The 95% effective dose of dexmedetomidine to induce adequate sedation in patients with chronic insomnia disorder: a biased coin design up-and-down sequential allocation trial

Abstract

Background

Chronic insomnia disorder is a common sleep disorder. Previous studies have reported increased demand for anesthetics in patients with chronic insomnia disorder. However, few studies have investigated the effective dose of dexmedetomidine for sedation in patients with chronic insomnia disorder. We aimed to determine the 95% effective dose of dexmedetomidine to induce adequate sedation in patients with chronic insomnia disorder.

Methods

This is a biased coin design up-and-down sequential allocation trial in patients aged 18–65 years with chronic insomnia disorder (n = 60). The dose of dexmedetomidine for the subsequent patient was determined based on the response of the previous patient, with an interval of 0.1 µg/kg. Sedation was assessed using the bispectral index. The 95% effective dose was calculated using isotonic regression.

Results

The 95% effective dose of dexmedetomidine to induce adequate sedation in patients with chronic insomnia disorder was 1.75 (95% confidence interval, 1.70–1.94) µg/kg. Four patients (6.7%) experienced bradycardia and no other adverse events were observed during the study.

Conclusions

The 95% effective dose of dexmedetomidine to induce adequate sedation in patients with chronic insomnia disorder was 1.75 µg/kg.

Trials registration

Chinese Clinical Trial Registry (ChiCTR2200063212); first registered 01/09 /2022.

Background

Insomnia disorder is one of the most common sleep disorders, characterized by difficulty falling asleep, difficulty maintaining sleep, and/or waking up earlier than intended, in addition to daytime functional impairment [1, 2]. Chronic insomnia disorder (CID) is defined by the presence of clinical symptoms of insomnia disorder for at least three nights a week that last for at least 3 months [3]. The pathological process of CID includes increased cortical activity, increased sympathetic excitability, and excessive arousal [4–6]. Additionally, the prevalence of anxiety disorder in patients with insomnia disorder is 20–30% [7]. Several clinical studies have reported increased demand for anesthetics in patients with insomnia or anxiety disorder [8–10].

Dexmedetomidine, a selective α2-adrenergic receptor agonist, has been widely used to induce perioperative sedation and analgesia [11]. Dexmedetomidine is anticipated to be a new type of sleep enhancer to treat the symptoms of insomnia since it can approximate physiological sleep with fewer adverse events [12]. However, the effective dose of intravenous dexmedetomidine for sedation in patients with CID has not yet been adequately described in the literature.

This study aimed to determine the 95% effective dose (ED₉₅) of dexmedetomidine to induce adequate sedation in patients with CID using the biased coin design up-and-down sequential method (BCD-UDM).

Methods

The Institutional Review Board of Shanghai Fourth People’s Hospital, Shanghai, China approved this study on 27 July 2022 (File No. 2022003-001; Chairperson, Hui Chen). Written informed consent was obtained from all patients. This study was registered on the Chinese Clinical Trial Registry (ChiCTR2200063212) on 01/09/2022. The research was conducted at Shanghai Fourth People’s Hospital, Shanghai, China between 1 September 2022, and 12 November 2023.

Patients were recruited from the outpatient sleep medicine clinics of Shanghai Fourth People’s Hospital. The criteria for inclusion were as follows: age, 18–65 years; fulfillment of the diagnostic criteria for CID per the third edition of the international classification of sleep disorders; subjective sleep onset latency > 30 min. The criteria for exclusion were as follows: lack of consent; the presence of bradycardia, hypotension, liver and kidney dysfunction; use of occupational sleep apnea solutions; restless legs syndrome; pregnancy or breastfeeding; shift work or irregular bedtime; untreated anxiety, depression or other mental disorders; allergy to dexmedetomidine.

