813
`
`M CLINICAL INVESTIGATIONS
`
`Anesthesiology
`78:813-820, 1993
`© 1993 AmericanSocicty of Anesthesiologists, Inc.
`J. B. Lippincott Company, Philadelphia
`
`The Pharmacokinetics and Hemodynamic Effects of
`Intravenous and Intramuscular Dexmedetomidine
`Hydrochloride in Adult Human Volunteers
`J. B, Dyck, M.D., F.R.C.P.C.,* M. Maze, M.B., Ch.B.,¢ C. Haack, A.N.,¢L. Vuorilehto, M.Sc.,§ S. L. Shafer, M.D.4
`
`Background: Dexmedetomidineis an a, agonist with poten-
`tial utility in clinical anesthesia for both its sedative and sym-
`patholytic properties.
`Methods: The pharmacokinetics and hemodynamic changes
`that occurred in ten healthy male volunteers were determined
`after administration of dexmedetomidine 2 yg/kg by intra-
`venousor intramuscular route in separate study sessions.
`Results: The intramuscular absorption profile of dexmede-
`tomidine, as determined by deconvolution of the observed
`concentrations against the unit disposition function derived
`from the intravenousdata, was biphasic. The percentage bio-
`availability of dexmedetomidine administered intramuscu-
`larly compared with the same dose administered intrave-
`nously was 73 + 11% (mean + SD). After intramuscular ad-
`ministration, the mean time to peak concentration was 12 min
`(range 2-60 min) and the mean peak concentration was 0.81
`+ 0.27 ng/ml. After intravenous administration of dexmede-
`tomidine, there were biphasic changes in blood pressure.
`During the 5-min intravenous infusion of 2 ng/kg dexmede-
`tomidine, the meanarterial pressure (MAP) increased by 22%
`and heart rate (HR) declined by 27% from baseline values.
`Over the 4 h after the infusion, MAP declined by 20% from
`baseline and HR rose to 5% below baseline values. The he-
`modynamicprofile did not show acute alterationsafter intra-
`muscular administration. During the 4 h after intramuscular
`administration, MAP declined by 20% and HR declined by 10%.
`
`
`* Assistant Professor, Department of Anesthesia, VA Medical Center,
`San Diego; and DepartmentofAnesthesia, UCSD, School of Medicine.
`t Associate Professor, DepartmentofAnesthesia, VA Medical Center,
`Palo Alto; and Departmentof Anesthesia, Stanford University, School
`of Medicine.
`
`# Research Assistant, Department of Anesthesia, VA Medical Center,
`Palo Alto,
`
`§ Orion Corporation FARMOS.
`{ Assistant Professor, Department of Anesthesia, VA Medical Center,
`Palo Alto; and Departmentof Anesthesia, Stanford University, School
`of Medicine,
`
`Received from the Departments of Anesthesia, VA Medical Center,
`San Diego, California; UCSD, School of Medicine, San Diego, Cali-
`fornia; VA Medical Center, Palo Alto, California; and Stanford Uni-
`versity, School of Medicine, Stanford, California. Accepted for pub-
`lication November 11, 1992. Supported by a grant from the Medical
`Research Council of Canada and the Orion Corporation FARMOS,
`Address reprint requests to Dr. Dyck: DepartmentofAnesthesiology,
`Veterans Administration Medical Center, 3350 La Jolla Village Drive,
`San Diego, California 92161-9125.
`
`Anesthesiology, V 78, No 5, May 1993
`
`Conclusions: The intramuscular administration of dexme-
`detomidine avoids the acute hemodynamic changes seen with
`intravenous administration, but results in similar hemody-
`namic alterations within 4 h. (Key words: Hemodynamics.
`Pharmacokinetics. Sympathetic nervous system, a, agonists:
`dexmedetomidine.)
