Does intranasal dexmedetomidine provide adequate plasma concentrations for sedation in children: a pharmacokinetic study

British Journal of Anaesthesia, 120 (5): 1056e1065 (2018) doi: 10.1016/j.bja.2018.01.035 Advance Access Publication Date: 13 March 2018 Paediatrics

Does intranasal dexmedetomidine provide adequate plasma concentrations for sedation in children: a pharmacokinetic study J. W. Miller1,*, R. Balyan1,2, M. Dong2, M. Mahmoud1, J. E. Lam1, J. N. Pratap1, J. R. Paquin1, B. L. Li3, J. P. Spaeth1, A. Vinks2,4 and A. W. Loepke5 1

Department of Anesthesiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA, 2Department of Clinical Pharmacology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA, 3Department of Anaesthesiology, Guangzhou Women and Children’s Medical Center and Guangzhou Medical University, Guangzhou, China, 4Department of Paediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, USA and 5Department of Anesthesiology and Critical Care Medicine, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA

*Corresponding author. E-mail: [email protected]

Abstract Background: Atomised intranasal dexmedetomidine administration is an attractive option when sedation is required for paediatric diagnostic procedures, as vascular access is not required. The risk of haemodynamic instability caused by dexmedetomidine necessitates better understanding of its pharmacokinetics in young children. To date, intranasal dexmedetomidine pharmacokinetics has only been studied in adults. Methods: Eighteen paediatric patients received dexmedetomidine 1 or 2 mg kg
1 intranasally or 1 mg kg
1 i.v. Plasma concentrations were determined by liquid chromatography/mass spectrometry. Non-compartmental analysis provided estimates of Cmax and Tmax. Volume of distribution, clearance, and bioavailability were estimated by simultaneous population PK analysis of data after intranasal and i.v. administration. Dexmedetomidine plasma concentration-time profiles were evaluated by simulation for intranasal and i.v. administration. Results: An average peak plasma concentration of 199 pg ml
1 was achieved 46 min after 1 mg kg
1 dosing and 355 pg ml
1 was achieved 47 min after 2 mg kg
1 dosing. A two-compartment pharmacokinetic model, with allometrically scaled parameters, adequately described the data. Typical bioavailability was 83.8% (95% confidence interval 69.5e98.1%). Conclusion: Mean arterial plasma concentrations of dexmedetomidine in infants and toddlers approached 100 pg ml
1, the low end reported for sedative efficacy, within 20 min of an atomised intranasal administration of 1 mg kg
1. Doubling the dose to 2 mg kg
1 reached this plasma concentration within 10 min and achieved almost twice the peak concentration. Peak plasma concentrations with both doses were reached within 47 min of intranasal administration, with an overall bioavailability of 84%. Keywords: anaesthesia; dexmedetomidine; intranasal; paediatrics; pharmacokinetics

Editorial decision: January 12, 2018; Accepted: January 12, 2018 © 2018 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved. For Permissions, please email: [email protected]

1056

Intranasal dexmedetomidine pharmacokinetics in children

Editor’s key points  Dexmedetomidine is a highly selective a2 adrenergic agonist with anxiolytic, sedative, and mild analgesic properties.  Although intranasal use is well-described in children, little is known of its pharmacokinetics in this setting.  The pharmacokinetics of intranasal and i.v. dexmedetomidine was studied in 18 children.  After intranasal administration, peak plasma concentrations occurred at 46 min, and bioavailability was >80%.

Intranasal dexmedetomidine is an attractive needle-free sedative for paediatric use where i.v. access is unnecessary or may be deferred until sedation is achieved. Dexmedetomidine has been widely used for paediatric sedation for nonpainful procedures.1 It has a short half-life, promotes a calm emergence, and is associated with maintenance of airway stability and spontaneous ventilation.2e4 Dexmedetomidine is effective and painless when administered by the nasal route.5 Pharmacodynamic studies of intranasal dexmedetomidine onset, efficacy, and adverse effects have been reported in children.6e10 To minimise side effects and instruct clinical dosing, it is imperative to gather pharmacokinetic information on dexmedetomidine in paediatric patients. The pharmacokinetics of i.v. dexmedetomidine have been studied in children, including premature infants.11e13 However, while one pharmacokinetic study of intranasal bolus dosing of dexmedetomidine has been performed in adult humans,14 no such data exist in children. This represents a substantial lack of information, as rapid administration of dexmedetomidine is associated with hemodynamic alterations, such as bradycardia, arterial hypotension, and hypertension.15 The margin of safety with intranasal dosing is currently unknown. Accordingly, this represents the first pharmacokinetic study of atomised intranasal dexmedetomidine in children. Cardiac surgery patients were selected for this study because of the availability of intra-arterial access for repeated blood sampling, routine use of dexmedetomidine as part of our anaesthetic plan, and our previously published pharmacodynamic data with intranasal dexmedetomidine in a similar patient population.7,10 The aim of the study was to characterise the peak drug concentration in plasma for two different doses of intranasal dexmedetomidine and to determine bioavailability compared with i.v. dosing. Based on prior adult pharmacokinetic and paediatric pharmacodynamic data, we hypothesised that intranasal atomised dexmedetomidine would reach plasma concentrations associated with sedation in approximately 20 min, with peak plasma concentrations within 60 min of administration.7,14

Methods This prospective pharmacokinetic study in a single quaternary paediatric referral centre was registered at clinicaltrials.gov on January 8, 2016 (NCT02836431) with principal investigator: J.W.M. Dexmedetomidine is not approved by the US Food and Drug Administration (FDA) for use in paediatric patients; accordingly, this study was conducted under an Investigational New Drug (IND # 76-346) application. After FDA review, the

