813
`
`I CLINICAL INVESTIGATIONS
`
`Anesthesiology
`78:813-320. 1993
`© I993 American Society of Anesthesiologists. inc.
`1. ll. Llpplncott Company. Phllzldclphla
`
`The Pharmacokinetics and Hemodynamic Efiects of
`Intravenous and Intramuscular Dexmedetomidine
`Hydrochloride in Adult Human Volunteers
`J. B. Dyck, M.D., F.Fi.C.P.C.,' M. Maze, M.B., Ch.B.,1 C. Haack, I-?.N.,¢ L. Vuorilehto, M.Sc.,§ S. L. Sharer, M.D.‘|
`
`Backgrotmd: Dexmedetomidine is an at. agonist with poten-
`tial utility in clinical anesthesia for both its sedative and sym-
`patholytic properties.
`Methods: The pharmacokineties and hemodynamic changes
`that occurred in ten healthy male volunteers were determined
`after administration of dexmedetomidine 2 pg/kg by intra-
`venous or 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 intravenous data, was biphasic. The percentage bio-
`availability of dexmedetomidinc administered intramuscu-
`larly compared with the same dose administered intrave-
`nously was 73 :l: 11% (mean 1: 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 dexmcde-
`tornidine, there were biphasic changes in blood pressure.
`During the S-min intravenous infusion of 2 ug/kg dexmede-
`tomldine, the mean arterial 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-
`modynamic profile did not show acute alterations after intra-
`muscular administration. Durlng 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 Department ofAncsthesia. UCSD, School of Medicine.
`f Associate Professor, Department ot'Anesthcsla. VA Medical Center,
`Palo Alto; and Department of Anesthesia, Stanford University. School
`of Medicine.
`
`# Research Assistant. Department of Anesthesia. VA Medical Center.
`Palo Alto.
`
`§ Orion Corporation FARMOS.
`1 Assistant Professor. Department of Anesthesia, VA Medical Center.
`Palo Alto; and Department of 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 Ccnter. Palo Alto. California; and Stanford Uni-
`versity. School of Medicine. Stanford. California. Accepted for pub-
`lication November 1 1, 1992. Supported by a grant from the Medical
`Research Council of Canada and the Orion Corporation FARMOS.
`
`Address reprint requests to Dr. Dyck: Department ofAnesthesiology.
`Veterans Administration Medical Center, 33 50 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:
`dexmedetomldine.)
`
`THE a2-adrcncrgic agonists are a new class of poten-
`tially useful adjunctive anesthetic agents. Clonidine,
`the prototypic ct;-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 intraoperativc
`hemodynamic fluctuations and is an eifective anxiolytic
`agent. Because clonidine has a long duration of action
`and is a partial agonist with only modest selectivity for
`the at; versus the oz. adrenoccptor, a second generation
`of oz; agonists is now being developed in an attempt to
`overcome the perceived shortcomings of clonidine in
`anesthetic settings. Dexmedetomidinc (1 ,620:1 [ot2:
`a.]) is more selective at the 0:; adrenoceptor than is
`clonidine (220:1) and is a full agonist.’
`is
`it
`To administer dexmedetomidinc accurately,
`necessary to characterize the pharmacokinetic profile
`using relevant doses via the intended routes of admin-
`istration, and to correlate side elfects, such as hemo-
`dynamic alterations, with the plasma concentrations of
`medication. Using a crossover study design, with dex-
`mcdetomidinc 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: Itttp://anestltesiologypubs.asaltq.orgIpdfaccess.asl|x?urI=/dataIJournaIsIJASAl93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 813
`
`

`
`814
`
`DYCK ET AL.