`
`THE a-adrenergic agonists are a new class of poten-
`tially useful adjunctive anesthetic agents. Clonidine,
`the prototypic a2-adrenergic agonist, is the most widely
`used drug of this class of compounds and decreases
`anesthetic and analgesic requirements in surgical pa-
`tients.’ Furthermore, clonidine administered before
`anesthetic induction may also minimize intraoperative
`hemodynamicfluctuationsandis an effective anxiolytic
`agent. Because clonidine has a long duration of action
`andis a partial agonist with only modestselectivity for
`the a2 versus the a, adrenoceptor, a second generation
`of a2 agonists is now being developed in an attempt to
`overcome the perceived shortcomings of clonidine in
`anesthetic settings. Dexmedetomidine (1,620:1 [ca2:
`@,]) is more selective at the a2, adrenoceptor than is
`clonidine (220:1) andis a full agonist.”
`is
`it
`To administer dexmedetomidine accurately,
`necessary to characterize the pharmacokinetic profile
`using relevant doses via the intended routes of admin-
`istration, and to correlate side effects, such as hemo-
`dynamic alterations, with the plasma concentrations of
`medication. Using a crossover study design, with dex-
`medetomidine administered intravenously and intra-
`muscularly, we characterized dexmedetomidine phar-
`macokinetics and hemodynamic alterations in ten
`healthy adult volunteers.
`
`Materials and Methods
`
`Subjects
`
`After approval by the Stanford University Investiga-
`tional Review Board, ten healthy male volunteers were
`recruited for this study. The average age of the subjects
`was 35.5 yr (range 29-44 yr) and the average weight
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 813
`
`

`

`814
`
`DYCK ETAL.
`
`was 79 kg (range 60-98 kg). Male subjects between
`the ages of 18-50 yr, with weight less than 100 kg and
`ASA physical status 1-2, were eligible for study.
`The volunteers were fasted from midnight before the
`study and were asked to abstain from any caffeine or
`alcohol consumptionfor the preceding 24 h. On arrival
`at the study site, an 18-G intravenous cannula wasin-
`serted, and 500 ml normalsaline was rapidly infused,
`followed by an infusion at 125 ml/h. A 20-G catheter
`was inserted into the radial artery and used both to
`measure arterial blood pressure and to sample blood
`for analysis of plasma dexmedetomidine concentra-
`tions. After fluid loading, 2 ug/kg dexmedetomidine
`hydrochloride was administered intravenously with an
`infusion pumpat a constant rate over 5 min. Subjects
`were kept in the supine position in a quiet room and
`disturbances were minimized until the initial 4 h of
`recording was completed. A minimum of 2 weeks after
`the intravenousstudy, the volunteer was given the same
`dose of dexmedetomidineas a single intramuscularin-
`jection into the deltoid muscle over 30 s during an
`otherwise similar study procedure.
`
`Blood Sampling
`Arterial blood was sampled at 0.5, 1.0, 1.5, 2.0, 2.5,
`3.0, 3.5, 4.0, 4.5, 5, 6, 8, 10, 12, 15, 20, 30, 45, 60,
`90, 120, 180, and 240 min after the start of the intra-
`venous infusion. The blood pressure transducer was
`exposedto valid arterial pressure waveform forat least
`15 s between eachof the blood samples obtained dur-
`ing the first 5 min of the intravenous infusion. During
`the intramuscular phase of the study, blood was sam-
`pled at 2,5, 10, 15, 20, 30, 45, 60, 90, 120, 180, and
`240 min after injection. Venous blood during both
`phases was sampled at 180, 240, 300, 450, 600, 900,
`1,200, and 1,440 min. The 5-ml K,EDTA anticoagu-
`lated samples were centrifuged and the plasma frozen
`at —40° C until the dexmedetomidine concentration
`was assayed. Blood sampling changed from arterial to
`venous at 4 h to minimize the length of time the vol-
`unteers were subjected to the presence of an arterial
`line.
