Feature Review Articles
`
`Dexmedetomidine: Applications in pediatric critical care and
`pediatric anesthesiology
`
`Joseph D. Tobias, MD
`
`Objective: To provide a general descriptive account of the
`end-organ effects of dexmedetomidine and to provide an evi-
`dence-based review of the literature regarding its use in infants
`and children.
`Data Source: A computerized bibliographic search of the lit-
`erature regarding dexmedetomidine.
`Main Results: The end-organ effects of dexmedetomidine have
`been well studied in animal and adult human models. Adverse
`cardiovascular effects include occasional episodes of bradycardia
`with rare reports of sinus pause or cardiac arrest. Hypotension
`has also been reported as well as hypertension, the latter thought
`to be due to peripheral ␣2B agonism with peripheral vasoconstric-
`tion. Although dexmedetomidine has no direct effects on myocar-
`dial function, decreased cardiac output may result from changes
`in heart rate or increases in afterload. There are somewhat
`conflicting reports in the literature regarding its effects on ven-
`tilatory function, with some studies (both human and animal)
`suggesting a mild degree of respiratory depression, decreased
`minute ventilation, and decreased response to CO2 challenge
`
`whereas others demonstrate no effect. The central nervous sys-
`tem effects include sedation and analgesia with prevention of
`recall and memory at higher doses. Dexmedetomidine may also
`provide some neuroprotective activity during periods of ischemia.
`Applications in infants and children have included sedation during
`mechanical ventilation, prevention of emergence agitation follow-
`ing general anesthesia, provision of procedural sedation, and the
`prevention of withdrawal following the prolonged administration
`of opioids and benzodiazepines.
`Conclusions: The literature contains reports of the use of
`dexmedetomidine in approximately 800 pediatric patients. Given
`its favorable sedative and anxiolytic properties combined with its
`limited effects on hemodynamic and respiratory function, there
`is growing interest in and reports of its use in the pediatric
`population in various clinical scenarios. (Pediatr Crit Care Med
`2007; 8:115–131)
`KEY WORDS: dexmedetomidine; ␣2-adrenergic agonist; opioid
`tolerance and withdrawal; emergence delirium; procedural-
`sedation
`
`D exmedetomidine (Precedex,
`
`Hospira Worldwide, Lake For-
`est, IL) is the pharmacologi-
`cally active dextro-isomer of
`medetomidine. It exerts its physiologic ef-
`fects via ␣2-adrenergic receptors. The ␣2-
`adrenergic agonists are subclassified into
`three groups: imidazolines, phenylethyl-
`amines, and oxalozepines. Dexmedetomi-
`dine and clonidine are members of the
`imidazole subclass, which exhibits a high
`ratio of specificity for the ␣2 vs. the ␣1
`receptor (Fig. 1). Clonidine exhibits an
`␣2:␣1 specificity ratio of 200:1 whereas
`that of dexmedetomidine is 1600:1,
`thereby making it a complete agonist at
`the ␣2-adrenergic receptor (1). Dexme-
`detomidine has a short half-life (2–3 hrs
`
`From the Departments of Anesthesiology and Pe-
`diatrics, University of Missouri, Columbia, MO.
`The author has consulted for and received hono-
`raria/speaking fees from Hospira.
`Copyright © 2007 by the Society of Critical Care
`Medicine and the World Federation of Pediatric Inten-
`sive and Critical Care Societies
`DOI: 10.1097/01.PCC.0000257100.31779.41
`
`Pediatr Crit Care Med 2007 Vol. 8, No. 2
`
`vs. 12–24 hrs for clonidine) and is com-
`mercially available for intravenous ad-
`ministration. An epidural clonidine for-
`mulation, although not marketed for
`intravenous administration, has been
`used for this purpose in clinical practice
`without consequences. Dexmedetomi-
`dine’s end-organ effects are mediated via
`postsynaptic ␣2-adrenergic receptors and
`subsequent activation of a pertussis tox-
`in-sensitive guanine nucleotide regula-
`tory protein (G protein) (2), which results
`in inhibitory feedback and decreased ac-
`tivity of adenylyl cyclase (3). A reduction
`of intracellular cyclic adenosine mono-
`phosphate and intracellular cyclic adeno-
`sine monophosphate-dependent protein
`kinase activity results in the dephosphor-
`ylation of ion channels (4). Alterations in
`ion channel function, ion translocation,
`and membrane conductance lead to de-
`creased neuronal activation and the clin-
`ical effects of sedation and anxiolysis (5).