Sedation protocol

All patients who volunteered to participate in the experiment were included. We asked the patients to fast for 4 h considering that they did not need to undergo surgery or an invasive exam. The patients were given the dexmedetomidine by an anesthetist in a monitored area. After measuring non-invasive blood pressure, heart rate (HR), and pulse oxygen saturation (SpO2) by BeneView T6 monitoring system (Mindray Medical International Co., Ltd, Guangdong, China), intravenous access was established. Bispectral index (BIS) monitoring (Aspect Medical Systems, Inc., MA, USA) was performed on all patients to evaluate the depth of sedation in this study.

Before administration, 10 µg/ml dexmedetomidine was prepared by diluting 200 µg dexmedetomidine (2 ml) (Youbituo™; Yangtze River Pharmaceutical Group Co., Ltd, China; 200 µg/ml) in saline 0.9% (18 ml). After being further diluted in saline 0.9% to 20 ml, the study dose was administered intravenously using an infusion pump (Mindray Medical International Co., Ltd, Guangdong, China). The infusion rate was 120 mL/h and the infusion time was 10 min.

Successful sedation was defined as BIS < 60 within 40 min following the initial dexmedetomidine administration. Failed sedation was defined as BIS ≥ 60 within 40 min following the initial dexmedetomidine administration. Sedation onset time was defined as the duration between the initial dexmedetomidine administration and BIS < 60. Adequate sedation time was defined as the time from sedation onset to BIS ≥ 60. Sedation time was defined as the time from sedation onset to BIS ≥ 90.

The mean arterial pressure (MAP), HR, SpO2 and BIS were collected at baseline and continuously monitored. The BIS recorded every 1 min until BIS ≥ 90. The MAP, HR and SpO2 recorded every 5 min until the patients were transported from the monitored area to the general ward. A Modified Aldrete score of 9 or higher was required to be transported to the general ward [13]. Given that residual dexmedetomidine may increase the risk of falls, particularly in patients taking hypnotic medications, all patients were admitted for one night of observation. The following morning, the patients were assessed in the general ward to document any adverse events such as nausea, vomiting and falls, through self-reporting.

Six milligrams of ephedrine and 0.5 mg of atropine were immediately accessible for every patient as rescue medications. Ephedrine or atropine was used to treat hypotension or bradycardia, and these boluses could be repeated as needed. Hypoxia was treated with oxygen inhalation through a nasal cannula or simple mask. Hypotension was defined as the MAP < 60 mmHg. Bradycardia was defined as HR < 50 beats per minute. Hypoxia was defined as SpO2 < 90%.

Biased coin design up-and-down sequential method

We used the BCD-UDM to determine the ED₉₅ of dexmedetomidine for the sedation of patients with CID. The starting dose of dexmedetomidine in this study was set at 0.8 µg/kg. The dose for each of the subsequent patients was determined based on the response of the previous patient. In case of a failed sedation, the dose was increased for the following patient by 0.1 µg/kg. While in the case of a successful sedation, the following patient was randomly chosen to either receive the same dose at P = 0.95 or a dose decreased by 0.1 µg/kg at P = 0.05. The biased coin allocation was achieved by a list of random responses, which was prepared by a statistician with Excel 2021. A research assistant used this random list to provide the dose of dexmedetomidine for the subsequent patient and set the delivery system. The anesthesiologists and patients remained blinded to the dose of dexmedetomidine throughout the study.

Outcome measurements

The primary outcome of the study was determining the ED₉₅ of dexmedetomidine to induce adequate sedation in patients with CID. The secondary outcomes included BIS values, sedation onset time, adequate sedation time, sedation time, and the incidence of bradycardia, hypotension, hypoxia, nausea, vomiting, and falls.

Statistical methods

Owing to the non-independence of the assignment of doses and the unknown distribution, the precise sample size for the BCD-UDM could not be determined [14]. A minimum sample size of 45 successful subjects is required for the BCD-UDM [15]. Owing to a lack of references, determining the appropriate starting dose of dexmedetomidine was challenging. Inappropriate starting doses may result in the need for more samples in BCD-UDM [15]. Taking into account these reasons, this study involved 60 patients with CID.