`
`Dexmedetomidine Assay
`The plasma was assayed for dexmedetomidine con-
`centration using a gas chromatograph (GC) with mass
`spectroscopy (MS) detection.’ The pentafluorobenzoyl
`derivatives of dexmedetomidine and the internal stan-
`dard detomidine were produced during extraction of
`the plasma into n-hexane in the presence of NazCO;
`
`Anesthesiology, V 78, No 5, May 1993
`
`and pentafluorobenzoyl chloride. The organic phase
`was evaporated and the residue reconstituted in tolu-
`ene. A 1-ul aliquot was injected onto a Hewlett-Packard
`fused silica capillary column cross linked with 5%
`phenyl methyl silicone (Part number 19091J-102,
`Hewlett-Packard Company,Little Falls, DE) of a Hew-
`lett-Packard gas chromatograph (Model 5890A, Hew-
`lett-Packard Company, USA) using helium as the carrier
`gas. The GC oven was programmed for 1 min at 90° C
`and 30° C/min up to 275° C with a 5.8-min hold at
`275° C. The MS (Finnigan MAT TSQ 70, Finnigan MAT)
`using methane as the carrier gas was operated in neg-
`ative ion chemical ionization and selected ion moni-
`toring mode with 70 eV ionization energy at 200° C.
`The pentafluorobenzoyl derivatives of detomidine were
`detected at 380.1 (mass/charge ratio) and dexmede-
`tomidine at 394.1. The lower limit of quantitation for
`this GC/MS technique was 50 pg/ml, recovery oftri-
`tiated dexmedetomidine was 81%, and the coefficient
`of variation for within-day assays at 75 pg/ml was 12%,
`at 350 pg/ml was 9%, and at 600 pg/ml was 17.1%.
`The coefficient of variation for between-day assays at
`212 pg/ml was 12.8%, and at 537 pg/ml was 11.3%.
`When three extractions were injected into the GC/MS
`system ten times each, at 75, 350, and 600 pg/ml,
`respectively,
`the coefficient of variation was 9.7%,
`7.5%, and 11.3%, respectively.
`
`Pharmacokinetic Analysis
`MomentAnalysis. Moment analyses were performed
`on boththe intravenous and intramuscular data to cal-
`culate the model independent parameters: area under
`the concentration versus time curve (AUC), area under
`the first momentof the concentration versus time curve
`(AUMC), clearance (Cl), volumeofdistribution (Vd,,),
`and mean residence time (MRT). Values for AUC and
`AUMCare intermediate steps in the calculations and
`are presented for the sake of continuity. The AUC was
`calculated using the trapezoidal method with linear
`interpolation when concentrations were increasing and
`log-linear interpolation when concentrations were de-
`creasing.‘ At time points where both arterial and venous
`concentrations were obtained, the venous values were
`used in the trapezoidal integration. Extrapolation from
`the terminal data point to infinity was accomplished
`using log-linear regression of the terminal elimination
`phase andis presented as the terminal elimination half-
`life or ln(2) divided by the slope of the terminal phase.
`In similar fashion, the AUMC was calculated as the
`trapezoidal integration of the curve generated by mul-
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 814
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`815
`
`80386-based computer. Calibration signals were re-
`corded from a Delta-Cal Transducer Simulator (Model
`650-905, Midvale, Utah) at 0, 50, 100, 150, and 200
`mmHg. Thedigitized binary file was read and analyzed
`with software that located the peak and trough of each
`wave, and calculated the MAP by integrating the area
`beneath the wave. The algorithm has specific criteria
`that define a wave, and rejected signals caused by
`opening the stopcock to draw a blood sample or flush-
`ing the arterial catheter. The heart rate was calculated
`as the reciprocal of the time interval between. wave
`peaks. The systolic and diastolic blood pressure, MAP,
`and heart rate were recorded for each wave on thear-
`terial pressure trace during the study. The hemody-
`namic data reported represents the median MAP and
`heart rate values for each 60-s period.
`
`Results
`
`Figure 1 shows the dexmedetomidine plasma con-
`centration versus time profiles for all ten volunteers
`during the 5-min intravenous infusion andfor the fol-
`lowing 24 h. At 3 and 4 h after the infusion, simulta-
`neous arterial and venous blood samples were drawn.
`This allowed us to remove the arterial catheter from
`the subject while still sampling pharmacokinetic data.
`The venous concentrations were consistently higher
`than the arterial concentrations, as would be expected
`during the elimination phase of the pharmacokinetic
`profile.* The rise in plasma concentration was probably
`not elution from storage sites in skeletal muscle, be-
`cause the subjects remained supine from the start of
`
`tiplying cach plasma concentration by its time. The
`volumeofdistribution at steady state was calculated as
`follows:>
`
`Vass
`
`- Dose X AUMC _ Dose X T
`(AUC)?
