2–3[1–5])	2(1–3[1–4])	0.749‡	0.945
Lidocaine dosage, mg/kg	3.6 ± 0.8	3.7 ± 0.8	3.7 ± 0.7	3.8 ± 0.9	3.8 ± 0.9	0.272†	0.895
Recall score to AFOI	3(3–3[2–3])	2(2–2[1–2])★	2(2–2[1–2])★	2(2–2[1–2])★	1(1–2[1–2])★	96.520‡	<0.001
Willingness to repeat AFOI	2(1–2[1–3])	2(1–2[1–3])	2(2–2[1–3])	2(1–2[1–3])	1(1–1[1–2])★*#	29.404‡	<0.001

Data are mean ± SD or medians (IQR [range]). AFOI, Awake fiberoptic intubation.

^† F values.

^‡ H values.

^★p < 0.005; Compared with Rf group.

^* p < 0.005; Compared with Rm group.

^# p < 0.005; Compared with RR0.2 group.

^§ p < 0.005; Compared with RR0.4 group.

3.5. Hemodynamic and respiratory variables during the AFOI

MAP at all-time points and HR at T1 and T2 were not statistically different among the five groups (p > 0.05). The HR at T3 was significantly decreased in the R0.2 and R0.6 groups compared with the Rf group (p < 0.005), but it was not statistically different among the four remimazolam groups (p > 0.005). The HR at T4 was significantly lower in the RR0.6 group than in the Rf and Rm groups (p < 0.005), but it was not statistically different among the RR0.2, RR0.4 and RR0.6 groups (p > 0.005). The SpO2 values at T1, T2, and T3 were not statistically different among the five groups (p > 0.005). The SpO2 value at T4 was significantly higher in the R0.2 group compared in the R0.6 group (p > 0.005). The first PETCO2 value was significantly higher in the RR0.4 and RR0.6 groups than in the Rf and Rm groups (p < 0.005). Furthermore, the first PETCO2 value was significantly increased in the RR0.6 group compared with the RR0.2 group (p < 0.005, Supplementary Table 1).

3.6. The occurrence of adverse events

Hypoxemia was observed only in RR0.2, RR0.4 and RR0.6 groups, with incidences of 13.3%, 44.8%, and 44.8%, respectively. The occurrence of deep sedation was significantly different among the five groups (p < 0.001). The incidence of hypoxemia was significantly higher in the R0.4 and R0.6 groups than in the other three groups (p < 0.05). The incidence of tachycardia was significantly decreased in the R0.6 group compared with the Rf and Rm groups (p < 0.05), but it was not statistically different among the Rf, Rm, R0.2 and 0.4 groups (p > 0.05). The five groups were comparable in terms of the incidence of hypertension, hypotension, and bradycardia (p > 0.05, Supplementary Table 2).

4. Discussion

The main aims of this clinical trial were to evaluate the efficacy and safety of different-dose remimazolam combined with remifentanil 0.05 mcg/kg/min for moderate sedation during AFOI procedure by comparing with alone use of remimazolam and remifentanil. The results showed that 3 studied doses of remimazolam (0.2, 0.4 and 0.6 mg/kg/h) combined with remifentanil provided comparable intubating conditions, as indicated by the reaction score to AFOI procedure, cough severity score and post-intubation tolerance. However, the occurrence of deep sedation increased with incremental doses of remimazolam, even with an incidence of deep sedation of up to 50% in the patients receiving remimazolam 0.6 mg/kg/h combined with remifentanil. On the whole, remimazolam 0.2 mg/kg/h combined with remifentanil for moderate sedation during AFOI procedure provided the benefit of less recall compared with alone remifentanil, and had significant merits of no deep sedation and less hypoxemia when compared to remimazolam 0.4 or 0.6 mg/kg/h combined with remifentanil. In addition, remimazolam 0.2 mg/kg/h combined with remifentanil and alone remimazolam 0.6 mg/kg/h produced comparable intubating conditions with no deep sedation and similar incidence of hypoxemia.