-

1057

Institutional Review Board of Cincinnati Children’s Hospital Medical Center approved this trial on December 1, 2015. This manuscript adheres to the applicable Equator guidelines.16

Study population The enrolment period was from January 2016 to December 2016. Paediatric patients for elective cardiac surgery aged 6e48 months were eligible for this study. All patients were scheduled to receive postoperative sedation with i.v. dexmedetomidine infusions. Written, informed consent was obtained from a parent or legal guardian. Eighteen children were sequentially assigned to three treatment groups, each containing six children. With no prior available paediatric data and given funding constraints, we chose six patients per group and eight to nine bioassay samples per patient. Terminal elimination was not studied because of the confounding effects of hemodilution with initiation of cardiopulmonary bypass (CPB) within 2 h of drug administration. Exclusion criteria were: patients admitted to the hospital before surgery, patients with a history of cardiac conduction system disease (e.g. sinus node disease, first or second degree atrioventricular block) or formally diagnosed channelopathy (e.g. "long QT syndrome"), current treatment with digoxin, alpha-adrenergic or beta-adrenergic agonists or antagonists, clonidine, anti-arrhythmic medications, anticonvulsants, presence of life-threatening medical conditions (ASA Physical Status 4 or 5), previous exposure to dexmedetomidine within 1 week, or unrepaired coarctation of the aorta. Patients with acute nasal or respiratory symptoms on the day of the study were excluded because of potential interference with intranasal absorption.

Clinical protocol All subjects received an inhalation induction of anaesthesia with sevoflurane followed by i.v. and intra-arterial cannula placement and tracheal intubation. Medications administered before surgery and during the study period were recorded. Standard monitoring per ASA guidelines was fully instituted before dexmedetomidine administration. Patients were in the supine position with their head positioned in the midline. The first six patients received undiluted intranasal dexmedetomidine (Mylan Pharmaceuticals, Canonsburg, PA, USA; 100 mg ml
1), 1 mg kg
1, by atomiser (Teleflex MAD Nasal; Research Triangle Park, NC, USA) in one naris as a rapid spray from a 1 ml syringe containing the dexmedetomidine and 0.2 ml air as a ‘chaser’ to ensure complete expulsion of the medication.17 The second six patients received 2 mg kg
1 intranasal dexmedetomidine in one naris in identical fashion. The maximum volume administered was 0.29 ml. The syringe was oriented approximately 30 degrees from horizontal in supine patients to direct the atomised medication cephalad toward the upper third of the nasal cavity.18 All nasal dexmedetomidine administration was performed by a single physician (J.W.M.) with extensive experience using the atomiser. To determine bioavailability, the final six patients received 1 mg kg
1 i.v. dexmedetomidine (Mylan Pharmaceuticals; 4 mg ml
1) over 10 min by infusion pump into a peripheral i.v. cannula. Supplemental anaesthetic drugs were reduced by the anaesthesia team in anticipation of the co-anaesthetic effects of the administered dexmedetomidine. The patients each

1058

-

Miller et al.

received less than 10 ml kg
1 i.v. fluid before CPB by commonly agreed practice in our institution. A case review for potential adverse events related to dexmedetomidine was completed for each group before the increase in dose.

Plasma sample collection and bioassay Arterial blood samples (1 mL) were obtained before administration of dexmedetomidine (‘baseline’) and then at 10 min intervals after administration for the first hour. The 10-min sample in the i.v. dexmedetomidine group was obtained as soon as the 1 mg kg
1 bolus was completed. If CPB was not initiated by 120 min after dexmedetomidine administration, then a 120-min sample was collected. A final arterial sample was obtained 2 min after administration of 400 units kg
1 of heparin for CPB, a time point that varied among study participants. Blood samples were centrifuged, and plasma was separated and stored in a polyethylene tube at
70 C until bioassay. Need for atropine, glycopyrrolate, epinephrine, ephedrine, or electrical pacing before cannulation for CPB were documented as potential adverse events. Arterial plasma dexmedetomidine was quantified by a validated isotope-dilution 2D high-performance liquid chromatography/tandem mass spectrometry (LC/LC-MS/MS) assay by iC42 Clinical Research and Development (University of Colorado Denver, Aurora, CO, USA). Plasma specimens were analysed by liquid chromatography tandem mass spectrometry (LC-MS/MS). Briefly, 100 ml of plasma was transferred into a 1.5 ml vial and 600 ml of 0.2 M ZnSO4 30% water/70% methanol (v/v) containing the internal standard d4-dexmedetomidine (25 pg ml
1) was added for protein precipitation. Samples were vortexed for 10 min and then centrifuged (at 26 000 g, 4 C, 10 min). The supernatant was transferred in a glass HPLC into the autosampler (4 C). Dexmedetomidine was quantified using LC-MS/MS on an ABSciex API5000 tandem mass spectrometer via a turbo V ion source operated in positive atmospheric pressure chemical ionisation mode (both Sciex, Concord, ON, Canada). For online extraction, 200 ml of the samples were loaded onto a 4.612.5 mm online extraction column (Zorbax XDB C8, Agilent Technologies, Santa Clara, California, USA) with a particle size of 5 mm. Samples were washed with a 5 ml min
1 flow of 0.1% formic acid in water/ methanol 95/5 (v/v). After 1 min, the switching valve was activated and the analytes were back-flushed from the extraction column onto a 4.6100 mm Kinetex F5, 2.6 mm analytical column (Phenomenex, Torrance, CA, USA). Mobile phase (A) was 0.1% formic acid in water and (B) was 0.1% formic acid in methanol at a flow rate of 0.8 ml min
1. The gradient program was 25% B for 1 min, to 85% B at 3.0 min, 95% B at 4.5 min, holding for 2 min, and then re-equilibrated to starting conditions 25% B over 0.1 min and held for 1.5 min before the next injection. The MS/MS system operated in the positive multiple reaction monitoring mode. The ion transitions monitored were m/z¼201.15/ 95.10 and m/z¼205.50/ 99.10 for dexmedetomidine and the internal standard d4dexmedetomidine, respectively. Source parameters were curtain gas flow 20 litre min
1, ion spray voltage 5500 V, source temperature 500 C, and ion source gas 240 litre min
1. MS/MS settings were declustering potential 80 V, exit potential 10 V, collision energy 25 eV, and cell exit potential 16 V. Quantification was based on plotting the dexmedetomidine/internal standard area-under-the-peak ratios and external calibration curves (linear fit, 1/X weighting) using the Analyst software