`2X AUC '
`where T was the duration of the infusion. Clearance
`was calculated as the ratio of dose to AUC:°
`
`_ Dose
`AUC ’
`
`CL
`
`and MRTasthe ratio of Vd,, to clearance:°
`
`LL
`MRT = —..
`CL
`
`Thebioavailability of intramuscular dexmedetomidine
`was calculated as the ratio of the AUC after intramus-
`cular versus intravenous administration of the same
`dose:
`
`% Bioavailability =
`
`
`AU
`
`Deconvolution Analysis. Based on the assumption
`that the pharmacokinetics of dexmedetomidine are
`linear andstationary, but making no assumptions about
`model structure, the absorption characteristics of in-
`tramuscular
`dexmedetomidine were
`determined
`through least-squares deconvolution of the intramus-
`cular concentration versus time function with the in-
`travenous unit disposition function (UDF)®’ for each
`individual patient. Knowingthat:
`
`C, = I* D(* = convolution operator).
`
`where Cp is the concentration in the plasma,I is the
`input function, and D is the unit disposition function,
`the known zero order intravenous infusion of dexme-
`detomidine can be deconvolved against the plasma
`versus time concentration profile to produce the in-
`travenous-UDF. The deconvolution wasconstrained to
`be positive and unimodal.
`
`Arterial Wave Form Recording and Analysis
`Theradial artery cannula was connected to a Deltran
`II transducer (Model 901-007, Utah Medical Products
`Inc., Midvale, Utah) on a Hewlett-Packard 78353A
`monitor. Analog output from the HP monitor was re-
`corded by a TEAC R-71 recorder and simultaneously
`digitized on a DT2801 Data Translation A/D board at
`128 Hz with 12-bit resolution to the hard disk of an
`
`Anesthesiology, V 78, No 5, May 1993
`
`
`
`
`
`DexmedetomidineConc,ng/ml
`
`
`
`
`
`20
`
`25
`
`Infusion
`
`I 0
`
`5
`
`15
`10
`Time, min
`
` 0
`
`4
`
`8
`
`12
`Time, hours
`
`16
`
`20
`
`24
`
`Fig. 1. Dexmedetomidine intravenous plasma concentration
`versus time.
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 815
`
`

`

`816
`
`DYCK ET AL.
`
`Subject
`No.
`
`Table 1. Moment Analysis Intravenous(IV) Data
`
`AUC IV 0 to
`Terminal
`Inifinity
`Half-life
`(ng-min+mi-*)
`(min)
`
`AUC
`(% under Data)
`
`Clearance
`(L/min)
`
`Ves (L)
`
`AUMC
`(ng + min?)
`
`AUMC
`(% underData)
`
`MAT
`(min)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`
`Mean*
`SD
`
`361
`356
`571
`335
`305
`353
`297
`235
`260
`251
`
`329
`101
`
`963
`475
`455
`499
`300
`410
`624
`277
`237
`185
`
`385
`144
`
`.
`
`62
`90
`86
`94
`90
`90
`94
`98
`95
`97
`
`90
`3.8
`
`»
`
`0.403
`0.440
`0.342
`0,502
`0.397
`0,397
`0.621
`0.672
`0.668
`0.562
`
`0.511
`0.125
`
`486
`211
`187
`203
`174
`210
`251
`186
`161
`161
`
`194
`28.7
`
`437,000
`172,000
`314,000
`136,000
`135,000
`188,000
`121,000
`65,900
`63,400
`72,300
`
`141,000
`79,000
`
`19
`60
`53
`66
`63
`63
`65
`89
`73
`84
`
`68
`11
`
`1207
`480
`547
`405
`440
`530
`404
`277
`241
`285
`
`401
`112
`
`AUC = area under the curve; V,, = volumeof distribution at steady state; AUMC = area under first moment curve; MRT = meanresidence time.
`* Subject 1 excluded from summary statistics.