For safety purposes, it is generally believed that sedatives and analgesics used for AFOI should be short-acting and easily titratable, with minimal spontaneous breathing depression and postoperative recall [29]. When remifentanil as a single agent is used for moderate sedation for AFOI, the dose regimens reported in available literatures vary, with loading doses of 0–0.75 mcg/kg over 0.5–5 min and continuous infusion rates of 0.05–0.15 mcg/kg/min [6,30–33]. Importantly, similar to other opioid µ receptor agonists, remifentanil can not only produce the dose-dependent analgesia and sedation but may also result in dose-related respiratory depression [15,16]. Thus, in our study, only a low continuous infusion rate of remifentanil 0.05 mcg/kg/min without a loading dose was selected to reduce the risk of respiratory depression during AFOI procedure. Remimazolam is a ultra-short-acting benzodiazepine agent, with the features of quick onset, rapid recovery, anterograde amnesia and few adverse effects, such as hypotension and respiratory depression [10,34–36]. Thus, we propose that the application of remifentanil combined with remimazolam for moderate sedation during AFOI procedure may provide potential advantages, such as easy titration, minimal adverse events, and postoperative recall.

In the available literature, remimazolam is usually administered as a bolus dose, followed by a supplemental or rescue dose, if necessary, to induce sedation or anesthesia. Furthermore, the efficacy of remimazolam in moderate sedation was assessed [19,20, 37,38]. In the male patients with age of 65–80 years, Zhao et al. [37] showed that a 95% effective dose of remimazolam for targeting the MOAA/S score of ≤ 3 and above 1 was 0.079 mg/kg, and a MOAA/S score of 4 could be achieved within 7 min after the administration of single-dose remimazolam. In the healthy volunteers, Antonik et al. [20] demonstrated that a single dose of remimazolam 0.05 mg/kg could only result in a light sedation level with a MOAA/S score of 4 and a BIS value of 75, while a single-dose remimazolam (0.075 mg/kg or more) could provide a deeper sedation level with a mean BIS value of 60–70 and a MOAA/S score of ≤ 2 immediately after administration. In the patients who were age of 18–85 years and underwent regional block under procedure sedation with a MOAA/S score of 4 or less, Li et al. [38] found that with sufentanil 0.08 mcg/kg for analgesia, remimazolam 0.08 mg/kg produced the best sedation efficacy in young patients who were age of less than 65 years, but remimazolam 0.04 mg/kg with the trend of fewer respiratory adverse events was more optimal for elderly patients aged ≥65 years. Pastis et al. [19] reported that remimazolam 5 mg (with 2.5 mg top-up doses) combined with fentanyl 25–75 mcg (with 25 mcg top-up doses) could effectively provide moderate sedation level with a MOAA/S score of 3 for bronchoscopy, without rescue sedative requirement in 84.2% of patients. Given that the dose ranges of remimazolam from 0.05 to 0.08 mg/kg were applied for moderate sedation in above previous studies, a maximal intravenous infusion rate of remimazolam 0.6 mg/kg/h was selected in our study. Our results showed that the interval from staring intravenous infusion of remimazolam to the completion of AFOI procedure was approximately 10 min and the cumulative dosage of remimazolam during this interval was approximately 0.1 mg/kg, similar to the dose of 0.093 mg/kg over 10 min used for AFOI by Chen et al. [13]. However, our pilot trial showed that the regimen of remimazolam 0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min tended to lead to deep sedation, which occurred in 4 out of 10 patients. To evaluate the effect of remimazolam dose on sedation depth, this study used the gradient doses of remimazolam at 0.2, 0.4, and 0.6 mg/kg/h. The design of the dose gradients is rational to a certain extent and is supported by published literatures about sedation during AFOI procedure [13,14], in which remimazolam is administered as a bolus dose of 0.05 mg/kg, or a infusion dose of 0.073 and 0.093 mg/kg over 10 min, equivalent to the dose of 0.2–0.6 mg/kg/h as observed in this study.