(version 1.6.2., Sciex, Framingham, Massachusetts, USA). Lower limit of quantifications was 20 pg ml
1 and linear range was 20e1000 pg ml
1 (r2>0.99) for dexmedetomidine. The four concentrations of the quality control samples were 64, 240, 600, and 800 pg ml
1. The extraction efficiency (recovery) ranged from 79.9% to 101.2% and was determined at the concentrations of the quality control samples following the recommendations by Matuszewski and colleagues.19 The imprecision was (coefficient of variance) 4.6e12.9%. No significant matrix effects or carry-over was detected.

Pharmacokinetic analysis Pharmacokinetic analysis was performed using descriptive non-compartmental analysis with WinNonlin (version 7.0, Certara, Princeton, NJ, USA) and the concentration-time plots for individual patient, maximum plasma concentration (Cmax), and time at which Cmax was observed Tmax estimates were derived. To compare our findings with literature reports, graphical coordinates of published plots were obtained by DIGIT software (version 1.0.4, Simulation Plus Inc., Lancaster, CA, USA) and a summary graph depicting published results along with our observations was developed using GraphPad software (GraphPad Prism version 7.02, La Jolla, CA, USA). The population pharmacokinetic modelling of data was performed by non-linear mixed effect modelling with NONMEM software (version 7.2, ICON, Ellicot City, MD, USA). One and two compartment models were tested to describe the dexmedetomidine plasma concentration-time profile, during model development. The first order conditional estimation method with interaction was used for all runs. A combined additive and proportional error model was used to estimate the residual error. The best model was selected using goodness of fit diagnostic plots, objective function value (OFV) and evaluation of population fixed effect and random parameters. To account for the influence of variation in body size on PK, PK parameters were allometrically scaled with body weight using a power coefficient of 0.75 for clearance (CL) and 1.0 for volume of distribution (V), as shown in the following equation:20

Pi ¼ Ppop x½BWi =70power

(1)

where BWi represents the body weight of individual (i) and 70 kg is a standard adult body weight. An intranasal absorption model was developed to obtain estimates of bioavailability (F). Dexmedetomidine intranasal administration and i.v. administration data were obtained from different patients. The sampling times were similar for the i.v. and intranasal groups. Simultaneous analysis of intranasal and i.v. administration data was performed to obtain population PK parameter estimates for CL, V, and bioavailability. A first order absorption rate constant (Ka) and lag time (ALAG) were included and fixed in the PK model to describe dexmedetomidine absorption. Different values for Ka and ALAG for first order absorption with lag time for intranasal dexmedetomidine administration were tried and best fit was obtained by fixing Ka and ALAG at 0.0157 min
1 and 5.00 min, respectively.

Model evaluation The goodness of fit diagnostic plots, OFV and Akaike information criterion (AIC) were used for model evaluation. The plots for consideration included observed value (DV) vs population predicted value (PRED), DV vs individual predicted

Intranasal dexmedetomidine pharmacokinetics in children

value (IPRED), conditional weighted residuals (CWRES) vs PRED, and CWRES vs time after dose. The final models were evaluated using visual predictive check (VPC) plots. One thousand datasets were simulated from the final model and median and 90% confidence intervals (CI) of simulated data were plotted along with DV.

Statistical analysis Patient characteristics along with estimated Cmax, Tmax, CL, V, and F values were summarised as median, mean [standard deviation (SD)] and range of observations, as appropriate. The plasma concentrations of dexmedetomidine at different time points are presented as mean (SD). All model parameters are reported as estimated values with associated relative standard errors (RSE) and 95% CI. A two-tailed unpaired t-test was performed to compare the dexmedetomidine Tmax values observed in the current study with values reported in the adult intranasal dexmedetomidine study.14,21

Results During the period from January 21, 2016 to December 16, 2016, 18 children were studied. Ninety-four patients were screened for eligibility and 24 patients met inclusion criteria when adequate research staff were available. The rate of consent to participate in this study was 88% (21 patients). Once enrolled, there were three dropouts because of a preoperative change in medical status observed before anaesthetic induction. Subject characteristics are presented in Table 1. The average age was 25.6 months (range 6e44 months). The study population included eight males and 10 females, two selfidentified as AfricaneAmerican and 16 as Caucasian, three prematurely born children (estimated gestational age less than 37 weeks), and four cyanotic subjects with baseline awake oxygen saturation less than 92%. Three subjects were taking enalapril (held on the day of surgery), one was taking aspirin, one was taking furosemide, and one was taking levothyroxine. No patient had a diagnosis of hepatic disease. All subjects received sevoflurane, vecuronium, fentanyl, and cefuroxime in addition to dexmedetomidine during the study period; all but one also received tranexamic acid. The diagnoses or procedures for each group were as follows: Group intranasal DEX 1 mg kg
1 included two atrial septal defects, one ventricular septal defect, and three right ventricular outflow tract repairs (one pulmonary valvar stenosis and two truncus arteriosus). Group intranasal DEX 2 mg kg
1 included two atrial septal defects, two aortic stenosis variants (valvar aortic stenosis, sub-aortic stenosis), one partial

-

1059

atrioventricular canal defect, and one aortopexy for tracheal compression. Group i.v. DEX 1 mg kg
1 included one atrial septal defect, one tetralogy of Fallot, two atrioventricular canal variants, and two single ventricle variants with cavopulmonary anastomoses undergoing a modified Fontan procedure. No major adverse events (defined as administration of a vasopressor, chronotropic agent, or electrical pacing) occurred after dexmedetomidine administration until cannulation for CPB in any treatment group.22 The duration of arterial blood sampling before CPB was at least 77 min in all patients (maximum 174 min).