The ED₉₅ was calculated using isotonic regression analysis, and the 95% confidence interval (CI) was generated from 2,000 bootstrapped samples [16]. An adjusted response probability was obtained using the pooled-adjacent-violators algorithm (PAVA) [16]. R for Windows version 4.2.2 was used to perform the statistical analysis.

Categorical variables were presented as frequency. The normality of continuous variables was assessed using the Shapiro − Wilk test. Normally distributed data were presented as the mean ± standard deviation, and non-normally distributed data were presented as the median (interquartile range). BIS variables were analyzed by one-way repeated-measures analysis of variance. We used SPSS 26.0 for Windows as the statistical software.

Results

Eighty-two patients with CID were evaluated for inclusion in this study between 25 September 2023, and 9 November 2023. Among them, 16 patients did not meet the inclusion criteria and 6 patients refused to participate. Finally, 60 patients were included, all of whom completed the study (Fig. 1); among them, 48 and 12 patients demonstrated successful and failed sedation, respectively.

Fig. 1.

Flow diagram depicting patient inclusion and exclusion criteria in the study

The patient characteristics, time of sedation onset, and sedation period are demonstrated in Table 1. The onset time was 13.6 ± 4.9 min, the adequate sedation time was 18.5 ± 7.6 min, and the sedation time was 34.1 ± 9.5 min.

Table 1.

Characteristics of patients who received dexmedetomidine sedation

	Total		Successful sedation	Failed sedation
Sex (M/F)	14/46	11/37		3/9
Age (yr)	43.6 ± 11.4	43.8 ± 11.7		42.8 ± 10.8
Height (cm)	161.0 (158.0–167.0)	161.5 (158.0–167.3)		160.5 (157.3–171.8)
Weight (kg)	58.2 ± 9.2	57.6 ± 9.6		58.4 ± 11.3
Body mass index (kg/m2)	21.7 ± 2.5	21.8 ± 2.5		21.6 ± 2.8
Sedation onset time (min)	/	13.6 ± 4.9		/
Adequate sedation time (min)	/	18.5 ± 7.6		/
Sedation time (min)	/	34.1 ± 9.5		/

Data represented in frequency, mean ± standard deviation, or median (interquartile range)

The sequence of patients and the response of each patient to the assigned dose are illustrated in Fig. 2. The ED₉₅ value of dexmedetomidine calculated using isotonic regression analysis was 1.75 (95% CI, 1.70–1.94) µg/kg for the sedation of patients with CID. The frequency of response to the assigned dexmedetomidine dose and the observed and PAVA-adjusted success rates are demonstrated in Table 2.

Fig. 2.

Process for determining the ED₉₅ of dexmedetomidine to induce adequate sedation of patients with chronic insomnia disorder. Successful doses are denoted by solid circles, failed doses are denoted by open circles. The horizontal line represents the ED₉₅; error bars represent the 95% confidence interval. ED₉₅, 95% effective dose. * The patients who experienced bradycardia

Table 2.

Observed and pooled-adjacent-violators algorithm (PAVA)-adjusted success rates

Assigned dose (µg/kg)	Success (number)	Tested (number)	Observed success rate	PAVA-adjusted success rate
0.8	0	1	0.000	0.000
0.9	1	2	0.500	0.143
1.0	0	1	0.000	0.286
1.1	1	2	0.500	0.375
1.2	0	1	0.000	0.438
1.3	1	2	0.500	0.500
1.4	2	3	0.667	0.667
1.5	4	5	0.800	0.747
1.6	8	10	0.800	0.823
1.7	20	22	0.909	0.909
1.8	11	11	1.000	1.000

Following the initial dexmedetomidine administration, a significant decline was observed in BIS from the baseline at 5 min; it reached the lowest level at 20 min and returned to baseline at 65 min (Fig. 3).

Fig. 3.