`
`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 calfeine or
`alcohol consumption for the preceding 24 h. On arrival
`at the study site, an 18-G intravenous cannula was in-
`serted, and 500 ml normal saline 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 dexmedetomidinc concentra-
`tions. After fluid loading, 2 ug/kg dexmedetomidinc
`hydrochloride was administered intravenously with an
`infusion pump at 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 intravenous study. the volunteer was given the same
`dose of dexmedetomidine as a single intramuscular in-
`iection 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
`exposed to valid arterial pressure waveform for at least
`15 s between each of 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 K2EDTA anticoagu-
`lated samples were centrifuged and the plasma frozen
`at —40° C until the dexmedetomidinc 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 dexmedetomidinc 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 Na2CO3
`
`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 19091]-I02,
`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, Finnlgan 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-
`tomidinc at 394.1. The lower limit of quantitarion for
`this GC/MS technique was 50 pg/ml, recovery of tri-
`tiated dexmedetomidinc was 81%, and the coefficient
`of variation for within-day assays at 75 pg/ml was 1 2%,
`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.
`
`Pbarmacokinetic Analysis
`Moment Analysis. Moment analyses were performed
`on both the intravenous and intramuscular data to cal-
`
`culate the model independent parameters: area under
`the concentration versus time curve (AUC), area under
`the first moment of the concentration versus time curve
`
`(AUMC), clearance (Cl), volume of distribution (Vds,),
`and mean residence time (MRT). Values for AUC and
`AUMC are intermediate steps in the calculations and
`are presented for the sake of continuity. The AUC was
`calculated rising 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 and is 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://anesthesiologypubs.asal|q.orgIpdfaccess.asI|x?urI=/dztalJournaIslJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 814
`
`

`
`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`815
`
`tiplying each plasma concentration by its time. The
`volume of distribution at steady state was calculated as
`follows?
`
`Vdss
`
`:3 Dose X AUMC _ Dose X T
`(AUc')’
`2 x AUC ’
`where T was the duration of the infusion. Clearance
`was calculated as the ratio of dose to AUC:’
`
`CL
`
`_ Dose
`AUC '
`
`and MRT as the ratio of Vd,.,. to clearance?
`
`Vdss
`MRT = -—- .
`CI.
`
`The bioavailability of intramuscular dexmedetomidine
`was calculated as the ratio of the AUC after intramus-
`cular versus intravenous administration of the same
`dose:
`
`AUG“;
`% Bioavailability = AUCW
`
`X 100.
`
`Deconvolution Analysis. Based on the assumption
`that the pharmacokinetics of dexmedetomidine are
`linear and stationary, 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)"7 for each
`individual patient. Knowing that:
`
`C,, = In«D(n= = 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 was constrained to
`
`be positive and unimodal.
`
`Arterial Wave Form Recording and Analysis
`The radial 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
`
`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
`mmilg. The digitized 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 betwecmwave
`peaks. The systolic and diastolic blood pressure. MAP,
`and heart rate were recorded for each wave on the ar-
`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 and for 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
`
`ng/ml 0.l
`
`DexmedeiomidincCone,
`
`0
`
`4
`
`3
`
`I2
`Time. hours
`
`I6
`
`20
`
`24
`
`Fig. 1. Dexmedetomldine Intravenous plasma concentration
`versus time.
`
`Downloaded From: ltttp://anesthesiologypnbs.as:l|q.orglpdfaccess.asl|x?urI=IdntalJournaIsIJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 815
`
`

`
`816
`
`DYCK ET AL.
`
`Table 1. Moment Analysis Intravenous (IV) Data
`AUC IV 0 to
`Terminal
`Inlllnlty
`Hall-Illa
`(ng « rnln - ml")
`(min)
`
`Subject
`No.
`
`AUC
`(% under Data]
`
`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
`
`i
`
`Clearance
`(L/mln)
`
`V. (L)
`
`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
`
`AUMC
`(ng - min’)
`
`437.000
`172.000
`314.000
`136.000
`135,000
`188.000
`121 .000
`65.900
`63.400
`72.300
`
`141.000
`79,000
`
`AUMC
`(% under Data)
`
`19
`60
`53
`66
`63
`63
`65
`89
`73
`84
`
`68
`11
`
`MRT
`(min)
`
`1 207
`480
`547
`405
`440
`530
`404
`277
`241
`285
`
`401
`112
`
`AUC = area under the curve; V.. = volume of distribution at steady state; AUMC - area under llrst moment curve; MRT - mean residence time.