Pharmacokinetic analysis A total of 148 plasma samples were collected from 18 subjects. Plasma concentrations of dexmedetomidine over time for the three groups are presented in Table 2 and Fig 1. Individual plasma concentrations at different time points along with fitted PK profiles are shown in Supplementary Fig S1. The average peak measured dexmedetomidine concentration (Cmax) was 199 (41) pg ml
1 at 46 (11) min (Tmax) after 1 mg kg
1 intranasal dexmedetomidine. The Cmax was higher at 355 (122) pg ml
1 with a Tmax of 47 (12) min after 2 mg kg
1 intranasal dexmedetomidine. To compare our findings with published data, our 2 mg kg
1 intranasal dexmedetomidine group plasma concentrations were overlaid with concentration time profiles after extravascular administration of dexmedetomidine at the same dose, reported by Anttila and colleagues23 (Fig 2). The concentration time profile of i.v. and intranasally administered dexmedetomidine were best described with a two-compartment model, compared with a one compartment model (DOFV¼419). Application of a three-compartment model did not further improve the model fit. This two compartment model was parameterised with zero order infusion for i.v. and first order absorption for intranasal with lag time and incorporating bioavailability. The best fit was obtained by fixing the absorption rate constant (Ka) and lag time (ALAG) at 0.0157 min
1 and 5.00 min, respectively. Age analysed as estimated gestational age (EGA in weeks) was not found to be a significant covariate.20 The final PK model captured the PK profiles well, as confirmed by the diagnostic plots (Fig 3). Dexmedetomidine CL, V, and bioavailability were estimated by NONMEM. The population PK parameters along with estimated value and 95% CI for the i.v. and intranasal groups are summarised in Supplementary Table S1. The population mean values (RSE%) for systemic CL, central V, distribution CL (Q), and peripheral volume (V2) were 1.04 (8.03%) litre 70 kg
1 min
1, 59.8 (27.8%) litre 70 kg
1, 2.54 (20.1%) litre 70 kg
1 min
1, and 95.1 (13.6%) litre 70 kg
1, respectively.

Table 1 Subject characteristics. Data presented as mean ± (standard deviation) or subject count. DEX, dexmedetomidine; F, female; M, male; SpO2, peripheral oxygen saturation

Age (months) Weight (kg) Gender M:F (n) Baseline cyanotic (SpO2<92) Haematocrit Creatinine Sampling duration (min)

Intranasal DEX 1 mg kg¡1

Intranasal DEX 2 mg kg¡1

I.V. DEX 1 mg kg¡1

21.5 (7e44) 11.2 (2.8) 5:1 1 39 (4) 0.29 (0.04) 100 (18)

27.5 (12e44) 12.1 (2.2) 1:5 0 38 (2) 0.30 (0.05) 119 (19)

27.8 (6e42) 12.4 (4.7) 2:4 3 43 (5) 0.33 (0.06) 121 (31)

1060

-

Miller et al.

Table 2 Pharmacokinetic parameters after atomised intranasal and i.v. dexmedetomidine administration of dexmedetomidine to paediatric cardiac surgical patients. Cmax, peak plasma concentration; Tmax, time of Cmax; SD, standard deviation

Intranasal 1 mg kg
1

Intranasal 2 mg kg
1

Intravenous 1 mg kg
1

Patient

Age (months)

Height (cm)

Weight (kg)

Cmax (pg ml¡1)

Tmax (min)

1 2 3 4 5 6 Median Range 7 8 9 10 11 12 Median Range 13 14 15 16 17 18 Median Range

7 18 7 44 17 36 17.5 7e44 12 27 34 17 44 31 29 12e44 32 42 39 6 7 41 35.5 6e42

71.0 90.5 66.0 102.5 82 92.5 86.3 66e102.5 78.5 91.0 91.0 87.0 98.1 88.9 90.0 78.5e98.1 95.5 99.5 101.6 63.5 64.0 90.3 92.9 63.5e101.6

8.4 11.8 9 15.7 9.8 12.8 10.8 8.4e15.7 9.2 12.5 14.4 11.4 14.6 10.5 11.9 9.2e14.6 15.2 14.8 18.2 7.2 6.6 12.5 13.7 6.6e18.2

187 250 163 251 163 177 182 163e251 319 329 338 318 597 229 324 229e597 813 1012 753 496 460 1030 783 460e1030

51 42 40 31 51 62 46.5 31e62 32 56 65 51 40 40 45.5 32e65 e e e e e e e e

The typical bioavailability was estimated as 83.8% with a 95% CI of 69.5e98.1% and 8.72% RSE. The estimated CL, V1, Q, V2, bioavailability, Ka, and ALAG for each individual subject are listed in Supplementary Table S2. The simulated PK profiles by VPC for both i.v. and intranasal administered dexmedetomidine (Fig 4) were in reasonable agreement with observed concentrations.