`Centrally acting ␣2-adrenergic agonists
`also activate receptors in the medullary
`vasomotor center, reducing norepineph-
`rine with a resultant central sympatho-
`
`lytic effect leading to decreased heart rate
`(HR) and blood pressure (BP). As the cen-
`tral presynaptic ␣2A-adrenergic receptor
`is a negative feedback receptor, agonists
`at this receptor result in decreased cate-
`cholamine release from the nerve termi-
`nal. Central nervous system stimulation
`of parasympathetic outflow and inhibi-
`tion of sympathetic outflow from the lo-
`cus ceruleus in the brainstem play a
`prominent role in the sedation and anxi-
`olysis produced by these agents. De-
`creased noradrenergic output from the
`locus ceruleus allows for increased firing
`of inhibitory neurons, most importantly
`the ␥-aminobutyric acid system (6 – 8).
`Primary analgesic effects and potentia-
`tion of opioid-induced analgesia result
`from the activation of ␣2-adrenergic re-
`ceptors in the dorsal horn of the spinal
`cord and the inhibition of substance P
`release. These interactions with central
`nervous system and spinal cord ␣2-
`adrenergic receptors mediate dexmedeto-
`midine’s primary physiologic effects in-
`cluding sedation, anxiolysis, analgesia, a
`decrease of the minimum alveolar con-
`
`115
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1061 – Page 115
`
`

`

`Figure 1. Representation of the chemical struc-
`ture of clonidine and dexmedetomidine, ␣2-
`adrenergic agonists of the imidazole subclass,
`which exhibit a high ratio of specificity for the ␣2
`vs. the ␣1 receptor.
`
`inhalational anesthetic
`centration of
`agents, decreased renin and vasopressin
`levels leading to diuresis, blunting of the
`sympathetic nervous system, and lower-
`ing of HR and BP (Fig. 2) (9, 10).
`Currently, dexmedetomidine’s only
`Food and Drug Administration (FDA)-
`approved indication is the provision of
`short-term sedation (⬍24 hrs) in adult
`patients in the intensive care unit (ICU)
`setting who are initially intubated and
`receiving mechanical ventilation (11). It
`is available in a water-soluble solution
`without the addition of lipid or propylene
`glycol and is not associated with pain
`following intravenous administration.
`There are no active or toxic metabolites.
`Given its favorable physiologic effects
`combined with a limited adverse effect
`profile reported to date, there is increas-
`ing use of this agent in the pediatric
`population. This article reviews the basic
`pharmacology of dexmedetomidine,
`its
`end-organ effects and adverse effect pro-
`file, and reports from the literature re-
`garding its use in various clinical scenar-
`ios in infants and children.
`
`PHARMACOKINETICS
`
`In healthy adult volunteers, dexme-
`detomidine’s pharmacokinetic profile in-
`
`Figure 2. The physiologic end-organ effects of dexmedetomidine.
`
`cludes a rapid distribution phase (distri-
`bution half-life of 6 mins), an elimination
`half-life of 2 hrs, and a steady-state vol-
`ume of distribution of 118 L (12). In the
`dosing range of 0.2– 0.7 ␮g/kg/hr deliv-
`ered via continuous intravenous infusion
`for up to 24 hrs, the pharmacokinetics
`are linear. Dexmedetomidine is 94% pro-
`tein bound to serum albumin and ␣1-
`glycoprotein. It undergoes hepatic me-
`tabolism with limited unchanged drug
`excreted in the urine or stool.