`
`known zero order input function to arrive at the cal-
`culated unit disposition function (UDF) for each sub-
`ject. Deconvolution was constrained to be positive and
`unimodal
`to restrict the output to physiologically
`meaningfulresults. Figure 3 shows average intravenous-
`UDF(+ SD) of the ten subjects calculated through the
`deconvolution technique. The resulting UDF for dex-
`medetomidine after intravenous administration was
`deconvolved against the concentration versus time
`profile after intramuscular administration on a patient-
`by-patient basis to produce the rate of intramuscular
`absorption shown in figure 4. Integration of the ab-
`sorption rate over time after intramuscular injection
`(figure 4) yields a total systemic dose of 133 ug and a
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`o2
`
`0.0
`
`the study until the 240-min sample, and were only
`starting to ambulate by 300 min. The plasma dexme-
`detomidine concentrations after intravenous adminis-
`tration decreased to less than the limit of quantitation
`in six patients by 20 h after administration.
`Momentanalysis of the intravenous data for the ten
`subjects is presented in table 1. The MRT ofsubject 1
`was so long that 24-h sampling did not adequately
`characterized the AUC. The AUC data for this subject
`encompassed only 62% of the total area and the AUMC
`19%. The means of the momentanalysis, therefore, do
`not include this subject. The mean clearance was 0.511
`+ 0.125 L/min, Vd,, was 194 + 28.7 L, and MRT was
`401 £112 min.
`Figure 2 shows the plasma concentration versus time
`profile after intramuscular administration of 2 ug/kg
`dexmedetomidine. The time to peak plasma concen-
`tration was 13 + 18 min and the mean peak concen-
`tration was 0.81 + 0.27 ng/ml(table 2). The variability
`in peak and time to peak concentrations was high. This
`was due, in large part, to the first two subjects who
`showed slower absorption with longer time to peak
`concentrations and lower peak concentrations. If the
`mean values are recalculated to include only subjects
`3-10, the time to maximum concentration was 6.1
`+ 4.4 min and the maximum concentration was 0.91
`+ 0,22 ng/ml. The average area under the concentra-
`tion versus time curve for all subjects was 243 + 78
`ng+min™’+ mi7and the average bioavailability was 73
`+ 11% (table 2).
`The concentration versus time profile for the intra-
`venous administrations was deconvolved against the
`
`Anesthesiology, V 78, No 5, May 1993
`
`4
`
`8
`
`ng/ml 0
`PlasmaDexmedetomidineConc,
`
`
`
`12
`Time, hours
`Fig. 2. Dexmedetomidine intramuscular plasma concentration
`versus time.
`
`16
`
`20
`
`24
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 816
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`817
`
`Table 2. Moment Analysis Intramuscular (IM) Data
`Terminal
`Half-life
`(min)
`
`AUC IM 0 toInfinity
`(ng+ min mi")
`
`Subject
`No.
`
`AUC
`(% underData)
`
`Bioavailability
`(%)
`
`Time to Peak
`Concentration
`(min)
`
`Peak Concentration
`(ng/ml)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`
`Mean
`SD
`
`329
`267
`394
`237
`262
`235
`237
`126
`170
`174
`
`243
`78
`
`AUC = area underthe curve,
`
`707
`291
`243
`314
`304
`254
`363
`57
`149
`131
`
`281
`177
`
`75
`95
`96
`96
`90
`94
`93
`99
`95
`99
`
`93
`6.9
`
`91
`75
`69
`71
`86
`67
`80
`54
`66
`69
`
`73
`11
`
`20
`60
`20
`i)
`10
`15
`5
`5
`2
`5
`
`13
`18
`
`0.37
`0.49
`0.81
`1,2
`0.61
`0.87
`0.88
`0.71
`1.2
`0.95
`
`0.81
`0.27
`
`bioavailability of 84% (133 ug systemically absorbed/
`158 wg average intramuscular dose). Figure 5 shows
`the cumulative absorption over time, as a percent of
`total absorption. The mean intramuscular dose was 158
`ug resulting in a bioavailability of intramuscular-to-in-
`travenousdosing of 84% using deconvolutionanalysis.
