8 13
`
`• CLINICAL INVESTIGATIONS
`
`Anesthesiology
`78:813-820, 1993
`© 1993 Alncrlcan Society of Ancsthcslologlsts, Inc.
`J. 8. Lippincott Company, Phlladclphln
`
`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.,t C. Haack, R.N.,"i L. Vuorilehto, M.Sc.,§ S. L. Shafer, M.D.11
`
`Backgrotmd: Dexmedetomldlne Is an a 2 agonlst wlth poten(cid:173)
`tial utUlty In clinical anesthesia for both its sedative and sym(cid:173)
`patholytlc properties.
`M ethods: The pharmacoklnetlcs and hemodyriamlc changes
`that occurred ln ten h ealthy male volunteers were determlned
`after administration of dexmedetomldlne 2 µg/ kg by Intra(cid:173)
`venous or Intramuscular route In separate study Se$Slons.
`Rcs11lts: The Intramuscular absorption pr ofile of dexmede(cid:173)
`tomldlnc, as dctcrmlned by deconvolution of the observed
`concentrations against the unit disposition function derived
`from the Intravenous data, was blphaslc. The percentage blo·
`availability of dexmedetomldlne administered lnU'amuscu(cid:173)
`larly compared with the same dose administered Intrave(cid:173)
`nously was 7~ ± 11% (mean ± SO). After Intramuscular ad(cid:173)
`ministration, the mean time to peak concentratJon was 12 min
`(range 2-60 min) and the mean peak concentration was 0.81
`± 0.27 ng/ml. After Intravenous admlnls1ratlon of dexmede(cid:173)
`tomidlne, there were blphaslc changes In blood pressure.
`During the S·mln Intravenous Infusion of 2 µg/kg dexmede·
`tomldlne, 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 decllned by 20% from
`baseline and HR rose to 5% below baseline values. The he(cid:173)
`modynamlc profile did not show acute alterations after Intra·
`muscular administration. During the ti h after Intramuscular
`administration, MAP declined by 2 0% and HR decllncd by 10%.
`
`•Assistant Professor, Oep.,tment of Ane•'thesl:t. VA Mcd!cal Center.
`San Diego; and Department of Anesthesia, UCSD, School of Medicine.
`t Associate Professor, Department of Anesthesia, VA Medical Center,
`PaloA110; and Department of Anesthesia, Stanford Univer;lcy, School
`of Medicine.
`
`* Research Assistant, Oepanmem of Aneschcsla, VA Medical Center,
`
`Palo Alto.
`§Orlon Corporation FARMOS.
`V ~~sistam Professor, Department of Anesthesia, VA Mecical Center,
`Palo Alto; and Department of Anesthesia, Stanford University, School
`of Medicine.
`Received from the Dcpar1mcntsof Anesthesia, VA Medical Center,
`San Diego, California; UCSD, School of Medicine, San Diego, Cali·
`fornia; VA Medical Center, Palo Alto, California; and Sttnford Uni·
`versicy, School of Medicine, Stanford, California. Accepted for pub·
`Uc:ulon November 11, 1992. Supported by a grant from the Medical
`Research Council of Canada and the Orlon Corporation FARMOS.
`Address reprint requests 10 Or. Dyck: Dcpanmcm of Ancs1hcslology,
`VctcransAdministratlon Mcdicnl Center, 3350 l..aJolla Vl!l~ge Drive,
`San Diego, California 92161-9125.
`
`Co11cl11sl o11s1 The lntrnmusculnr ndmlnlstratlon of dexme(cid:173)
`detomldlne avoids the acute hemodynamlc changes seen with
`Intravenous administration, but results In similar hemody·
`namlc alter ations within ti h. (Key words: Hemodynamlcs.
`Pharmneoklnetlcs. SympathetJc nervous system, a 2 agonists:
`dexmedetomldlne.)