`Cunningham et al. (13) evaluated
`dexmedetomidine pharmacokinetics fol-
`lowing administration (0.6 ␮g/kg infused
`over 10 mins) in five adults with severe
`
`hepatic failure and compared the results
`with five age-matched controls with nor-
`mal hepatic function. When compared
`with age-matched controls with normal
`hepatic function, there was an increased
`volume of distribution at steady state (3.2
`vs. 2.2 L/kg, p ⬍ .05), an increased elim-
`ination half-life (7.5 vs. 2.6 hrs, p ⬍ .05),
`and a decreased clearance (0.32 vs. 0.64
`L/hr/kg; p ⬍ .05) in patients with hepatic
`dysfunction. In a subsequent study in six
`adult patients with severe renal disease
`(24-hr creatinine clearance ⱕ30 mL/min)
`who were not receiving dialysis, there
`was no statistically significant difference
`between renal disease and control pa-
`
`116
`
`Pediatr Crit Care Med 2007 Vol. 8, No. 2
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1061 – Page 116
`
`

`

`tients in the volume of distribution at
`steady state (1.81 ⫾ 0.55 vs. 1.54 ⫾ 0.08
`L/kg) or the elimination clearance
`(12.5 ⫾ 4.6 vs. 8.9 ⫾ 0.7 mL/min/kg)
`(14). However, the elimination half-life
`was decreased with renal disease (113.4 ⫾
`11.3 mins vs. 136.5 ⫾ 13.0 mins, p ⬍
`.05). Despite the shorter elimination half-
`life, there was prolonged sedation in pa-
`tients with renal disease. The 1-hr postin-
`fusion visual analog score of sedation
`(scale of 0 to 100) was 49.2 ⫾ 25.4 in
`patients with renal disease compared
`with 26.2 ⫾ 18.3 in patients with normal
`function (p ⬍ .05). The authors
`renal
`speculated that the increased sedation
`with renal failure resulted from decreased
`protein binding and an increased free
`fraction of the drug. Venn et al. (15) eval-
`uated the impact of acute surgical inter-
`vention and critical illness on dexmedeto-
`midine pharmacokinetics in ten adult
`patients following complex abdominal or
`pelvic surgical procedures. Dexmedeto-
`midine administration included a loading
`dose of 0.4 ␮g/kg over 10 mins followed
`by an infusion of 0.7 ␮g/kg/hr. When
`compared with data from healthy volun-
`teers, there was no difference in half-life,
`volume of distribution, or clearance.
`Data regarding dexmedetomidine
`pharmacokinetics in the pediatric popu-
`lation have been presented in one recent
`manuscript and two abstracts (16 –18).
`Petroz et al. (16) randomized 36 children,
`ranging in age from 2 to 12 yrs, to receive
`dexmedetomidine infused for 10 mins at
`2, 4, or 6 ␮g/kg/hr (0.33, 0.6, and 1 ␮g/
`kg). Using a two-compartment model,
`they reported that the pharmacokinetics
`of dexmedetomidine in children are sim-
`ilar to adults with no dose-dependent ki-
`netics, protein binding of 92.6%, weight-
`adjusted total body clearance of 13 mL/
`kg/min, a volume of distribution of the
`peripheral compartment of 1.0 L/kg, and
`a terminal elimination half-life of 1.8 hrs.
`Rodarte et al. (17) administered a contin-
`uous infusion in a dose ranging from 0.2
`to 0.7 ␮g/kg/hr for 8 –24 hrs to ten chil-
`dren (0.3–7.9 yrs of age) following cardiac
`procedures (n ⫽ 9) or craniofacial proce-
`dures (n ⫽ 1). Using a two-compartment
`model, they reported a volume of distri-
`bution of 1.53 ⫾ 0.37 L/kg, a clearance of
`0.57 ⫾ 0.14 L/kg/hr (approximately 9.5
`mL/kg/min), and a terminal elimination
`half-life of 2.65 ⫾ 0.88 hrs. They com-
`mented that their data demonstrated that
`the pharmacokinetics of dexmedetomi-
`dine in children were predictable and
`consistent with results reported in adults.
`
`Pediatr Crit Care Med 2007 Vol. 8, No. 2
`
`The final pharmacokinetic study in chil-
`dren included infants, ranging in age
`from 1 to 24 months, following surgery
`for congenital heart disease (18). The au-
`thors reported a median clearance of 27.2
`mL/kg/min, peripheral volume of distri-
`bution of 2.5 L/kg, and terminal elimina-
`tion half-life of 83 mins. They concluded
`that infants appear to clear dexmedeto-
`midine more quickly than adults or older
`children.
`
`END-ORGAN EFFECTS OF
`DEXMEDETOMIDINE
`
`Cardiovascular and
`Hemodynamic Effects
`
`Heart Rate, Blood Pressure, Cardiac
`Output, and Myocardial Contractility. Hy-
`potension and bradycardia have been re-
`ported in adult patients, especially in the
`presence of comorbid cardiac disease,
`when administered with other medica-
`tions that possess negative chronotropic
`effects or following large or rapid bolus
`doses. In healthy adult volunteers, there
`is a biphasic effect following dexmedeto-
`midine with an initial increase in systolic
`blood pressure (sBP) and a reflex decrease
`in HR followed by a stabilization of sBP
`and HR at values below the baseline (19).
`Stimulation of peripheral postsynaptic
`␣2B-adrenergic receptors results in vaso-
`constriction and the initial increase in
`sBP, whereas the eventual decrease in BP
`and HR results from central presynaptic
`␣2A-adrenergic receptor stimulated sym-
`patholysis.