Discussion

for non-painful paediatric diagnostic procedures such as transthoracic echocardiography, computerised tomography, magnetic resonance imaging, and auditory brainstem response testing.1 Dexmedetomidine is an attractive sedative because of its lack of respiratory depression, even at very high doses,25 however, it can lead to hemodynamic compromise, emphasising the need for identifying the lowest dose to provide therapeutic effects. Although i.v. dexmedetomidine pharmacokinetics has been studied, no pharmacokinetic data of intranasal dexmedetomidine in children has previously

Since the first description by Yuen and colleagues24 in 2007, intranasal dexmedetomidine sedation is increasingly utilised

Fig 1. Dexmedetomidine concentrations versus time. Dexmedetomidine concentrations in arterial plasma (mean, SD) versus time after administration of 1 mg kg
1, 2 mg kg
1 intranasally (i.n.) compared to intravenous (i.v.) 1 mg kg
1 infused over 10 minutes.

Fig 2. Extravascular administration of dexmedetomidine. Comparison of extravascular administration of dexmedetomidine. Our data was overlaid on the plot from Anttila. Concentration time profile of dexmedetomidine level following administration of 2 mg kg
1 dose by the intranasal route along with mean curves observed after administration of 2 mg kg
1 by intramuscular (i.m.), intravenous (i.v.), buccal and oral routes by Anttila.

Intranasal dexmedetomidine pharmacokinetics in children

-

1061

Fig 3. Goodness-of-fit plots for pharmacokinetic model. Goodness-of-fit plots for the final pharmacokinetic model of simultaneous analysis of the intravenous and intranasal administered dexmedetomidine. Observed vs. (A) population-predicted and (B) individual-predicted dexmedetomidine concentrations (line of identity shown for clarity). The conditional weighted residuals (CWRES) vs. (C) time after dose and (D) population-predicted dexmedetomidine concentration.

been reported. The present study fills this gap in knowledge, by demonstrating that plasma concentrations exceeding 100 pg ml
1 can be achieved in 6e44-months-old children after administration of 1 mg kg
1 of intranasal dexmedetomidine within approximately 20 min. Peak plasma concentrations of 199 pg ml
1 were achieved within 46 min of administration. Doubling the intranasal dexmedetomidine dose to 2 mg kg
1 achieved peak concentrations (355 pg ml
1) almost twice that of 1 mg kg
1, but with a similar time to peak concentration. The overall bioavailability of intranasal dexmedetomidine in our model was 83.8%. The plasma concentrations observed after intranasal administration in our study were higher than plasma concentrations after oral and buccal administration, but lower than after i.m. and i.v. administration in adults.23 The only other intranasal dexmedetomidine study was performed in adults by Iirola and colleagues.14 In their study, intranasal dexmedetomidine was administered to six adult male volunteers by nasal spray device using a veterinary formulation of dexmedetomidine with a concentration of 500 mg ml
1 (0.2 ml). Median peak arterial plasma concentration of 340 (range 230e700 pg ml
1) was reached within 38 min (range 15e60 min) and absolute bioavailability by non-compartmental analysis was 65% (range 35e93%). There was a large variation in time to peak concentration in both studies. The mean Tmax value in the current study (47 min) was longer than what

has been reported for adults (38 min).14 However, this difference was not statistically significant (P¼0.128). The target plasma concentration for paediatric procedural sedation using dexmedetomidine without concomitant opioids and benzodiazepines is unknown. The plasma concentration of dexmedetomidine to achieve sedation in adults may be approximately 200e400 pg ml
1.21,23 Su and colleagues26 reported minimal, moderate, and deep sedation with median plasma concentrations of 120, 370, and 430 pg ml
1 of dexmedetomidine in paediatric patients in the intensive care unit (ICU) with infusions of dexmedetomidine after cardiac surgery; however, these patients received supplemental midazolam and analgesics. Even higher plasma concentrations of 400e800 pg ml
1 have been reported for i.v. ICU sedation in children and adults.27 Ebert and colleagues25 reported that in adult volunteers, despite significant sedation, recall and picture recognition were preserved at plasma concentrations below approximately 1000 pg ml
1. From these studies, a dexmedetomidine therapeutic range for paediatric sedation with dexmedetomidine as a sole agent may be approximately 200e600 pg ml
1. This concentration was reached in the group receiving 2 mg kg
1 intranasally (peak plasma concentration 333 pg ml
1). Larger doses such as 3e4 mg kg
1 intranasal dexmedetomidine may produce plasma concentrations within this range and provide an earlier onset time, albeit with a higher potential for adverse hemodynamic effects.

1062

-

Miller et al.

Fig 4. Visual predictive check for pharmacokinetic model. Visual predictive check (VPC) for the final model of dexmedetomidine administered by (A) 1 mg kg
1 intravenous (B) 1 mg kg
1 intranasal, and (C) 2 mg kg
1 intranasal routes. Open circles represent observed plasma concentrations; lines represent the median, 5th and 95th percentiles of the simulated data (n ¼ 1000).

The haemodynamic effects of dexmedetomidine are biphasic depending on plasma concentration and speed of injection.2 Plasma concentrations of dexmedetomidine greater than 1000 pg ml
1 can be associated with peripheral vasoconstriction and bradycardia.13,28 Initial hypertension with intranasal administration of dexmedetomidine in