`The AUMC offigure 4 was 277 ug: h?. The mean ab-
`sorption time (MAT) calculated as AUMC/AUC was 2.08
`h, and the meanfirst order rate constant (Ka)for intra-
`muscular absorption as the reciprocal of MAT was
`0.48 hh.
`Figures 6 and 7 show the mean MAP (+SD) of the
`ten volunteers during intravenous and intramuscular
`dexmedetomidine. The peak rise in MAP after intra-
`
`venous dexmedetomidine occurred at 5 min and was
`22% above baseline values. A much smaller increase
`in MAP occurred after intramuscular injection, but was
`even earlier in onset and was probably caused by the
`anxiety induced by the intramuscular injection. By 4
`h, both intravenous and intramuscular dexmedetomi-
`dine resulted in a 20% decline in MAP from baseline.
`The blood pressure disturbance at 140-150 min was
`caused by subjects waking up abruptly, rather than by
`ambulation of the subjects. Figures 8 and 9 show the
`mean heart rate (£SD) for the ten volunteers after in-
`travenous and intramuscular dexmedetomidine,
`re-
`spectively. The decline in HR after intravenous dex-
`medetomidine was 27% below baseline 4-5 min after
`
`2i. Me
`:"
`
`Za
`
`dy“ny
`
`| |
`
`|
`
`
`
`4
`
`g
`
`z
`
`100
`
`od
`:
`. 10
`:
`a
`<
`3
`3a
`
`1
`
`0.1
`
`_ 101%
`=
`Ay
`x
`"hes
`5 10
`
`Soy,
`
`;
`
`
`0
`5
`0 15
`20
`25
`30
`
`by
`
`by
`
`Time,min
`eee ga zaa ay by
`
`0
`
`4
`
`8
`
`12
`Time, hours
`
`16
`
`20
`
`24
`
`Fig. 3. Mean unit disposition function + SD of intravenous
`dexmedetomidine.
`
`Time, hours
`Fig. 4. Dexmedetomidine intramuscular rate of absorption
`(+ SD) versus time.
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 817
`
`

`

`MAP -IM
`
`120
`
`110
`
`= 100
`Sa
`
`70
`
`60
`
`50
`
`0
`
`50
`
`100
`Time, min
`
`150
`
`200
`
`250
`
`Fig. 7. Mean arterial pressure (+ SD) of intramuscular dex-
`medetomidine.
`
`818
`
`DYCK ET AL.
`
`100
`
`80
`
`60
`
`40
`
`20
`
`
`
`
`
`IMDexmedetomidineCumulativeAbsorption%TotalDose
`
`4
`
`8
`
`a0
`
`12
`
`16
`
`20
`
`24
`
`Fig. 5. Dexmedetomidine cumulative absorption (+ SD) versus
`time.
`
`Time, hours
`
`starting the infusion. Again, this was not seen after in-
`tramuscular injection. Dexmedetomidine given by in-
`tramuscular and intravenous routes caused a 5% and
`10% decline in HR, respectively, by the end of the 4-
`h recording period.
`
`Discussion
`
`Hemodynamic alterations after intravenous admin-
`istration preclude the use of dexmedetomidine as a
`rapid intravenous infusion or bolus. Compartmental
`pharmacokinetic analysis would be required to admin-
`ister dexmedetomidine via a computer-controlled in-
`fusion pump, but dexmedetomidine would be more
`commonly administered by slow intravenous infusions
`130
`
`MAP - IV
`
`to steady state or by intramuscular injections, for which
`momentanalysis is adequate. Momentanalysis is model
`independent and allows calculation of fundamental
`pharmacokinetic parameters, such as volumeofdistri-
`bution and clearance. Momentanalysis cannot describe
`the multiple distribution phases, as are commonly
`modeled by compartmental pharmacokinetic analysis.
`The venous dexmedetomidine concentrations at 3 and
`4 h were slightly greater than arterial concentrations
`and the calculated area under the curves maybeslightly
`greater than an entirely arterial concentration profile.