The MOAA/S score is the scoring system most frequently used for assessment of sedation level in clinical studies and practice. As AFOI requires satisfactory cooperation of the patient, particularly when he/she has a known difficult airway, a moderate sedation level with the MOAA/S scores of 3–4 is often considered as desirable. This is different from the sedation managements for clinical bronchoscopy reported in previous works [39–41], in which a deep sedation with the MOAA/S score of 1–2 is often required to promote the endoscopic procedure. Cooperation loss, airway loss and respiratory depression are all potential sequelaes of deep sedation [8], which are best avoided during AFOI, particularly in managing a predicted or known difficult airway. Our results demonstrated that remimazolam 0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min produced an adequate moderate sedation with the MOAA/S scores of 2–4 for AFOI procedure in all patients. However, when using this sedation regimen, approximately 50% of patients achieved a deep sedation with the MOAA/S score of 2, and 44.8% of patients experienced hypoxemia during AFOI procedure. Evidently, this sedation scheme is not desirable for managing difficult airways. In non-intubated older patients aged ≥ 70 years, Deng et al. [42] demonstrated that when remimazolam 0.075 mg/kg as a loading dose following by a continuous infusion of 0.12 mg/kg/h was used to provide sedation during surgery, the incidences of oversedation and hypoxemia (SpO₂< 90%) were 27% and 13.5%, respectively. When remimazolam is used for intraoperative sedation, thus, the authors suggest that intravenous infusion rate should be titrated from 0.1 mg/kg/h. Nonetheless, in our patients receiving alone remimazolam 0.6 mg/kg/h, the incidence of deep sedation only was 3.3% and hypoxemia did not occur in any patient. Zhou et al. [43] confirm that there is a synergistic sedation interaction between remimazolam and remifentanil. A recent study also demonstrated that fentanyl 1 mcg/kg reduced the effective dose of remimazolam-induced sedation by 50% [44]. In the present study, high incidences of deep sedation and hypoxemia in the patients who received remimazolam 0.4 and 0.6 mg/kg/h combined with remifentanil may contribute to the application of combined two drugs.

The combinations of sedatives and opioid agents are often recommended for sedation management of AFOI [6,29], but respiratory depression and hypoxemia are always the most concerning adverse events [3,8]. In the patients aged 18–65 years who underwent nerve block under procedure sedation with remimazolam 0.08 mg/kg and sufentanil 0.08 mcg/kg, Li et al. [38] reported that the incidence of hypoxemia (SpO₂ < 90%) was 30%. Our results showed that the incidence of hypoxemia (SpO₂ < 92%) was approximately 45%, with a high median value of first P_ETCO₂ (43–44 mmHg) after AFOI in the patients using the regimens of remimazolam 0.4 and 0.6 mg/kg/h combined with remifentanil. In contrast, incidence of hypoxemia only was 13.3% with a low median value of first P_ETCO₂ (38 mmHg) following the AFOI in the patients receiving remimazolam 0.2 mg/kg/h combined with remifentanil 0.05 mcg/kg/min. These results indicate that remimazolam combined with a fixed dose of remifentanil can result in breathing inhibition and hypoxemia, which increases with incremental doses of remimazolam. Fortunately, none of patients in the five groups of this study suffered from severe hypoxemia. Oxygen desaturation in patients with hypoxemia was also transient and could be effectively rectified by increasing oxygen supplement, engaging in verbal and tactile stimuli to awaken the patient, and using jaw thrust to keep airway open during AFOI. However, it must be emphasized that severe hypoxemia is hazardous in the event of airway loss and failed intubation. To avoid the condition, it is important to keep the patient awake and cooperative and to avoid sedatives that lower airway tone.