BIS response to dexmedetomidine administration. Significant decrease in BIS from baseline was observed at 5 min after initial dexmedetomidine administration; it reached the lowest level at 20 min and returned to baseline at 65 min. *P < 0.001 compared to baseline. BIS, bispectral index

Four patients (6.7%) experienced bradycardia and recovered after receiving 0.5 mg atropine (Fig. 2). No other adverse events including hypotension, hypoxia, nausea, vomiting or falls were observed in this study.

Discussion

Our study demonstrated that the ED₉₅ of dexmedetomidine to induce adequate sedation in patients with CID was 1.75 (95% CI, 1.70–1.94) µg/kg. In this study, four patients (6.7%) experienced bradycardia and no other adverse were events observed.

Insomnia disorder is the most common sleep disorder affecting 6–10% of the general population [17]. In addition, it is estimated that 260–360 million surgeries are performed annually worldwide [18]. Previous studies have demonstrated that intraoperative dexmedetomidine use can improve postoperative sleep quality in patients undergoing surgery [19, 20]. It can be inferred that a large number of patients with CID require perioperative sedation each year. Therefore, it is necessary to determine the effective dose of dexmedetomidine for sedation in patients with CID. To our knowledge, this was the first study to investigate the effective dose of dexmedetomidine to induce sedation in patients with CID, which could provide clinical guidance.

Previous studies have investigated the 50% effective doses (ED₅₀) of dexmedetomidine using the conventional up-and-down sequential method (UDM) [21, 22]. However, ED₅₀ has limited clinical significance since it indicates a 50% failure rate by definition. Unlike ED₅₀, ED₉₅ indicates the dose of drug required to produce effects in most patients. In clinical practice, knowing the ED₉₅ helps anaesthetists determine the anaesthetic dose needed to ensure most patients achieve adequate anaesthesia to prevent movement or pain during surgery. The conventional UDM extrapolates ED₉₅ from ED₅₀, which cannot be verified, whereas BCD-UDM can directly estimate ED₉₅ [15]. In many UDM studies, parameter estimators (logistic or probit regression) were used to calculate the CI of potency doses [23]. However, it has been demonstrated that the parameter estimators underestimate the statistical variation in potency doses, leading to an unreasonably narrow CI [24]. The isotonic regression technique obtains adjusted response probabilities using PAVA, estimates ED₉₅ using nonparametric isotonic regression, and calculates CI using a parametric bootstrap routine. These methods reduce bias by relying less on untestable hypotheses [15]. In this study, we used BCD-UDM and isotonic regression analysis to estimate the ED₉₅ of dexmedetomidine for the sedation of patients with CID, which can reduce bias and variance.

We selected BIS to asses sedation levels since recoding some subjective scores, such as observer’s assessment of alert/sedation (OAA/S) and Ramsay sedation scores, requires the repeated stimulation of patients, which results in an uncomfortable experience, especially in patients with CID. According to previous studies, there is a significant correlation between BIS and subjective scores during dexmedetomidine sedation [25–27]. Since 85% of the BIS values were between 40 and 60 when the OAA/S responsiveness was 3 during dexmedetomidine sedation [26], we defined successful sedation as BIS < 60 after dexmedetomidine administration in this study.

Intravenous dexmedetomidine may cause hypotension and bradycardia due to its central sympatholytic effects [28]. For procedural sedation, a loading dose of dexmedetomidine of 1 µg/kg followed by a maintenance infusion beginning at 0.6 µg/kg/h is recommended [29]. Candiotti et al. and Bergese et al. demonstrated that the incidence of hypotension and bradycardia in procedural sedation with dexmedetomidine was 8.5–27.3% and 4.7–7.3%, respectively. In this study, 56 patients (93.3%) received a loading dose of dexmedetomidine higher than 1 µg/kg; however, no patient experienced hypotension only four patients (6.7%) experienced bradycardia. Our study revealed that dexmedetomidine can induce adequate sedation in patients with CID safely. We speculate that the increased sympathetic excitability associated with insomnia may play a role in modulating patients with CID tolerance to the cardiovascular effects of dexmedetomidine.