`Dexmedetomidine appears to have systemic clearance
`of approximately 0.5 L/min, approximately one-half
`of hepatic blood flow. Overall, the volume and clear-
`anceare fairly similar to those of fentanyl? with an ex-
`90
`
`HR-IV
`
`80
`
`& 70a2
`
`60
`
`s&
`
`= so
`
`40
`
`120
`
`110
`
`=P 100
`
`&E 3 7
`
`0
`
`60
`
`50
`30
`ereseeeeee:ee
`eereaes|
`0
`50
`100
`150
`200
`250
`0
`50
`100
`150
`200
`250
`Time, min
`Time, min
`
`Fig. 6. Mean arterial pressure (+ SD) of intravenous dexme-
`detomidine.
`
`Fig. 8. Mean heart rate (+ SD) of intravenous dexmedetomi-
`dine.
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 818
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`819
`
`HR-IM
`
`”
`
`HeartRate,bpm3S3gS
`
`eyoS
`
`sumption of linearity is violated frequently in both
`compartmental pharmacokinetic analysis and moment
`analysis. As violation of linearity is part of model mis-
`specification, the magnitude of such a violation can be
`roughly estimated from the size of the residual error
`when compartmental models are fit to the data. In
`practice, most pharmacokinetic studies simply ignore
`the issue of nonlinear pharmacokinetics because the
`extent of the violation is fairly small, and pharmaco-
`kinetics based on the assumption oflinearity provide
`a succinct, easily estimated, and clinically useful de-
`scription of pharmacokinetic behavior. The hyperten-
`sion and bradycardia seen after intravenous dexmede-
`re
`0
`50
`100
`150
`200
`250
`tomidine were not seen after intramuscular adminis-
`tration.
`Time, min
`The peak plasma concentrations were an order of
`magnitude lower after intramuscular administration.
`On the assumptionthatthe differences in hemodynamic
`profiles may have been a result of concentration-de-
`pendentperipheral vasoconstriction, one might strive
`to maintain a plasma dexmedetomidine concentration
`of less than 1.0 ng/ml. From the presented moment
`analysis data, and knowing that clearance times targeted
`concentration will yield a corresponding infusion rate,
`the steady-state concentration of 1.0 ng/ml could be
`achieved through an infusion of dexmedetomidine at
`0.511 wg/min. The plasma concentration will asymp-
`totically approach the targeted steady-state concentra-
`tion of 1.0 ng/ml and would bevery close to the steady-
`state concentration after three elimination half-lives or
`1,155 min. If it is desirable to attain the target con-
`centration before 19.25 h, a loading dose, calculated
`as targeted steady-state concentration times Vd,, or 194
`ug, may be administered and followed by the mainte-
`nance infusion. The loading dose should not be ad-
`ministered as a bolus, but can be given as an infusion
`over 30-45 min with minimal increased risk of adverse
`hemodynamic alterations.
`Twosubjects lost consciousness when they assumed
`the upright posture, approximately 5 h after the intra-
`venous infusion of dexmedetomidine. During these
`events, both subjects had bradycardia. Thelikely etiol-
`ogy for this loss of consciousnessis the sympatholytic
`effect of the dexmedetomidine leaving unopposed va-
`gal tone. Both subjects recovered from their vasovagal
`events spontaneously and uneventfully. No analogous
`events occurred after intramuscular administration, but
`increased caution on the part of both the investigators
`and the subjects during the second phase of the study
`may have prevented similar episodes.
`
`Fig. 9. Mean heartrate (+ SD) of intramuscular dexmedetom-
`idine.
`
`tensive tissue distribution (fentanyl Vd,, approximately
`300 L) and a moderately large hepatic clearance (i.e.,
`large CL). The MRTis a term unfamiliar to many anes-
`thesiologists, but one that might serve a useful purpose
`for comparison of medications given the misleading
`characteristics of the compartmental elimination half-
`life.'° The MRT is the momentanalysis equivalent of
`the half-life in compartmental analysis and represents
`the time required to eliminate 63.2% of an intravenous
`bolus dose. The effective half-life of a medication is
`0.693 times the MRT.