`' subject 1 excluded lrom 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
`meaningful results. Figure 3 shows average intravenous-
`UDF (_-t SD) of the ten subjects calculated through the
`deconvolution technique. The resulting UDF for dex-
`mederomidine 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 pg and a
`
`
`
`S P
`
`at
`
`0.6
`
`0.4
`
`0.2
`
`0.0
`
`
`
`
`
`
`
`PlasmaDcxrnedetomidineCone,ng/ml
`
`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.
`Moment analysis of the intravenous data for the ten
`subjects is presented in table 1. The MRT of subject 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 moment analysis. therefore. do
`not include this subject. The mean clearance was 0.5 1 1
`:t 0.125 L/min. Vd,,. was 194 :1: 28.7 L. and MRT was
`401 1 1 12 min.
`
`Figure 2 shows the plasma concentration versus time
`profile after intramuscular administration of 2 jig/kg
`dexmedetomidlne. The time to peak plasma concen-
`tration was 13 i 18 min and the mean peak concen-
`tration was 0.81 i 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
`t 4.4 min and the maximum concentration was 0.91
`:1: 0.22 ng/ml. The average area under the concentra-
`tion versus time curve for all subjects was 243 1- 78
`ng - min" - ml" and the average bioavailability was 73
`i 11% (table 2).
`The concentration versus time profile for the intra-
`venous administrations was deconvolved against the
`
`Anesthesiology. V 78. No 5. May 1993
`
`0
`
`4
`
`8
`
`12
`Time, hours
`Fig. 2. Dexmedetomidlne intramuscular plasma concentration
`versus time.
`
`16
`
`TD
`
`24
`
`Downloaded From: http://anesthesiologypubs.asal|q.orgIpdfaccess.asI|x?urI=/dztalJournaIslJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 816
`
`

`
`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`817
`
`Table 2. Moment Analysis Intramuscular (KM) Data
`
`11rne to Peak
`Terminal
`Subject
`AUC IM 0 to infinity
`Hall-Ilia
`AUG
`Bloavallabllliy
`Concentration
`Peak Concentration
`
`No.
`(ng - mln - ml“)
`(min)
`(% under Data)
`(96)
`(mln)
`(ng/ml)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`1 0
`
`Mean
`SD
`
`329
`267
`394
`237
`262
`235
`237
`126
`170
`174
`
`243
`78
`
`AUG = area under the curve.
`
`707
`291
`243
`314
`304
`254
`363
`57
`149
`1 31
`
`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
`5
`10
`1 5
`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 pg systemically absorbed]
`158 pg average intramuscular dose). Figure 5 shows
`the cumulative absorption over time. as a percent of
`total absorption. The mean intramuscular dose was 158
`pg resulting in a bioavailability of intramuscular-to-in-
`travenous dosing of 84% using deconvolution analysis.
`The AUMC of figure 4 was 277 ug- h‘. The mean ab-
`sorption time (MAT) calculated as AUMC/AUC was 2.08
`h, and the mean first order rate constant (Ka) for intra-
`muscular absorption as the reciprocal of MAT was
`0.48 h".
`
`Figures 6 and 7 show the mean MAP (iSD) of the
`ten volunteers during intravenous and intramuscular
`dexmedetomidine. The peak rise in MAP after intra-
`
`venous dexmcdetomidine 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
`11, 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 (iSD) 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
`
`100
`
`Q
`
`3
`
`°
`3?
`g
`§
`:5
`.9
`30
`
`§D
`
`.0
`
`1
`
`0.1
`
`0
`
`.
`
`4
`
`_ '00 8
`3%
`
`1%
`3"*‘%s
`
`g »o
`8
`
`5 6°
`O
`6
`
`
`1,3;
`
`33
`
`20
`0
`5
`l0
`I5
`Tlmc,min
`3
`833353353;
`
`25
`L
`
`8
`
`I2
`Time, hours
`
`16
`
`20
`
`30
`
`J.