Xiong et al. demonstrated that the ED₉₅ of a single bolus of dexmedetomidine for adequate sedation (OAA/S ≤ 3) was 1.80 µg/kg in parturients and 1.10 µg/kg in non-pregnant females, respectively [30]. Kim et al. revealed that the ED₉₅ for light sedation (OAA/S 3/4) was 0.57 µg/kg in patients aged 45 − 64 years undergoing spinal anesthesia for transurethral prostate excision [31]. Our study revealed that 1.75 µg/kg dexmedetomidine can induce adequate sedation (BIS < 60) in patients with CID. Owing to the different definitions of successful sedation, we could not determine that patients with CID require a higher dose of dexmedetomidine than the general population to achieve the same level of sedation; thus, further controlled trials are necessary to establish this.

Dexmedetomidine decreases rapid eye movement (RME) sleep and increases non-REM stage 2 (NREM2) sleep in patients without CID, and an increased dose of dexmedetomidine can induce non-REM stage 3 (NREM3) sleep when deep sedation is achieved [32, 33]. A preliminary study has explored the feasibility of using dexmedetomidine in the treatment of CID albeit they did not use polysomnography to compare sleep reported and did not determine the sleep-inducing dose of dexmedetomidine [34]. The ability and dose of dexmedetomidine to induce NREM2 or NREM3 sleep in patients with CID must be further explored.

This study has some limitations that should be considered. First, the design of this study was uncontrolled, limiting the conclusions that could be drawn and not determining without doubts that patients with CID have a higher demand for dexmedetomidine. We could not determine that patients with CID have a higher demand for dexmedetomidine. Second, the study is that we only enrolled patients aged 18–65 years and not older patients. Elderly patients are more susceptible to hypotension and bradycardia following the administration of dexmedetomidine than are younger patients owing to the drug’s decreased clearance rate and prolonged half-life [35, 36]. Dexmedetomidine is therefore prescribed at a lower dose for elderly patients. Third, patients with untreated anxiety, depression or other mental disorders were excluded, which impacts the clinical interpretation. Future studies should include recruiting patients with co-medications and co-morbidities. Fourth, patients with CID usually take sedative drugs, such as benzodiazepines, tricyclic antidepressants, histamine receptor antagonists, and antipsychotics [37, 38], which might influence the degree of sedation. Fifth, this study did not assess and grade patients based on CID severity.

Conclusion

In conclusion, this study demonstrated that the ED₉₅ of dexmedetomidine to induce adequate sedation in patients with CID who met the study criteria was 1.75 µg/kg.

Abbreviations

CID

Chronic insomnia disorder

ED95

95% effective dose

BCD-UDM

Biased coin design up-and-down sequential method

HR

Heart rate

SpO2

Pulse oxygen saturation

BIS

Bispectral index

MAP

Mean arterial pressure

CI

Confidence interval

PAVA

Pooled-adjacent-violators algorithm

ED50

50% effective doses

UDM

Up-and-down sequential method

OAA/S

Observer’s assessment of alert/sedation

RME

Rapid eye movement

NREM2

Non-rapid eye movement stage 2

NREM3

Non-rapid eye movement stage 3

Author contributions

LMY contributed to design the study，collect the date and draft the manuscript. LGF contributed to interpret and analyze the data. LQF and CQ contributed to design and conduct the study, and revise the manuscript. All authors reviewed the manuscript.

Funding

This study was supported by grants from Shanghai Fourth People’s Hospital, School of Medicine, Tongji University (No.SY-XKZT-2022-1005).

Data availability

The datasets of this study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of Shanghai Fourth People’s Hospital, Shanghai, China. (File No. 2022003-001). Written informed consent was obtained from all patients. All methods in this study were performed in accordance with the relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Conflict of interest

The authors declare that they have no conflicts of interest.