`
`THE a 2-adrenergic agonists are a new class of poten(cid:173)
`tially usefu l adjunctive anesthetic agents. Clonidine,
`the p rototypic a 2-adrenergic agonist, is the most widely
`used drug of this c lass of compoun ds and decreases
`anesthetic and analgesic requirements In surgical pa·
`ticms.' Furthermore, clonidine administered b efore
`anesthetic induction may also minimize intraoperative
`h emodynamic fluctuations and is an effective anxio lytic
`agent. Because clonidine has a long duration o f action
`an d is a partial agon ist with only modest select ivity for
`the a 2 versus the a, adrenoceptor , a second generation
`of a 2 agonists is now being developed in an attempt to
`overcome the p erceived shortc omings of clonidine in
`anesthetic settings. Dcxmedetomidine (1,620: 1 [a2 :
`a 1]) is more selective at the a 2 adrenoceptor t han is
`clonidine (220:1) and is a full agonist. 2
`To administer dexmedetomidine accurately, it is
`n cccs:sary to characterize the p harmacokinctic pcofilc
`u sing r elevan t doses via the in tended routes of admin(cid:173)
`istration, and to correlate side effects, such as hemo(cid:173)
`d ynam ic alterations , w ith the plasma con centrations of
`medication. Using a crossover study design , with dex(cid:173)
`medetomidine administered intravenously and intra(cid:173)
`muscularly, we characterized d exmedetomldine p har(cid:173)
`macokinetics and h emodynamic a lterat ions in ten
`healthy adult volunteers.
`
`Materials and Methods
`
`Su bjects
`
`After approval by t he Stan ford University Investiga(cid:173)
`tional Review Board, ten healthy male v olunteers were
`recruited for this study. The average age of the subjects
`was 35.5 yr (range 29-44 yr) and the average weight
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From: bttp:/lanestbesiology.pubs.asabq.org/pdfaccess.asbx?ul'l=/data/Joumals/J ASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`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 caffeine or
`alcohol consumption for the preceding 24 h. On arrival
`at the study site, an l S·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 dexmedetomidine concentra·
`tions. After fluid loading, 2 µg/kg dexmedecomidine
`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(cid:173)
`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
`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(cid:173)
`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 dexmedctomidine concentration
`was assayed. Blood sampling changed from arterial to
`venous at 4 h to minimize the length of time the vol·
`umcers were subjected to the presence of an arterial
`line.
`
`Dexmetletomidine Assay
`The plasma was assayed for dexmedetomidine con(cid:173)
`centration using a gas chromatograph (GC) with mass
`spectroscopy (MS) detection. 3 The pentatluorobenzoyl
`derivatives of dexmedetomidlne and the internal stan·
`dard detomidine were produced during extraction of
`the plasma into n-hexane in the presence of Na2C03
`
`Ancsthcslolob'Y· V 78, No 5, May 1993
`
`and pentafluorobenzoyl chloride. The organic phase
`was evaporated and the residue reconstituted in tolu·
`enc. A l·ul aliquot was injected onto a Hewlett-Packard
`fused silica capillary column cross linked with 5%
`phenyl methyl silicone (Part number 19091)-102,
`Hewlett-Packard Company, Little Falls, DE) of a Hew(cid:173)
`lett-Packard gas chromatograph (Model 5890A, Hew·
`lctt·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 MATTSQ 70, Finnigan MAT)
`using methane as the carrier gas was operated in neg(cid:173)
`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. I. The lower limit of quantltation for
`this GC/MS technique was 50 pg/ml, recovery of tri·
`tiated dexmedetomidine was 81 %, and the coefficient
`of variation for within-day assays at 7 5 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.
`
`Pbarmacoktnetic Analysts
`Moment Analysis. Moment analyses were performed
`on both the intravenous and intramuscular data to cal(cid:173)
`culate the model independent parameters: area under
`the concentration versus time curve (AUC), :1rea under
`the first moment of the concentration versus time curve
`(AUMC), clearance (Cl), volume of distribution (Vd.,),
`and mean residence time (MRT). Values for AUC and
`AUMC are intermediate steps in the calculations and
`arc presented for the sake of continu ity. The AUC was
`calculated using the trapezoidal method with linear
`interpolation when concentrations were increasing and
`log-linear interpolation when concentrations were de·
`creasing. 4 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 In (2) divided by the slope of the terminal phase.