children has not been reported, although most studies have deferred blood pressure measurement before the onset of sedation.7,10 In the present study, no hypertension requiring treatment was observed by arterial pressure monitoring, albeit in anaesthetised patients. The relatively slow absorption and low maximum plasma concentration observed after nasal administration may explain why initial hypertension and bradycardia are not commonly reported with this route. Administration of intranasal dexmedetomidine has typically been reported in doses ranging from 1 to 3 mg kg
1.29,30 Published pharmacodynamic data in young children sedated with intranasal dexmedetomidine from our institution and others show adequate sedation for non-painful procedures at approximately 20e30 min after administration, with duration of approximately 30e60 min.7,10,29 We performed intranasal administration by atomisation, which has several potential advantages over simple nasal drop administration, such as uniform application throughout the nasal mucosa to maximise the area of absorption. However, a recent clinical study showed no difference between the two techniques with slow incremental administration of the nasal drops.31 Our i.v. dexmedetomidine group was similar to a study by Petroz and colleagues32 with 1 mg kg
1, which also infused this dose over 10 min. Our finding is close to their reported estimates of volume of 56.7 litre 70 kg
1. However, our CL estimate is larger than their reported value of 0.315 litre 70 kg
1 min
1. This could be because of differences in age (2e12 yr) and design (longer sampling time in Petroz). Moreover, our i.v. results are in agreement with those of Vilo and colleagues,33 which showed a CL of 0.794 litre 70 kg
1 min
1 and a terminal half-life of 139 min. Terminal elimination parameters in our study with i.v. dexmedetomidine are also consistent with prior studies of i.v. dexmedetomidine showing a 1.8 h terminal half-life.32 These similarities support our model of i.v. administration and intranasal application. Our population PK estimate of V1 is similar to Potts and colleagues,27 who also used an allometric model for i.v. dexmedetomidine in children. However, for CL, Q, and V2, our values are around 1.43, 1.37, and 1.93 times of their estimated values. Their data were pooled from four different studies including one in patients undergoing cardiac surgery and three others involving urologic, abdominal, plastic, craniofacial surgeries, and bronchoscopy. The estimated bioavailability of intranasal dexmedetomidine in this study is similar to the bioavailability estimated in healthy adults (82.0%) as reported by Yoo and colleagues.21 Other population PK parameters in the adult study were significantly lower than our results. In the adult study, both intranasal and i.v. doses were administered to a homogeneous population of healthy young males. In our study, separate patients received either intranasal or i.v. dose. Our patients were young children with congenital heart disease. Our sampling window was limited compared with the 10 h post-dose sample collection period in the adult study. All these factors might have contributed towards variation in the findings of the two studies. The variability in observed pharmacodynamic effects of intranasal dexmedetomidine reported in other studies may be partially explained by the variation in intranasal absorption as noted in the present study in children and the previous study by Iirola and colleagues14,21 in adults. This suggests that there exists a continued need for an improved delivery system and optimised formulation for intranasal administration of

Intranasal dexmedetomidine pharmacokinetics in children

dexmedetomidine. However, Vilo and colleagues33 reported that even with i.v. dosing, marked inter-individual variation in pharmacokinetic parameters exists in young children. A more detailed study of covariates, including pharmacogenetics, may be required to attain a more personalised sedation protocol.34 Intranasal administration of medications is thought to provide direct systemic absorption by bypassing hepatic firstpass metabolism, similar to buccal or sublingual administration. However, the nasal route may also produce effects by direct nose to brain delivery, potentially via the olfactory and trigeminal nerves, bypassing the bloodebrain barrier.35 Extracellular transport is thought to occur within minutes of administration for some intranasal medications including peptides, such as insulin, and lipophilic drugs, such as cocaine.18 We previously reported that intranasal dexmedetomidine produces sedation in a similar cohort of children with congenital heart disease within 26e28 min on average.7,10 Our current PK data (minimal lag time to absorption and peak plasma concentration in 47 min) suggests that a direct comparison of intranasal and buccal administration of dexmedetomidine with pharmacodynamic and plasma concentration measurements may be required to resolve whether a nose to brain component exists after intranasal administration. Our data contributes to the pharmacokinetic description of extravascular dexmedetomidine administration. With intranasal dexmedetomidine, we observed peak plasma concentrations exceeding those reported for buccal (5 min ‘swish and spit’) or orogastric administration (150 ml water) in adults (Fig 2).23 Peak plasma concentrations after buccal administration in adults occurred in 1.5 h on average and has not been studied in children. The buccal dexmedetomidine bioavailability determined by Anttila and colleagues23 is confounded by correction for the ‘amount of dexmedetomidine expelled from the mouth’ after 5 min measured to be a mean of 44% of the administered dexmedetomidine. Thus, their published bioavailability for buccal administration of 82% is probably an overestimate in a practical paediatric sedation setting. Peak plasma concentrations after oral administration have been reported to average 2.2 h with a lag time of detectable plasma concentrations of 0.6 h and a bioavailability of 16%. Despite this, successful oral dexmedetomidine sedation with doses of approximately 5 mg kg
1 has been reported.36

Limitations Our study was designed to characterise the absorption and peak plasma concentrations of dexmedetomidine after intranasal administration; however, in this clinical setting, the terminal elimination half-life could not be determined because of termination of sampling at the initiation of CPB which occurred approximately 2 h after the dexmedetomidine dose. This may have resulted in higher CL estimates. The metabolism and CL of i.v. dexmedetomidine has been well characterised in the postoperative ICU in patients after congenital cardiac surgery.11,37 Using our experimental model, we were unable to study i.v. and intranasal PK in the same patients and instead had to use different, though similar, individuals. Another limitation of our study was the administration to supine anaesthetised, tracheally intubated patients. Awake, spontaneously breathing patients sitting erect may achieve different results. While the volume of medication atomised was minimised, mucosal saturation and