`In healthy, adult volunteers, dexme-
`detomidine doses of 0.25, 0.5, 1.0 and 2.0
`␮g/kg administered over 2 mins resulted
`in a decrease from baseline of the mean
`arterial pressure (MAP) at 60 mins of
`14%, 16%, 23%, and 27% (19). Following
`a dose of 1 ␮g/kg, cardiac output, mea-
`sured by thoracic bioimpedance, was
`81 ⫾ 13% of baseline at 1 min, 88 ⫾ 14%
`of baseline at 10 mins, and 91 ⫾ 11% of
`baseline at 60 mins. With a dose of 2
`␮g/kg, cardiac output was 58 ⫾ 32% of
`baseline at 1 min, 76 ⫾ 33% of baseline
`at 10 mins, and 85 ⫾ 28% of baseline at
`60 mins.
`for adverse hemody-
`The potential
`namic effects with dexmedetomidine in
`patients with comorbid features is illus-
`trated in an adult ICU population of 98
`cardiac and general surgery patients who
`received dexmedetomidine for sedation
`during mechanical ventilation (11).
`
`Dexmedetomidine was dosed as a bolus
`dose of 1 ␮g/kg over 10 mins followed by
`an infusion of 0.2– 0.7 ␮g/kg/hr. Hypo-
`tension (MAP ⱕ60 mm Hg or a ⱖ30%
`decrease from baseline) occurred in 18 of
`66 patients. Eleven of the episodes oc-
`curred during the bolus. Hypertension
`was noted in six of the 66 patients during
`the loading dose. Although no morbidity
`or mortality was noted, the infusion was
`temporarily (n ⫽ 3) or permanently (n ⫽
`3) discontinued, and treatment with at-
`ropine (n ⫽ 2) or temporary cardiac pac-
`ing (n ⫽ 4) was necessary.
`Bradycardia and sinus arrest have
`been reported with dexmedetomidine
`(20, 21). In a study combining dexme-
`detomidine with propofol to induce anes-
`thesia, two of the first four patients had
`brief and self-limited sinus arrest after
`laryngoscopy (20). Dexmedetomidine was
`administered as a bolus dose of 1 ␮g/kg
`over 15 mins followed by an infusion of
`0.4 ␮g/kg/hr resulting in the administra-
`tion of an average dose of 1.47 ␮g/kg
`before anesthetic induction with propo-
`fol. The protocol was amended (decrease
`of the dexmedetomidine dose to 0.7
`␮g/kg over 15 mins followed by an infu-
`sion of 0.27 ␮g/kg/hr), and no subsequent
`problems were noted.
`We reported bradycardia in a 5-wk-old
`infant with trisomy 21 who was receiving
`dexmedetomidine for sedation during
`mechanical ventilation (22). Concomi-
`tant medications included digoxin for the
`treatment of chronic congestive heart
`failure due to an unrepaired atrioventric-
`ular canal defect. Twelve hours after the
`initiation of the dexmedetomidine infu-
`sion, the infant’s HR decreased to 40 –50
`beats/min with a stable BP. The dexme-
`detomidine infusion was discontinued
`without other therapy, and the HR re-
`turned to baseline within 60 mins.
`In a study of 192 patients with Amer-
`ican Society of Anesthesiologists ratings
`of 1 or 2, randomized to receive either
`intramuscular dexmedetomidine and in-
`travenous saline, intramuscular dexme-
`detomidine and intravenous fentanyl, or
`intramuscular midazolam and intrave-
`nous fentanyl, followed by maintenance
`anesthesia (70% nitrous oxide in 30%
`oxygen, fentanyl, and either enflurane or
`isoflurane),
`intraoperative bradycardia
`and hypotension were significantly more
`common in the patients who received
`dexmedetomidine compared with those
`receiving midazolam (23). In one patient,
`bradycardia (HR 35 beats/min) required
`pharmacologic therapy. Khan et al. (24),
`
`117
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1061 – Page 117
`
`

`

`in a study of nine male volunteers assess-
`ing the effects of low (0.3 ng/mL) and
`high (0.6 ng/mL) dexmedetomidine
`plasma concentrations on isoflurane re-
`quirements, reported five hypotensive
`events in the low concentration group
`and seven in the concentration group.
`Interventions including crystalloid, crys-
`talloid and methoxamine, or atropine
`were necessary in five patients. The ma-
`jority of the hemodynamic events (75%)
`occurred at an end-tidal
`isoflurane of
`ⱖ1%.
`In a cohort of 80 children, ranging in
`age from 1 to 12 yrs, no clinically signif-
`icant hypotension or bradycardia oc-
`curred with the intraoperative adminis-
`tration of dexmedetomidine (0.5 ␮g/kg)
`during anesthesia at 1 minimum alveolar
`concentration with either desflurane or
`sevoflurane (25). However, there was a
`greater decrease in HR in patients anes-
`thetized with sevoflurane compared with
`those receiving desflurane (104 ⫾ 16 vs.