`In similar fashion, the AUMC was calculated as the
`trapezoidal integration of the cu rve generated by mul·
`
`Downloaded From: bttp:/lanestbesiology.pubs.asabq.org/pdfaccess.asbx?ul"l=/data/Joumals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Exhibit 1021 - Page 814
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`815
`
`80386-based computer. Calibration signals were re(cid:173)
`corded from a Delta-Cal Transducer Simulator (Model
`650-905, Midvale, Utah) at 0, 50, 100, 150, and 200
`mmHg. 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 between-. wave
`peaks. The systolic and diastolic blood pressure, MAP,
`and heart rate were recorded for each wave on the ar(cid:173)
`terial pressure trace during the study. The · hemody(cid:173)
`namic data reported represents the median MAP and
`heart rate values for each 60-s period.
`
`Results
`Figure 1 shows the de:xmcdetomidine plasma con(cid:173)
`centration versus time profiles for all ten volunteers
`during the 5-min intravenous infusion and for the fol(cid:173)
`lowing 24 h. At 3 and 4 h after the infusion, simulta(cid:173)
`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.8 The rise in plasma concentration was probably
`not elution from storage sites in skeletal muscle, be(cid:173)
`cause the subjects remained supine from the start of
`
`10
`
`e "Cl)
`
`c
`u
`c:
`8
`
`Infusion
`
`20
`
`25
`
`10
`
`~
`c
`u
`c:
`8
`" c
`~
`E
`g
`.g
`() e
`"' " 0 0.1
`
`tiplying each p lasma concentration by its time. The
`volume of distribution at steady state was calculated as
`follows: 5
`
`Dosex AUMC
`Vd., = -
`-(A_ U_C)_2_
`
`Dosex T
`2 X AUC'
`
`where T was the duration of the infusion. Clearance
`was calculated as the ratio of dose to AUC:5
`
`CL= Dose
`AUC'
`
`and MRT as the ratio of Vd .. , to clearance: 5
`
`MRT= Vd.,.
`CL
`
`The bloavallabillty of Intramuscular dexmede1om!dlne
`was calculated as the ratio of the AUC after intramus(cid:173)
`cular versus intravenous administration of the same
`dose:
`
`AUC1v
`.1 b'l'
`% B .
`toavat a 1 tty = - - - X 1
`00
`.
`AUC1i11
`
`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(cid:173)
`tramuscular dexmedetomidine were determined
`through least-squares deconvolution of the intramus(cid:173)
`cular concentration versus time function w ith the in(cid:173)
`travenous unit disposition function (UDF)6
`7 for each
`•
`individual patient. Knowing that;
`Cp = 1*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(cid:173)
`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(cid:173)
`corded by a TEAC R-71 recorder and simultaneously
`digitized on a DT280 I Data Translation A/D board at
`128 Hz with 12-bit resolution to the hard disk o f an
`
`Anesthesiology, V 78, No 5, May 1993
`
`0
`
`8
`
`ll
`Time, hllurs
`Fig. L Dcxmedetomldlne Intravenous plasma concentration
`vcrs11s time.
`
`16
`
`20
`
`24
`
`Downloaded From : bttp:/lanestbesiology.pubs.asabq.org/pdfaccess.asbx?ud=/data/Jounials/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Ex hibit 1021 - Page 81 5
`
`

`

`816
`
`DYCK ET AL.
`
`Table 1. Moment Anal ysis Intravenous (IV) Dat:a
`
`Subject
`No.