`The bioavailability of intramuscular dexmedetomi-
`dine was between 70% and 80%. On average, peak
`plasma concentrations of dexmedetomidine were ob-
`tained within 15 min after intramuscular injection, al-
`though the time to peak concentration after intramus-
`cular injection varied widely. The intramuscular ab-
`sorption profile was biphasic with early rapid
`absorption.
`Intravenous dexmedetomidine as a rapid infusion
`caused biphasic changes in HR and MAP similar to those
`seen after administration of clonidine.'''? The clinical
`utility of intravenous dexmedetomidine will be limited
`by these hemodynamicalterations. Bolus intravenous
`administration of dexmedetomidine would be unwise
`in most circumstances.It is possible that dexmedetom-
`idine pharmacokinetics are not linear secondary to the
`concentration-dependent hemodynamic alterations.
`Manyofthe drugs usedin anesthesia practice (#.e., pro-
`pofol and thiopental) affect hemodynamics and prob-
`ably have nonlinear pharmacokinetics. Thus, the as-
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 819
`
`

`

`820
`
`DYCK ET AL.
`
`We concludethat, although intramuscular absorption
`of dexmedetomidine is rapid, the peak plasma con-
`centrations that result are less than those after a 5-min
`intravenous infusion with the same dose, and hemo-
`dynamicalterations are less severe.
`
`References
`
`1. Aho M, Lehtinen A-M, Erkola O, Kallio A, Kortilla K: The effect
`of intravenously administered dexmedetomidine on perioperative
`hemodynamics and isoflurane requirements in patients undergoing
`abdominal hysterectomy. ANESTHESIOLOGY 74:997~1002, 1991
`2. Virtanen R, Savola JM, Saano U, Nyman L: Characterization of
`selectivity, specificity, and potency of medetomidine as an alpha-2
`adrenoceptor agonist. Eur J Pharmacol 150:9-14, 1988
`3. Vuorilehto L, Salonen JS, Anttila M: Picogram level determi-
`nation of medetomidine in dog serum by capillary gas chromatog-
`raphy with negative ion chemical ionization mass spectrometry. J
`Chromatogr 497:282-287, 1989
`4. Gibaldi M, Perrier D: Absorption kinetics and bioavailability,
`Pharmacokinetics, 2nd edition. Edited by Swarbrick J. New York,
`Marcel Dekker, 1982, pp 145-198
`
`5. Gibaldi M: Biopharmaccutics and Clinical Pharmacokinetics,
`4th edition. Malvern, Pennsylvania, Lea & Febiger, 1991, pp 14-23
`6. Verotta D: An inequality-constrainedleast squares deconvolution
`method. J Pharmacokinet Biopharm 17:269-289, 1989
`7. Streisand JB, Varvel JR, Stanski DR, LeMaire L, Ashburn MA,
`Hague BI, Tarver SD, Stanley TH: Absorption and bioavailability of
`oral transmucosal fentanyl citrate. ANESTHESIOLOGY 75:223-229, 1991
`8. Chiou WL: The phenomenonand rationale of marked depen-
`dence of drug concentration on blood sampling site implications in
`pharmacokinetics pharmacodynamics, toxicology and therapeutics
`(part 1). Clin Pharmacokinet 17:175-199, 1989
`9. Scott J, Stanski DR: Decreased fentany!] and alfentanil dose re-
`quirements with age: A simultaneous pharmacokinetic and phar-
`macodynamic evaluation. ) Pharmacol Exp Ther 240:159-166, 1987
`10. Hughes MA, Glass PSA, Jacobs JR: Context-sensitive half-time
`in multicompartment pharmacokinetics models for intravenous an-
`esthetic drugs. ANESTHESIOLOGY 76:334~—341, 1992
`11. Rhee HM,LappJD:Are opioid receptors involved in the bra-
`dycardic and hypotensive action of clonidine. Am J Hypertens 1:
`249S-2545S, 1988
`12. Frisk-Holmberg M:Effect of clonidine at steady-state on blood
`pressure in spontancously hypertensive rats: Interaction of various
`alpha-adrenoceptorantagonists. Acta Physiol Scand 120:37-42, 1984
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From:http://anesthesiology.pubs.asahq.org/pdfaccess.ashx?urF/data/Journals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,455,527
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 820
`
`