BIS monitoring is widely used to quantify the level of sedation or general anesthesia. In the patients with age of 18–60 years and difficult airways, Gnaneswaran et al. [45] determined the optimal BIS level of conscious or moderate sedation for the AFOI and showed that a BIS range of 80–86 was appropriate. Our findings indicated that a moderate sedation level with the MOAA/S score of less than 3 seemed suitable for AFOI. Furthermore, the BIS values were 79–91 when a sedation level with the MOAA/S score of 3 was achieved, which is in agreement with the findings of Gnaneswaran et al.’s study [45]. However, we noted that some patients achieved a deep sedation level with the MOAA/S score of 2, but their BIS values were larger than 82, which was within the above-mentioned range of BIS values. In the elderly patients aged 60–69 years receiving remimazolam for anesthesia induction, Liu et al. [46] also demonstrated that the BIS value associated with 50% of patients who fell asleep was 86, with a 95% confidence interval of 83.7–88.6. Thus, we consider that remimazolam 0.2 mg/kg/h combined with remifentanil 0.05 mcg/kg/min or alone remimazolam 0.6 mg/kg/h may be effective and safe scheme of moderate sedation for AFOI procedure, as they can produce moderate sedation level with the MOAA/S score of 3–4, BIS value ≥ 87 and few adverse respiratory events.

Moderate sedation for AFOI can not only alleviate patient anxiety, but also may improve patient comfort and intubation conditions [29]. The parameters, such as the patient’s reaction score to the AFOI procedure, cough severity during AFOI procedure, tolerance to tracheal tube following successful intubation, postoperative recall of AFOI procedure, and willingness to repeat the procedure, are frequently used to assess intubation conditions or patient comfort. In our study, cough severity during AFOI procedure, patient’ reactions to AFOI procedure, tolerance to tracheal tube following successful intubation, postoperative recall of the procedure, the number of attempts and total dosage of lidocaine were not significantly different among the 3-dose remimazolam-remifentanil groups. It suggests that these combined drug schemes can provide similar intubation conditions and patient comfort. In available randomized controlled trials regarding sedation management of AFOI, benzodiazepines such as midazolam are often applied as sedatives to decrease postoperative recall of the procedure [47]. Our results of postoperative recall scores for AFOI procedure indicated that all four remimazolam dosing regimens used in this study were effective in reducing postoperative recall of the procedure compared with remifentanil alone. Additionally, the patients in the RR0.6 group showed a higher acceptance of repeating the procedure than those in the RR0.2 and Rm groups, implying that a larger dose of remimazolam combined with remifentanil may improve patient comfort during AFOI procedure. However, it must be noticed that a larger dose of remimazolam with remifentanil 0.05 mcg/kg/min is associated with increased risks of deep sedation and adverse events including hypoxemia. Deep sedation may cause a loss of response to verbal commands and reduce patient cooperation. Even it may also increase collapsibility of the upper airway and decreases respiratory drive [48], producing loss of a patent airway, respiratory depression and hypoxemia.

It is well known that tracheal intubation can induce a strong stress response with elevated blood pressure and increased HR. Therefore, the hemodynamic changes associated with AFOI procedure are also used as indicators to assess the efficacy of a sedation regimen, as reported in previous studies [3,18]. In the present study, MAP changes during AFOI procedure did not statistically differ among the five sedation regimens. The dosing regimen of remimazolam 0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min seemed conducive to control HR changes during AFOI procedure compared to alone remifentanil 0.05 mcg/kg/min or alone remimazolam 0.6 mg/kg/h, as shown by lower HR and less tachycardia, in spite of higher incidences of deep sedation and hypoxemia. That is, sedation should not be strengthened as a substitute for airway topical anesthesia to inhibit cough reflexes and cardiovascular responses caused by airway manipulation. As a vital part of successful AFOI, airway topical anesthesia was carried out with the preferred ‘spray-as-you-go’ technique in this study [3,18]. The mean doses of lidocaine required for airway topical anesthesia in this study were not statistically different among the five sedation regimens and were less than 4 mg/kg, which is similar to those reported in previous studies [3,47] and significantly lower than the recommended maximal dose of 9 mg/kg for airway topical anesthesia in the available literature [8].