`
`AUCIVOto
`lnlfinlty
`(ng ·min· ml"')
`
`Terminal
`Half.Ille
`(min)
`
`At:C
`(% und« Data)
`
`aearance
`(L/mln)
`
`V.,(L)
`
`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
`95
`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
`
`AUMC
`(ng·min2)
`
`437,000
`172,000
`314,000
`136,000
`135,000
`188,000
`121,000
`65,900
`63,400
`72.300
`
`141,000
`79,!JOO
`
`AUMC
`(°k under Data)
`
`19
`60
`53
`66
`63
`63
`65
`89
`73
`84
`
`68
`11
`
`MRT
`(min)
`
`1207
`480
`547
`405
`440
`530
`404
`277
`241
`285
`
`401
`112
`
`AUC = area under tna curve; v .. • voome or distribution at steady state; AUMC = area under first moment curve; MRT = mean residence time.
`• Subject 1 excluded from summaiy statistics.
`
`known zero order input function to arrive at the cal(cid:173)
`culated unit disposition function (UDF) for each sub(cid:173)
`ject. Deconvolution was constrained to be positive and
`unimodal to restrict the output to physiologically
`meaningful results. Figure 3 shows average intravenous(cid:173)
`UDF (±SD) of the ten subjects calculated through the
`deconvolution technique. The resulting UDP for dex(cid:173)
`medetomidine after intravenous administration was
`deconvolved against the concentration versus time
`profile after intramuscular administration on a patient(cid:173)
`by-patient basis to produce the rate of intramuscular
`absorption shown in figure 4. Integration of the ab(cid:173)
`sorption rate over time after intramuscu lar injection
`(figure 4) yields a total systemic dose of 13 3 µg and a
`
`1.4
`1.2
`]
`1.0
`l 0.8
`g 0.6
`8 0.4
`0.2
`
`1.0
`
`~
`" U' 0.8
`" 8
`" .s 0.6
`..., ·a
`2 u
`'O " ~ 0
`.. 8
`i::i 0.2
`t(l
`~ o.o
`
`0.4
`
`10
`Time, min
`
`15
`
`20
`
`the study until the 240-min sample, and were only
`starting to ambulate by 300 min. The plasma dexme·
`detomldine concentrations after intravenous adminis·
`Cration decreased to less than the limit of quantltation
`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. 511
`± 0.125 L/min, Vd .. was 194 ± 28.7 L, andMRTwas
`401±112min.
`figure 2 shows the plasma concentration versus time
`profile after intramuscular administration of 2 ,ug/kg
`dexmedetomidine. The time to peak plasma concen(cid:173)
`tration was 13 ± 18 min and the mean peak concen(cid:173)
`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(cid:173)
`tion vetsus time curve for all subjects was 243 ± 78
`ng · min- 1
`• m1- 1 and the average bioavailability was 73
`± 11% (table 2) .
`The concentration versus time profile for the intra(cid:173)
`venous administrations was deconvolved against the
`
`Anesthesiology, V 78, No 5, May 1993
`
`0
`
`8
`
`12
`Time, hours
`Fig. 2. DexmedetomJdlne intramuscular plasma conccntratJon
`versus time.
`
`16
`
`24
`
`Downloaded From : bttp:/lanestbesiology.pubs.asabq.org/pdfaccess.asbx?ul'l=/data/Joumals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Ex hibit 1021 - Page 816
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDINE
`
`817
`
`Table 2. Moment An alysis Intramuscular (IM) Data
`
`Subject
`No.
`
`AUC IM 0 to Infinity
`(ng ·min· m1-•1
`
`Term In al
`Half-Ille
`(min)
`
`AUC
`(% under Data)
`
`Bloavailablllty
`(o/o)
`
`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 under the 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
`5
`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 µg systemically absorbed/
`158 µg average intramuscular dose). Figure 5 shows
`the cumulative absorption over time, as a percent of
`total absorption. The mean intramuscular dose was 158
`µg resulting in a bioavailability of intramusclllar-to-in(cid:173)
`travenous dosing of 84% using deconvolution analysis.