`120 ⫾ 17 beats/min, p ⬍ .01).
`Lowering of HR and thereby myocar-
`dial oxygen consumption may provide
`beneficial effects in patients with coro-
`nary artery disease. Talke et al. (26) ran-
`domized 24 adult patients undergoing
`vascular surgery to placebo or one of
`three plasma concentrations of dexme-
`detomidine: 0.15 ng/mL (low dose), 0.3
`ng/mL (medium dose), or 0.45 ng/mL
`(high dose). Dexmedetomidine was
`started 1 hr before anesthetic induction
`and continued for 48 hrs. Although there
`was an increased intraoperative need for
`atropine and/or phenylephrine with
`dexmedetomidine, no such difference was
`noted postoperatively. In the placebo
`group, there was an increased incidence of
`tachycardia (23 mins/hr) when compared
`with the low-dose (9 mins/hr, p ⫽ .006),
`medium-dose (0.5 mins/hr, p ⫽ .004),
`and high-dose (2.3 mins/hr, p ⫽ .004)
`dexmedetomidine groups. In an anec-
`dotal report, the negative chronotropic
`effect of dexmedetomidine was used as a
`therapeutic maneuver during off-pump
`coronary artery bypass surgery when
`tachycardia was unresponsive to ␤-adren-
`ergic blockade (27).
`The potential for significant negative
`chronotropic effects appears to be greater
`when dexmedetomidine is administered
`with medications that have negative
`chronotropic effects (propofol, succinyl-
`choline, digoxin, pyridostigmine) or dur-
`ing vagotonic procedures (laryngoscopy)
`(20 –22). Animal studies have not demon-
`strated direct effects on myocardial con-
`
`118
`
`tractility or intracellular calcium regula-
`tion (28). When studied in an isolated
`right ventricular papillary muscle prepa-
`ration, dexmedetomidine had no effect on
`the amplitude and time variables of iso-
`metric, isotonic, or zero-loaded-clamped
`twitches and intracellular calcium cur-
`rents (28).
`Sympathetic Nervous System and En-
`dogenous Catecholamine Release. Bio-
`chemical data from a cohort of eight
`adult postoperative patients demonstrate
`the sympatholytic effects of dexmedeto-
`midine (29). Following a 60-min dexme-
`detomidine infusion administered by a
`computer-controlled infusion protocol
`to achieve a plasma concentration of
`600 pg/mL, the plasma norepinephrine
`concentration decreased from 2.1 ⫾ 0.8
`to 0.7 ⫾ 0.3 nmol/L, the plasma epi-
`nephrine concentration decreased from
`0.7 ⫾ 0.5 to 0.2 ⫾ 0.2 nmol/L, HR
`decreased from 76 ⫾ 15 to 64 ⫾ 11
`beats/min, and sBP decreased from 158
`⫾ 23 to 140 ⫾ 23 mm Hg. The same
`investigators evaluated changes in
`plasma and urinary catecholamines in
`41 adult patients undergoing vascular
`surgery (30). Dexmedetomidine was
`started intraoperatively and continued
`for the first 48 postoperative hours.
`When compared with patients receiving
`dexmedetomidine, plasma norepineph-
`rine concentrations were two to three
`times higher at the time of tracheal
`extubation and at 60 mins after arrival
`in the postanesthesia care unit than in
`the control group. Urinary normeta-
`nephrine levels increased significantly
`in the placebo group, whereas no
`change was noted in patients receiving
`dexmedetomidine.
`A similar sympatholytic effect has
`been demonstrated following the intraop-
`erative administration of dexmedetomi-
`dine to pediatric patients undergoing car-
`diopulmonary bypass and surgery for
`congenital heart disease (31). Muktar et
`al. (31) randomized 30 infants and chil-
`dren to placebo or dexmedetomidine (bo-
`lus of 10 ␮g/kg over 10 mins followed by
`an infusion of 0.5 ␮g/kg/hr), which was
`administered after anesthetic induction
`and placement of arterial and venous can-
`nulae. Although plasma cortisol, norepi-
`nephrine, epinephrine, and glucose con-
`centrations
`increased in both the
`dexmedetomidine and the placebo groups
`after sternotomy and following cardiopul-
`monary bypass, the increase was signifi-
`cantly less in patients receiving dexme-
`detomidine. Additionally, when weaning
`
`less so-
`from cardiopulmonary bypass,
`dium nitroprusside was required in pa-
`tients receiving dexmedetomidine (0.3 ⫾
`0.36 vs. 1.3 ⫾ 0.68 ␮g/kg/min, p ⬍ .05).