-

1063

oropharyngeal runoff cannot be excluded.38 Because of the complexity of cardiac surgical preparation, sternotomy, and cardiac dissection, we could not include haemodynamic data in this study. However, before surgical manipulation, no haemodynamic instability requiring treatment was observed. Future studies should also compare the pharmacokinetic profile of intranasal dexmedetomidine with simple nasal drop administration and by small volume buccal (or sublingual) administration in children.31 We had one cyanotic patient in the 1 mg kg
1 intranasal dexmedetomidine group. Observed Cmax, Tmax, and estimated PK parameters in this cyanotic patient were 70e90% compared with values for non-cyanotic patient in the same dose group. However, no statistical conclusion can be drawn because of the small sample size. No significant difference was observed in estimated PK parameters of cyanotic and non-cyanotic patients in the 1 mg kg
1 i.v. dexmedetomidine group. The population PK parameters in our study were scaled with body weight. There is a strong theoretical basis of using allometric scaling to account for the influence of body size on PK parameters.20 Given the limited sample size, we did not evaluate which body size metric (body weight or body suface area (BSA)) would better correlate with PK parameters, but applied allometry theory. Future studies with more patients and in a younger age population are warranted to better understand the best body size metric associated with dexmedetomidine PK. We analysed EGA for covariate analysis and it was not found to be a significant covariate. However, the CL of most drugs reaches adult value by 1 yr of age20 and we had very few patients younger than 1 yr. Future studies with younger patients will help understand the effect of maturation on dexmedetomidine PK.

Conclusions Mean arterial plasma concentrations of dexmedetomidine in infants and toddlers approached 100 pg ml
1, the low end reported for sedative efficacy, within 20 min of an atomised intranasal administration of 1 mg kg
1. Doubling the dose to 2 mg kg
1 reached this plasma concentration within 10 min and achieved almost twice the peak concentration. Peak plasma concentrations with both doses were reached within 47 min of intranasal administration, with an overall bioavailability of 84%. The concentration time profile was best described by a two-compartment model by population pharmacokinetic modelling.

Authors’ contributions Study design: J.W.M., M.M., B.L.L., J.P.S., A.V., A.W.L. Patient recruitment: J.W.M., J.R.P., J.P.S. Data collection: J.W.M., J.E.L., J.N.P., J.R.P., J.P.S., A.L. Data analysis: J.W.M., R.B., M.D., A.V., A.W.L. Data interpretation: J.W.M., R.B., M.D., B.L.L., A.V., A.W.L. Drafting and revising the manuscript: J.W.M., R.B., M.D., J.N.P., B.L.L., A.V., A.W.L. Approval of the final version: all authors.

Acknowledgements The authors would like to thank Megan Kalin MS, Kristie Geisler, and Millicent Frimpong-Manso for their assistance as research coordinators for this study and Cara L. Sparks, MD and Maria E Ashton MS RPh MBA for editorial assistance.

1064

-

Miller et al.

Declaration of interest None declared. 11.

Supplementary material Supplementary data related to this article can be found at https://doi.org/10.1016/j.bja.2018.01.035.

12.

Funding Children’s Heart Association of Cincinnati and Department of Anesthesiology, Cincinnati Children’s Hospital Medical Center. Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health (UL1 TR001425). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References 1. Sulton C, McCracken C, Simon HK, et al. Pediatric procedural sedation using dexmedetomidine: a report from the Pediatric Sedation Research Consortium. Hosp Pediatr 2016; 6: 536e44 2. Mahmoud M, Mason KP. Dexmedetomidine: review, update, and future considerations of paediatric perioperative and periprocedural applications and limitations. Br J Anaesth 2015; 115: 171e82 3. Kandil A, Subramanyam R, Hossain MM, et al. Comparison of the combination of dexmedetomidine and ketamine to propofol or propofol/sevoflurane for drug-induced sleep endoscopy in children. Paediatr Anaesth 2016; 26: 742e51 4. Mahmoud M, Jung D, Salisbury S, et al. Effect of increasing depth of dexmedetomidine and propofol anesthesia on upper airway morphology in children and adolescents with obstructive sleep apnea. J Clin Anesth 2013; 25: 529e41 5. Yuen VM, Hui TW, Irwin MG, et al. A randomised comparison of two intranasal dexmedetomidine doses for premedication in children. Anaesthesia 2012; 67: 1210e6 6. Miller J, Xue B, Hossain M, Zhang MZ, Loepke A, Kurth D. Comparison of dexmedetomidine and chloral hydrate sedation for transthoracic echocardiography in infants and toddlers: a randomized clinical trial. Paediatr Anaesth 2016; 26: 266e72 7. Miller JW, Divanovic AA, Hossain MM, Mahmoud MA, Loepke AW. Dosing and efficacy of intranasal dexmedetomidine sedation for pediatric transthoracic echocardiography: a retrospective study. Can J Anaesth 2016; 63: 834e41 8. Miller JW, Loepke A, Bernard P. Nasal dexmedetomidine sedation for pediatric transthoracic echocardiography; select abstracts from the 18th annual update on pediatric and congenital cardiovascular disease, Scottsdale, Arizona, February 11e15, 2015. World J Pediatr Congenit Heart Surg 2015; 6: 153e88 9. Yuen VM, Hui TW, Irwin MG, Yao TJ, Wong GL, Yuen MK. Optimal timing for the administration of intranasal dexmedetomidine for premedication in children. Anaesthesia 2010; 65: 922e9 10. Miller JW, Ding L, Gunter JB, et al. Comparison of intranasal dexmedetomidine and oral pentobarbital sedation for transthoracic echocardiography in infants and toddlersda

13.

14.

15.