`The AUMC of figure 4 was 277 µg · h 2• The mean ab(cid:173)
`sorption time (MAT) calculatedasAUMC/AUCwas 2.08
`h, and the mean first order rate constant (Ka) for intra(cid:173)
`muscular absorption as the reciprocal of MAT was
`0.48 h- 1
`•
`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 dcxmcdctomidine occurred at S 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(cid:173)
`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(cid:173)
`travenous and intramuscular dexmedetomidinc, re·
`spectively. The decline in HR after intravenous dex(cid:173)
`medetomidine was 27% below baseline 4- 5 min after
`
`~ 100
`.;-
`;;
`
`b¢bbb
`
`!
`
`1111111
`1111111
`
`l l
`
`0
`
`8
`
`12
`Time, hours
`Fig. 4. Dexmedetom ldlne Intramuscular rate of absorption
`(± SD) vet•s11S time.
`
`16
`
`20
`
`24
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded Fl'om : bttp:/lanestbesiology.pubs.asabq.org/pdfaccess.asbx?ul'l=/data/Joumals/JASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Ex hibit 1021 - Page 817
`
`100 ~
`
`~ :s
`
`'O
`tll?
`0
`c
`8
`
`10
`
`~!6
`b~b 0
`
`60
`
`0
`
`6
`
`100
`
`~
`:§
`~
`~ 10
`u·
`c
`
`8 ..,
`.s
`'O
`'§
`B .,
`-g
`~
`0 0.1
`
`4
`
`12
`Time, hours
`Fig. 3, Mean unit disposition function ± SD of lntmvenous
`dexmedetomldlne.
`
`16
`
`20
`
`Z4
`
`~ l!
`
`0
`
`10
`
`"' c
`""
`e-0
`B <
`" .s
`:!! e
`B
`~ 0.1
`~
`" 0
`~ O.Ql
`
`

`

`818
`
`DYCK ET AL.
`
`.,
`0
`·~
`~
`.Q
`<I;
`u
`
`100
`
`80
`
`~ ~
`~ 8 60
`a5
`g ~
`:§ ~
`e
`g .,
`~
`" Cl
`~
`
`'O
`
`40
`
`20
`
`0
`
`130
`
`l:W
`
`MAP-IM
`
`11! I !\ !I \! II 11! I! I I I
`
`1111
`
`110
`&" 100
`E
`e 90
`~ :a 80
`
`70
`
`60
`
`S-0
`
`0
`
`lO
`
`100
`Time, min
`Fig. 7. Mean arterial p r essu re (± SD) o f Intramuscular dex(cid:173)
`medetomldlne.
`
`ISO
`
`200
`
`0
`
`8
`
`12
`Time, hours
`Fig. 5. Dexm edetom ldlne cum ulative absorption(± SD) VCl'SllS
`t ime.
`
`16
`
`20
`
`24
`
`starting the infusion. Again, this was not seen after In·
`tramuscular injection. Dexmedetomidine given by in(cid:173)
`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(cid:173)
`istration preclude the use of dexmedetomidinc as a
`rapid intravenous infusion or bolus. Compartmental
`pharmacokinetic analysis would be required to admin(cid:173)
`ister dexmedecomlcline via a computer-comrolled in(cid:173)
`fusion pump, but dexmcdctomid inc would be more
`commonly ad ministered by slow intravenous infusions
`
`to steady state or by lmramuscular injections, for which
`moment analysis is adequate. Moment analysis is model
`independent and allows calculation of fundamental
`pharmacokinetic par:1metcrs, such as volume of distri(cid:173)
`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 h were slightly greater than arterial concentrations
`and the calculated area under the curves may be slightly
`greater than an entirely arterial concentration profile.
`Dexmedctomidine appears to have systemic clearance
`of approximately 0.5 L/min, approximately one-half
`of hepatic blood flow. Overall, the volume and clear(cid:173)
`ance nre fairly similar to those of fentnnyl9 with nn ex-
`
`130
`
`t:W
`
`110
`&" 100
`e
`E 90
`~ 80
`
`10
`
`60
`
`50
`
`MAP - IV
`
`90
`
`80
`
`.D
`
`.,.
`e 70
`<i ;;
`" 60
`t:
`~ so
`
`40
`
`30
`
`HR-IV
`
`so
`
`100
`Time, min
`Pig. 6. Mean arterial pr essure (± SD) o f intravenous dexm e(cid:173)
`deto mldine.