`
`24
`
`Fig. 3. Mean unit disposition function i SD of intravenous
`dexmedetomidine.
`
`Anesthesiology. V 78. No 5. May 1993
`
`Ka
`
`Em
`3%
`I
`
`éfisblsllllll
`
`ill 111 11
`
`
`
`Time. hours
`Fig. 4. Dexmedetomidine intramuscular rate of absorption
`(3 SD) versus time.
`
`Downloaded From: http://anesthesiologypnbs.as:l|q.orglpdfaccess.asl|x?urI=IdatalJournaIsIJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 817
`
`

`
`818
`
`DYCK ET AL.
`
`§ 3
`
`G
`
`
`
`
`
`[MDexmedetornidincCumulativeAbsorption 96TotalDose 88
`
`o<O:‘
`
`AO
`
`MAP - [M
`
`'30
`I20
`
`I10
`
`100
`
`MAP,mmHg 8
`
`80
`
`70
`
`60
`
`??-T-
`0
`S0
`100
`150
`200
`250
`Time, min
`
`Fig. 7. Mean arterial pressure (1 SD) of intramuscular dex-
`medetomidine.
`
`to steady state or by intramuscular injections, for which
`moment analysis is adequate. Moment analysis is model
`independent and allows calculation of fundamental
`pharmacokinetic parameters, such as volume of distri-
`bution and clearance. Moment analysis cannot describe
`the multiple distribution phases, as are commonly
`modeled by compartmental pharmacokinetic analysis.
`The venous dexmedetomidine concentrations at 3 and
`
`4 11 were slightly greater than arterial concentrations
`and the calculated area under the curves may be slightly
`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-
`ance are fairly similar to those of fentanyl’ with an ex-
`90
`
`HR-IV
`
`80
`
`E 70
`n.D
`.9’
`:2 60
`
`=1
`
`so
`
`40
`
`30
`
`’*'f
`0
`50
`I00
`150
`N)
`250
`Time, min
`
` 0
`
`4
`
`8
`
`I2
`
`16
`
`20
`
`2-:
`
`Fig. 5. Dcxmedctomidinc cumulative absorption (1 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
`
`MAP - IV
`
`"°
`13)
`
`H0
`
`MAP.mrnHg 88
`
`S
`
` 0
`
`S0
`
`100
`Time, min
`
`I50
`
`200
`
`250
`
`Fig. 6. Mean arterial pressure (: SD) of intravenous dexme-
`detomldine.
`
`Fig. 8. Mean heart rate (1: SD) of intravenous dexmedetomi-
`dine.
`
`Anesthesiology. V 78. No 5. May 1995
`
`Downloaded From: Itttp://anesthesiologypubs.asal|q.orgIpdfaccess.asI|x?urI=/datalJournaIslJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 818
`
`

`
`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`819
`
`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 of linearity provide
`a succinct, easily estimated, and clinically useful de-
`scription of pharmacokinetic behavior. The hyperten-
`sion and bradycardia seen after intravenous dexmede-
`tomidine were not seen after intramuscular adminis-
`tration.
`
`The peak plasma concentrations were an order of
`magnitude lower after intramuscular administration.
`On the assumption that the differences in hemodynamie
`profiles may have been a result of concentration-de-
`pendent peripheral 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 pg/min. The plasma concentration will asymp-
`totically approach the targeted steady-state concentra-
`tion of 1.0 ng/ml and would be very 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 beforc 19.25 h, a loading dose, calculated
`as targeted steady-state concentration times Vd., or 194
`pg, 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
`hemodynamie alterations.
`Two subjects 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. The likely etiol-
`ogy for this loss of consciousness is the sympatholytlc
`elfect of the dexmedetomidine leaving unopposed va-
`gal tone. Both subjects recovered from their vasovagal
`events spontaneously and unevcntfully. 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.
`
`HR-lM
`
`200
`
`250
`
` 0
`
`50
`
`150
`
`I00
`Time. min
`
`90
`
`80
`
`E 70
`.8‘
`2'
`E 60
`E0
`=1
`
`so
`
`40
`
`30
`
`Fig. 9. Mean heart rate (i SD) of intramuscular dexmedetom-
`idine.