`No adverse effects were noted.
`In specific clinical scenarios such as
`hemorrhage, hypovolemia, or conges-
`tive heart failure, there is the potential
`for dexmedetomidine’s sympatholytic
`effect to be detrimental by offsetting
`the protective function of the sympa-
`thetic nervous system. In an animal
`model, Blake et al. (32) evaluated the
`effects of dexmedetomidine on the BP
`response during incremental decreases
`in intravascular blood volume induced
`by a gradual
`inflation of an inferior
`vena cava cuff. In control animals, the
`gradual reduction of intravascular vol-
`ume resulted in a progressive increase
`in HR with peripheral vasoconstriction
`to maintain MAP until cardiac index
`was approximately 40% of baseline, at
`which time there was failure of vaso-
`constriction and a decrease in MAP.
`Dexmedetomidine, administered intra-
`venously or directly into the fourth
`ventricle of the central nervous system,
`resulted in a decrease of HR and MAP
`from baseline and an earlier decompen-
`sation during inflation of the inferior
`vena cava cuff. Similar findings were
`reported in rabbits treated with doxo-
`rubicin to induce a chronic congestive
`heart failure and then subjected a re-
`duction of intravascular volume by in-
`flation of an inferior vena cava cuff (33).
`Myocardial Oxygen Consumption and
`Perioperative Ischemia. Clinical studies
`in adults have shown that the periopera-
`tive administrative of ␣2-adrenergic ago-
`nists may modify the incidence of adverse
`cardiovascular events including myocar-
`dial
`ischemia (34, 35). In an animal
`model of coronary artery stenosis,
`dexmedetomidine reduced blood flow in
`the nonischemic myocardium and in the
`ischemic epicardial layer; however, there
`was no effect on blood flow in the isch-
`emic mid-myocardial and subendocardial
`layers, thereby increasing the ischemic-
`nonischemic blood flow ratio (36). Myo-
`cardial oxygen demand also decreased
`with dexmedetomidine, thereby further re-
`ducing the ischemic myocardium’s oxygen
`deficiency
`Similar findings were reported by
`Willigers et al. (37) in their animal model
`using graded coronary stenosis to pro-
`duce lactate release from the poststenotic
`myocardium. Lactate production oc-
`curred in zero of eight dogs receiving
`
`Pediatr Crit Care Med 2007 Vol. 8, No. 2
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1061 – Page 118
`
`

`

`dexmedetomidine compared with four of
`seven in the control group (p ⫽ .03).
`With dexmedetomidine,
`lactate release
`was 46% less during emergence from an-
`esthesia, and the endocardial/epicardial
`blood flow ratio increased by 35% com-
`pared with the control group. Decreased
`levels of plasma epinephrine (158 vs.
`1909 pg/mL) and norepinephrine (126 vs.
`577 pg/mL) and decreased HR (123 ⫾ 6
`vs. 160 ⫾ 10 beats/min) were noted. The
`authors postulated that this may account
`for the anti-ischemic effect of dexmedeto-
`midine.
`Additional potentially protective ef-
`fects of dexmedetomidine on myocar-
`dial performance include preservation
`of myocardial dysfunction following
`ischemia and prevention of catechol-
`amine-induced arrhythmogenesis (38,
`39). Hypoxia followed by reoxygenation
`exposes the myocardium to an oxidative
`stress, resulting in tissue injury/death
`and myocardial dysfunction. In rats ex-
`posed to 60 mins of hypoxia, dexme-
`detomidine administered before but not
`after hypoxia significantly improved left
`ventricular-developed pressure after
`reoxygenation (38). The effect was
`blocked by yohimbine, an ␣2-adrenergic
`antagonist. In a separate study, dexme-
`detomidine increased the dysrhythmo-
`genic dose of epinephrine in halothane-
`anesthetized dogs (mean dose of 3 ␮g/kg/
`min in control animals vs. 6 ␮g/kg/min in
`animals receiving dexmedetomidine) (39).
`Pulmonary Vascular Resistance.
`There is limited information regarding
`dexmedetomidine’s effects on the pulmo-
`nary vasculature and pulmonary vascular
`resistance (PVR). In six instrumented
`sheep, dexmedetomidine (2 ␮g/kg over 1
`min) transiently increased mean pulmo-
`nary artery pressure (MPAP) and PVR (40).