`
`ISO
`
`100
`
`zso
`
`0
`
`so
`
`100
`Timo-,min
`Fig. 8. Mean h eart r ate (± SD) of Intravenous dexmedetoml(cid:173)
`dln e.
`
`200
`
`250
`
`Anesthesiology, V 78, No 5, May 1993
`
`Dowuloaded From: bttp:/lauestbesiology.pubs.asabq.org/pdfaccess.asbx?ul"l=/data/J oumals/J ASA/931315/ ou 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Exhibit 1021 - Page 81 8
`
`

`

`PHARMACOKINETICS AND HEMODYNAMICS OF DEXMEDETOMIDlNE
`
`819
`
`HR - IM
`
`90
`
`80
`
`E 70
`0..
`~ u
`;:;
`"' 60
`'5
`" :i: so
`
`40
`
`30
`
`sumption of linearity is violated frequently in both
`compartmental pharmacokinetic analysis and moment
`analysis. As violation of linearity is part of model mis(cid:173)
`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(cid:173)
`kinetics based on the assumption of linearity provide
`a succinct, easily estimated, and clinically useful de(cid:173)
`scription of pharmacokinetic behavior. The hyperten(cid:173)
`sion and bradycardia seen after intravenous dexmede(cid:173)
`tomidine were not seen after intramuscular adminis(cid:173)
`tration.
`The peak plasma concentrations were an order of
`magnitude lower after intramuscular administration.
`On the assumption that the differences in hemodynamic
`profiles may have been a result of concentration-de(cid:173)
`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 cou ld be
`achieved through an infusion of dexmedetomidine at
`0. 511 µg/min. The plasma concentration will asymp(cid:173)
`totically approach the targeted steady-state concentra(cid:173)
`tion of 1.0 ng/ml and would be very close to the steady(cid:173)
`state concentration after three elimination hall-lives or
`1, 1 5 5 min. If it is desirable to attain the target con(cid:173)
`centration b efore 19.25 h, a loading dose, calculated
`as targeted steady-state concentration times Yd« or 194
`µg, may be administered and followed by the mainte(cid:173)
`nance infusion. The loading dose should not be ad(cid:173)
`ministered as a bolus, but can be given as an infusion
`over 30-45 min with minimal increased risk of adverse
`hemodynamic 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(cid:173)
`ogy for this loss of consciousness is the sympatholytic
`effect of the dexmedetomidine leaving unopposed va(cid:173)
`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 slmllar episodes.
`
`0
`
`so
`
`100
`Time, min
`Fig. 9. Mean heart rate(± SD) of lntramuscular dexmedetom(cid:173)
`ld lne.
`
`150
`
`200
`
`250
`
`tensive tissue distribution (fentanyl Vd., approximately
`300 L) and a moderately large hepatic clearance (i.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. 10 The MRT ls 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(cid:173)
`dine was between 70% and 80%. On average, peak
`plasma concentrations of dexmedetomidine were ob·
`tained within l 5 min after intramuscular injection, al(cid:173)
`though the time to peak concentration after intramus(cid:173)
`cular injection varied widely. The intramuscular ab(cid:173)
`sorption prolile was b iphasic with early rapid
`absorption.
`Intravenous dexmedecomldine as a rapid infusion
`caused bi phasic changes in HR and MAP similar to those
`seen after administration of clonidine. 11
`• 12 The clinical
`utility of intravenous dexmedetomidine will be limited
`by these hemodynamic alterations. Bolus intravenous
`administration of dexmedetomidine would be unwise
`in most circumstances. It is possible that dexmedetom(cid:173)
`id in e pharmacokinetics are not linear secondary to the
`concentration·dependent hemodynamic alterations.