16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

27.

prospective, randomized, double-blind trial. Anesth Analg 2018 Jan 23. https://doi.org/10.1213/ANE.0000000000002791 [Epub ahead of print] PubMed PMID: 29369091 Su F, Gastonguay MR, Nicolson SC, DiLiberto M, OcampoPelland A, Zuppa AF. Dexmedetomidine pharmacology in neonates and infants after open heart surgery. Anesth Analg 2016; 122: 1556e66 Liu HC, Lian QQ, Wu FF, et al. Population pharmacokinetics of dexmedetomidine after short intravenous infusion in Chinese children. Eur J Drug Metab Pharmacokinet 2017; 42: 201e11 Potts AL, Anderson BJ, Holford NH, Vu TC, Warman GR. Dexmedetomidine hemodynamics in children after cardiac surgery. Paediatr Anaesth 2010; 20: 425e33 Iirola T, Vilo S, Manner T, et al. Bioavailability of dexmedetomidine after intranasal administration. Eur J Clin Pharmacol 2011; 67: 825e31 Dawes J, Myers D, Gorges M, Zhou G, Ansermino JM, Montgomery CJ. Identifying a rapid bolus dose of dexmedetomidine (ED50) with acceptable hemodynamic outcomes in children. Paediatr Anaesth 2014; 24: 1260e7 Simera I, Moher D, Hirst A, Hoey J, Schulz KF, Altman DG. Transparent and accurate reporting increases reliability, utility, and impact of your research: reporting guidelines and the EQUATOR network. BMC Med 2010; 8: 24 Reynolds J, Rogers A, Medellin E, Guzman JA, Watcha MF. A prospective, randomized, double-blind trial of intranasal dexmedetomidine and oral chloral hydrate for sedated auditory brainstem response (ABR) testing. Paediatr Anaesth 2016; 26: 286e93 Dhuria SV, Hanson LR, Frey 2nd WH. Intranasal delivery to the central nervous system: mechanisms and experimental considerations. J Pharm Sci 2010; 99: 1654e73 Matuszewski BK, Constanzer ML, Chavez-Eng CM. Strategies for the assessment of matrix effect in quantitative bioanalytical methods based on HPLC-MS/MS. Anal Chem 2003; 75: 3019e30 Anderson BJ, Holford NH. Tips and traps analyzing pediatric PK data. Paediatr Anaesth 2011; 21: 222e37 Yoo H, Iirola T, Vilo S, et al. Mechanism-based population pharmacokinetic and pharmacodynamic modeling of intravenous and intranasal dexmedetomidine in healthy subjects. Eur J Clin Pharmacol 2015; 71: 1197e207 Bhatt M, Kennedy RM, Osmond MH, et al. Consensusbased recommendations for standardizing terminology and reporting adverse events for emergency department procedural sedation and analgesia in children. Ann Emerg Med 2009; 53: 426e35 Anttila M, Penttila J, Helminen A, Vuorilehto L, Scheinin H. Bioavailability of dexmedetomidine after extravascular doses in healthy subjects. Br J Clin Pharmacol 2003; 56: 691e3 Yuen VM, Irwin MG, Hui TW, Yuen MK, Lee LH. A doubleblind, crossover assessment of the sedative and analgesic effects of intranasal dexmedetomidine. Anesth Analg 2007; 105: 374e80 Ebert TJ, Hall JE, Barney JA, Uhrich TD, Colinco MD. The effects of increasing plasma concentrations of dexmedetomidine in humans. Anesthesiology 2000; 93: 382e94 Su F, Nicolson SC, Zuppa AF. A dose-response study of dexmedetomidine administered as the primary sedative in infants following open heart surgery. Pediatr Criti Care Med 2013; 14: 499e507 Potts AL, Anderson BJ, Warman GR, Lerman J, Diaz SM, Vilo S. Dexmedetomidine pharmacokinetics in pediatric

Intranasal dexmedetomidine pharmacokinetics in children

28.

29.

30.

31.

32.

intensive careda pooled analysis. Paediatr Anaesth 2009; 19: 1119e29 Hammer GB, Drover DR, Cao H, et al. The effects of dexmedetomidine on cardiac electrophysiology in children. Anesth Analg 2008; 106: 79e83 Li BL, Ni J, Huang JX, Zhang N, Song XR, Yuen VM. Intranasal dexmedetomidine for sedation in children undergoing transthoracic echocardiography study-a prospective observational study. Paediatr Anaesth 2015; 25: 891e6 Behrle N, Birisci E, Anderson J, Schroeder S, Dalabih A. Intranasal dexmedetomidine as a sedative for pediatric procedural sedation. J Pediatr Pharmacol Ther 2017; 22: 4e8 Li BL, Zhang N, Huang JX, et al. A comparison of intranasal dexmedetomidine for sedation in children administered either by atomiser or by drops. Anaesthesia 2016; 71: 522e8 Petroz GC, Sikich N, James M, et al. A phase I, two-center study of the pharmacokinetics and pharmacodynamics of dexmedetomidine in children. Anesthesiology 2006; 105: 1098e110

-

1065

33. Vilo S, Rautiainen P, Kaisti K, et al. Pharmacokinetics of intravenous dexmedetomidine in children under 11 yr of age. Br J Anaesth 2008; 100: 697e700 34. Anderson BJ, Meakin GH. Scaling for size: some implications for paediatric anaesthesia dosing. Paediatr Anaesth 2002; 12: 205e19 35. Ruigrok MJ, de Lange EC. Emerging insights for translational pharmacokinetic and pharmacokineticpharmacodynamic studies: towards prediction of nose-to-brain transport in humans. AAPS J 2015; 17: 493e505 36. Konia MR. Oral dexmedetomidine for preoperative sedation in an adult uncooperative autistic patient. J Clin Anesth 2016; 34: 29e31 37. Su F, Nicolson SC, Gastonguay MR, et al. Population pharmacokinetics of dexmedetomidine in infants after open heart surgery. Anesth Analg 2010; 110: 1383e92 38. Wolfe TR, Braude DA. Intranasal medication delivery for children: a brief review and update. Pediatrics 2010; 126: 532e7 Handling editor: A.R. Absalom