5. Limitations

This study has several limitations. First, this was a single centre clinical study with the small sample size. For safety and ethical reasons, only patients with normal airways were included. Thus, generalization of our results to the clinical scenario of managing predicted difficult airways should be done with caution because abnormal airway anatomy may make the airway open during sedation management and AFOI procedure more difficult. Although patients who experienced deep sedation in the RR0.4 and RR0.6 groups did not need naloxone or flumazenil antagonism in present study, these dose regimens indeed had a potential risk of losing a patent airway in patients with predicted difficult airways and should be used cautiously in such patients. Second, to our best knowledge, this is the first clinical trial determining the efficacy and safety of a combined drug regimen including the short-acting benzodiazepine and opioid (remimazolam and remifentanil) for moderate sedation during AFOI procedure. Given that opioids can cause dose-dependent respiratory depression [30], however, only a low intravenous infusion rate of remifentanil was chosen to minimize the risk of respiratory depression in the current study. Available evidence indicates that remifentanil 0.05 mcg/kg/min cannot provide adequate sedation and analgesia for airway manipulation [49]. The effective dose of remifentanil to suppress the cardiovascular changes associated with tracheal intubation is 0.1 mcg/kg/min after a loading dose of 1 or 0.2 mcg/kg/min when anesthesia is induced with propofol [50,51]. In the combination with remimazolam for moderate sedation during AFOI procedure, thus, the optimal dose of remifentanil deserves further study. Third, because the onset of remimazolam needs 1–3 min [11,20], AFOI procedure was started following intravenous infusion of remimazolam for 3 min in this study. Furthermore, it is reported that maximum plasma concentration of remimazolam is reached at 5 (3–20) min after starting intravenous infusion [35]. In the present study, the mean time required for AFOI procedure 6.1–7.5 min. That is, the peak effect time of remimazolam coincided with the time points that implement the critical steps of AFOI procedure, such as lower airway topical anesthesia, fiberscopic manipulations and advancement of tracheal tube. However, our results cannot determine whether these pharmacological features are advantages of remimazolam used for moderate sedation of the AFOI, as we did not establish a definite correlation between the pharmacologic features of remimazolam and the critical steps of AFOI procedure. This issue is very interesting and deserves further study. Fourth, the sample size estimation was based on the incidence of deep sedation. The sample size may be under-powered to show the between-group differences in some secondary outcomes including patient’ reaction score to the AFOI, cough severity score, MAP and HR. Fifth, intubation-related complications, such as sore throat and hoarsenesss, postoperative recovery and length of stay were not assessed in this study. To address above limitations, further clinical trials with strict design and sufficient samples are needed.

6. Conclusions

Remimazolam 0.2–0.6 mg/kg/h combined with remifentanil 0.05 mcg/kg/min are effective and feasible sedation regimens for AFOI procedure in patients with normal airway. Although the regimen including remimazolam 0.6 mg/kg/h may be more comfortable for patients who receive AFOI, the regimen including remimazolam 0.2 mg/kg/h should be the better choice for balancing efficacy and safety, especially for patients with potential difficult airway. Thus, an optimal dosing regimen of remimazolam combined with remifentanil for sedation during AFOI procedure in patients with predicted difficult airway is warranted for further study.

Funding Statement

This trial was supported by the 2020 Perioperative Sedation and Analgesia Funding of the Bethune Charitable Foundation (Grant number: BCF-RF-WSQZTZJ-202011-012 to CWL). The funder has no role in the study design, data collection or analysis, decision to publish, or preparation of the manuscript.

Author contribution statement

CRediT: Xiao-Rui Zhou: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft; Cheng-Wen Li: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Writing – original draft; Kai Su: Conceptualization, Data curation, Investigation, Methodology, Writing – review & editing; Yi Cheng: Conceptualization, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing; Mu Jin: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing; Fu-Shan Xue: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data related to this trial can be obtained by contacting the corresponding authors, if reasonable (jinmu0119@hotmail.com or xuefushan@aliyun.com).