`
`tensive tissue distribution (fentanyl Vdt, approximately
`300 L) and a moderately large hepatic clearance (1. e.,
`large CL). The MRT is 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 moment analysis 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. ' "'2 The clinical
`
`utility of intravenous dexmedetomidine will be limited
`by these hemodynamie alterations. 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 hemodynamie alterations.
`Many of the drugs used in anesthesia practice (1. e., pro-
`pofol and thiopental) afi'ect hemodynamics and prob-
`ably have nonlinear pharmacokinetics. Thus, the as-
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From: Itttp://anesthesiologypubs.asal|q.orgIpdfacress.asI|x?urI=/dztalJournaIslJASAI93l3l5/ on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 819
`
`

`
`820
`
`DYCK ET AL.
`
`We conclude that. 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-
`dynamic alterations are less severe.
`
`References
`
`1. Aho M. behtinen A-M, Erkoia 0. Kallio A. Kortilia K: The effect
`of intravenously administered dcxmedetomidine on perloperative
`hemodynamlcs and isoilurane requirements in patients undergoing
`abdominal hysterectomy. ANESTHESIOLOGY 74:997-1002. 1991
`2. Virtancn R. Savola JM. Saano U. Nyman L: Characterization of
`selectivity. specificity. and potency of medetomidlne as an alpha-2
`adrcnoceptor agonist. Eur] Pharmacol 150:9-14. 1988
`3. Vuorilehto L, Salonen JS, Anttiia M: Plcogram level determi-
`nation of medetomidlne 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 bioavallabillty.
`Pharmacoltinetlcs. 2nd edition. Edited by Swarbrlck J. New York.
`Marcel Dekker. 1982, pp 145-198
`
`5. Gibaldi M: Biopharmaceutics and Clinical Pharmacokinetics.
`4th edition. Malvern. Pennsylvania. Lea & Fcbiger. 1991. pp 14-23
`6. Verona D: An inequality-constrained least squares deconvolution
`method.) Pharmacokinet Biopharm 17:269-289. 1989
`7. Streisand J13. Varvel JR. Stanski DR. LeMaire L. ltshburn MA.
`Hague Bl. Tarver SD. Stanley TH: Absorption and bioavailabllity of
`oral transmucosal fentanyl citrate. ANISFHIOLOGY 75:223-229. 1991
`8. Chiou WL: The phenomenon and rationale of marked depen-
`dence of drug concentration on blood sampling site implications in
`pharmacokinetlcs pharmacodynamics. toxicology and therapeutics
`(part 1). Clin Pharmacokinet 17:175-199. 1989
`9. Scott J, Stanskl DR: Decreased fentanyi and aifentanil dose re-
`quirements with age: A simultaneous pharmacokinetic and phar-
`macodynamic evaluation.) Pharmacol Exp Ther 240:1S9-166, 1987
`10. Hughes MA. Glass PSA. Jacobs JR: Context-sensitive half-time
`in multicompartment pharmacokineties models for intravenous an-
`esthetic drugs. Anmuesiowov 76:334-341. 1992
`l l. Rhee HM, Lapp JD: Are opioid receptors involved in the bra-
`dycardic and hypotenslvc action of clonidine. Am] Hypertens 1:
`2498-2543. 1988
`12. Frisk-Holmberg M: Effect of clonidine at steady-state on blood
`pressure in spontaneously hypertensive rats: interaction of various
`alpha-adrenoceptor antagonists. Acta Physiol Scand l20:37—42. 1984
`
`Anesthesiology. V 78. No 5. May 1993
`
`Downloaded From: ltttp://anesthesiologypnbs.as:l|q.org/pdfaecess.asl|x?urI=IdatalJournnIsIJASAI93l3l5I on 08/09/2016
`Petition for Inter Partes Review of US 8,648,106
`Amneal Pharmaceuticals LLC — Exhibit 1021 — Page 820