`Many of the drugs used in anesthesia practice (i.e., pro(cid:173)
`pofol and thiopental) affect hemodynamics and prob(cid:173)
`ably have nonlinear pharmacokinetics. Thus, the as-
`
`Anesthesiology, V 78, No 5, May 1993
`
`Downloaded From : http:/lanesthesiology.pubs.asahq.org/pdfaccess.ashx?ul'l=/data/J ounials/J ASA/931315/ on 08/09/2016
`Petitio n for Inter Partes Review of US 8,242, 158
`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(cid:173)
`centrations that result arc less than those after a 5-min
`intravenous infusion with the same dose, and hemo(cid:173)
`dynamic alterations are less severe.
`
`References
`1. Aho M, Lehtinen A·M, Erkola 0, K.'tlllo A, Kortilb K: The effect
`of intravenously administered dcxmedetomldinc on periopcrative
`hemodynamics and lsoflurane requirements In patients undergoing
`abdominal hysterectomy. ANmmtEStOLOGY 74:997- 1002, 1991
`2. Vlrtanen R, Savolo JM, Sa2no U, Nyman L: Char:ictcriz:uion of
`selectivity, specificity, and potency of mcdetomidlnc as an alpha-2
`adrcnoccptor agonist. Eur J Pharmacol 150:9-14, 1988
`3. Vuorllchto L, Salonen JS, Anttila M: Plcogram level determi(cid:173)
`nation of medctomidinc In dog scrum by capillary &'ts chromatog(cid:173)
`raphy with negative Ion chemical ionlza1lon mass spectrometry. J
`Chromatogr497:282-287, 1989
`4. Gibaldl M, Perrier D: Absorption kinetics and bioavailabillty,
`Pharmncoklnetics, 2nd edition. Edited by Swarbrick J. New York,
`Marcel Dekker. 1982. pp 145-198
`
`5. Glbaldi M: Dlopharmaccutlcs and Cllnical Pharmacokinetlcs,
`4th edition. Malvern, Pennsylvania, I.ca & Fcbigcr, 1991 , pp 14-23
`6. Verona D: An Inequality-constrained least sqllares deconvolution
`method. J Pharmacokinet D!opharm 17:269-28?, 1989
`7. Streisand JD, Varvel JR, Stanski DR, LcMaire L, Ashburn MA,
`Hague m, Tarver so. Stanley TH: Absorption and bioavailabillty of
`oral transmucosal fentanyl citrate. ANES1llllSIOLOGY 75:223-229, 1991
`8. Chiou WL: The phenomenon and rationale of marked depen·
`dcncc of drug concemratlon on blood sampling site lmpllcttlons In
`phnrmacokinetlcs pharm2codynomics, toxicology and ther:1pcutlcs
`(pan I). Clin Pharmacoklnct 17:175-199, 1989
`9. Scott), Stanskl DR: Decreased fcmanyl and alfcmanil dose re·
`qutremcnts with age: A s1mu1mneous pharmacok1ncuc and phar(cid:173)
`macodynamie cwluatlon.J Ph.irmacol Exp Thcr 240:159- 166, 1987
`JO. Hughes MA, Glass PSA,JacobsJR: Context-sensitive balf-tlmc
`in muhlcompanmcnt pharmacoklnetics models for Intravenous an·
`CStilCtiC drugs. ANr:STHESIOLOGY 76:334-341, 1992
`11. Rhee HM , L.'tpp JD: Arc opioid receptors involved ill the bra·
`dycardlc and hypotensive action of clonidine. Am J Hypertcns 1:
`2495-2545, 1988
`12. Frisk-Holmberg M: Effect of clonidlnc at steady-state on blood
`pressure Jn spontaneously hypertensive rats: lnceractlon of various
`alpha-aclrenoccptoramagonislS. Acta Physlol Scand 120: 3 7-42, 1984
`
`An~sthcsiology, V 78, No 5, May 1993
`
`Downloaded F1·om : http:/lanesthesiology.pubs.asahq.org/pdfaccess.ashx?ul'l=/data/Joul'Dals/J ASA/931315/ on 08/09/2016
`Petition for Inter Partes Review of US 8,242, 158
`Amneal Pharmaceuticals LLC - Exhibit 1021 - Page 820
`
`