`PVR increased from a baseline of 81 ⫾ 16
`dynes·sec·cm⫺5 to a maximum of 141 ⫾
`27 dynes·sec·cm⫺5, whereas MPAP in-
`creased from 15 ⫾ 1 to 18⫾ 0 mm Hg.
`MAP also increased (86 ⫾ 2 to 93⫾ 6
`mm Hg), as did systemic vascular resis-
`tance (1416 ⫾ 83 to 1889 ⫾ 64
`dynes·sec·cm⫺5). There was no change in
`pulmonary artery occlusion pressure.
`Similar transient pulmonary hemody-
`namic changes have been reported in
`healthy human volunteers with graded
`dexmedetomidine infusions to a plasma
`concentration of 1.9 ng/mL (19). Given
`the potential impact of these findings,
`especially in patients with elevated MPAP
`
`Pediatr Crit Care Med 2007 Vol. 8, No. 2
`
`or PVR, future studies are needed to de-
`fine these effects.
`
`Respiratory Effects
`
`Ventilation. The ventilatory effects of
`increasing doses of dexmedetomidine
`(0.25, 0.5, 1, and 2 ␮g/kg over 2 mins)
`have been evaluated in healthy adult vol-
`unteers by measurement of oxygen satu-
`ration, PaCO2, CO2 response curves with
`CO2 rebreathing, and respiratory induc-
`tance plethysmography (10, 41). With
`doses of 1 or 2 ␮g/kg, PaCO2 increased
`significantly with a maximum effect
`noted 10 mins following the dose. The
`mean PaCO2 increase from baseline was
`5.0 and 4.2 mm Hg with the 1.0 and 2.0
`␮g/kg doses, respectively. The effect per-
`sisted for 60 mins following 1 ␮g/kg and
`for 105 mins following 2 ␮g/kg. Follow-
`ing 2.0 ␮g/kg, minute ventilation de-
`creased from 8.7 ⫾ 0.7 to 6.3 ⫾ 1.5 L/min
`(p ⬍ .05). The decrease resulted predom-
`inantly from a decreased tidal volume
`with less effect on respiratory rate. Sig-
`nificant changes were also noted using
`CO2 response curves, as minute ventila-
`tion at an end-tidal CO2 of 55 mm Hg was
`depressed following the both the 1- and
`2-␮g/kg doses. The authors also noted
`short episodes of apnea and irregular
`breathing in some subjects, which oc-
`curred more commonly with the two
`highest doses (seven of ten patients with
`2 ␮g/kg and five of six patients with 1
`␮g/kg vs. one of six with either 0.5 ␮g/kg
`or 0.25 ␮g/kg). Respiratory inductance
`plethysmography was used to demon-
`strate that these problems were obstruc-
`tive and not central. Although oxygen
`saturation decreased with the obstructive
`episodes, the average room air oxygen
`saturation remained ⬎95% following all
`doses of dexmedetomidine. The oxygen
`saturation decrease was greatest at 10
`mins following 1 ␮g/kg (decrease from
`98.5 ⫾ 0.7% to 96.2 ⫾ 1.3%) and at 60
`mins following 2 ␮g/kg (decrease from
`98.3 ⫾ 0.8% to 95.4 ⫾ 1.2%). Similar
`respiratory effects have been demon-
`strated in experimental animals although
`a paradoxic effect has been noted with
`more of an effect on ventilation with 1 vs.
`10 ␮g/kg in one study and 10 or 30 ␮g/kg
`vs. 50 ␮g/kg in another (42– 44).
`Conflicting results were reported
`when comparing the respiratory effects of
`dexmedetomidine with remifentanil
`in
`six healthy adult volunteers (45). When
`compared with baseline, a remifentanil
`infusion to achieve a stepwise plasma
`
`concentration of 1, 2, 3, and 4 ng/mL
`resulted in respiratory depression mani-
`fested as a decrease in respiratory rate
`and minute ventilation, increased PaCO2,
`blunting of the CO2 response curve, and
`apnea with oxygen desaturation. During
`stepwise dexmedetomidine infusions to
`achieve plasma concentrations of 0.6, 1.2,
`1.8, and 2.4 ng/mL, there was an increase
`in respiratory rate, a decrease in the hy-
`popnea/apnea index, and no change in
`the end-tidal CO2 when compared with
`baseline values. With dexmedetomidine,
`some patients demonstrated a periodic in-
`crease in minute ventilation during CO2
`response curves (hypercapnic arousal) that
`correlated with changes in the Bispectral
`Index. The authors noted that similar
`changes occur during natural sleep and
`that these findings may result from
`dexmedetomidine’s mechanism of action
`in the locus ceruleus and its convergence
`on the natural sleep pathway. The au-
`thors concluded that dexmedetomidine
`stands apart from other sedatives in that
`it appears to be clinically safe from a
`respiratory point of view even in doses
`high enough to cause unresponsiveness.
`Similar findings were reported from an
`evaluation of the respiratory effects of
`dexmedetomidine (10 and 30 ␮g/kg) and
`alfentanil in an animal model (rats) (46).
`Neither dose of dexmedetomidine had an
`effect on PaO2, PaCO2, or pH, whereas the
`administration of alfentanil resulted in a
`decrease in pH and PaO2 and an increase in
`PaCO2. Dexmedetomidine had no additional
`effect when administered after alfentanil,
`and in fact, dexmedetomidine in a dose of
`30 ␮g/kg decreased the acidosis and hyper-
`capnia that occurred following alfentanil.
`Despite these findings, monitoring of respi-
`ratory function during the administration
`of dexmedetomidine in high-risk patients
`or those receiving other agents that may
`depress respiratory function appears war-
`ranted given the recent report of central
`apnea after a general anesthetic that in-
`cluded dexmedetomidine (47).
`Airway Reactivity. In mongrel dogs,
`the intravenous but not the inhalational
`administration of dexmedetomidine has
`been shown to prevent histamine-in-
`duced bronchoconstriction (48). Bron-
`choconstriction was provoked with aero-
`solized histamine, and its effect on airway
`caliber was evaluated using high-resolu-
`tion computed tomography with an eval-
`uation of airway cross-sectional area.
`Aerosolized histamine constricted the
`airways to 66 ⫾ 27% of baseline com-
`pared with 87 ⫾ 30.4% of baseline when
`
`119
`
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC – Exhibit 1061 – Page 119
`
`

`

`the animals were pretreated with intrave-
`nous dexmedetomidine.
`
`Central Nervous System Effects
`
`Sedation. Clinical studies in humans
`and experimental trials in both humans
`and animals have demonstrated the sed-
`ative effects of dexmedetomidine (9, 10,
`41, 49, 50). In ten healthy adult male
`volunteers, sequential 40-min infusions
`of dexmedetomidine were administered
`to achieve plasma concentrations of 0.5,
`0.8, 1.2, 2.0, 3.2, 5.0, and 8.0 ng/mL (45).
`The visual analog sedation score (0 ⫽
`very alert and 100 ⫽ very sedated) in-
`creased to 36 ⫾ 27 and 62 ⫾ 18 from a
`baseline of 0 with the first two targeted
`infusion levels (0.5 and 0.8 ng/mL). The
`two volunteers who received the highest
`incremental dose (calculated to achieve a
`plasma concentration of 8.0 ng/mL) were
`not arousable even with vigorous shak-
`ing. Picture recall and recognition were
`preserved during the lowest incremental
`infusion (0.5 ng/mL) but were 0% (0 of
`10) and 20% (2 of 10), respectively, with
`the second and third infusion levels (0.8
`and 1.2 ng/mL).
`Dexmedetomidine’s sedative response
`has been shown to have properties that
`parallel natural sleep (49, 50). Using
`functional magnetic resonance imaging
`(MRI), the blood oxygen level dependent
`signal, a correlate of local brain activity,
`changes with dexmedetomidine-induced
`sedation in a similar fashion to that seen
`during natural sleep (50). This is different
`from the pattern that occurs following
`the administration of midazolam. Using
`immunohistochemistry and in situ hy-
`bridization, dexmedetomidine has also
`been shown to induce a qualitatively sim-
`ilar pattern of c-fos expression in sleep-
`promoting brain nuclei of rats as that
`seen during nonrapid eye movement
`sleep (a decrease in the locus ceruleus
`and tuberomammillary nucleus and an
`increase in the ventrolateral nucleus)
`(49). These effects were attenuated by ati-
`pamezole, an ␣2-adrenergic antagonist,
`and did not occur in rats that lacked
`␣2-adrenergic receptors. These findings
`suggest a clinical advantage of dexmedeto-
`midine with its pattern of sedation parallel-
`ing natural sleep when compared with
`other agents (barbiturates, benzodiaz-
`epines, and propofol) commonly used for
`ICU sedation. These agents disrupt the nor-
`mal electroencephalographic patterns of
`sleep, and these effects may be responsible
`for the delirium seen in the ICU setting.
`
`120
`
`Given that delirium has been shown to be
`an independent risk factor of mortality in
`the adult ICU setting, avoidance of the dis-
`ruption of the natural sleep cycle