`
`European Journal of
`Clinical Pharmacology
`© by Springer—Verlag 1978
`
`Absorption of Oral and Intramuscular Chlordiazepoxide"<
`
`D. J. Greenblatt**, R. I. Shader, S. M. MacLeod, E. M. Sellers, K. Franke, and H. G. Giles
`
`Clinical Pharmacology Unit, Massachusetts General Hospital, Boston, Psychopharmacology Research Laboratory,
`Massachusetts Mental Health Center, Boston, USA, Clinical Institute, Addiction Research Foundation, Toronto,
`and Departments of Medicine and Pharmacology, University of Toronto, Canada
`
`Summary. The absorption of oral and intramuscular
`(i. m.) chlordiazepoxide hydrochloride (CDX - HCl)
`was compared in two pharmacokinetic studies. In
`Study One, single 50-mg doses of CDX - HCl were
`administered orally and by i. m. injection to 14 heal—
`thy volunteers using a crossover design. Whole—blood
`concentrations of chlordiazepoxide (CDX) and its
`first active metabolite, desmethylchlordiazepoxide
`(DMCDX), were determined in multiple samples
`drawn after the dose. Mean pharmacokinetic vari—
`ables for CDX following oral and i. In. administra—
`tion, respectively, were: highest measured blood con-
`centration, 1.65 vs 0.87 pg/ml (p < 0.001); time of
`highest concentration, 2.3 vs 7.6 h after dosing (p <
`0.001); apparent absorption half-life, 0.71 vs 3.39 h
`(p < 0.001). Biphasic absorption after i. m. injection,
`consistent with precipitation at the injection site, was
`observed in 9 of 14 subjects. Based upon comparison
`with previous intravenous data, the completeness of
`absorption was 100% for oral vs 86% for i. m. CDX
`' HCl (p < 0.1). In Study Two, 28 male chronic
`alcoholics with clinical manifestations of the acute
`
`randomly
`syndrome were
`alcohol withdrawal
`assigned to one of four treatment conditions: 50 or
`100 mg doses of CDX - HCl, by mouth or by i. m.
`injection. Concentrations of CDX and DMCDX,
`determined in plasma samples drawn every 20 min
`for 5 h following the dose, were significantly higher
`after oral administration of a given dose than at cor—
`responding points in time after i. m. injection after
`the same dose. Thus absorption of oral CDX is
`reasonably rapid and complete, whereas the absorp—
`tion rate of i. m. CDX is slow.
`
`* Supported in part by Grant MH-12279 from the United States
`Public Health Service (Drs. Greenblatt and Shader)
`** To whom reprint requests should be sent
`
`Key words: Chlordiazepoxide, benzodiazepines,
`pharmacokinetics,
`bioavailability,
`intramuscular
`injection, alcohol withdrawal.
`
`Chlordiazepoxide hydrochloride (Libriurn) is com—
`monly administered by intramuscular injection in
`clinical situations requiring sedative effects of rapid
`onset [8]. Previous studies, however, suggest that the
`rate of absorption of chlordiazepoxide following
`intramuscular injection is
`slow [1, 4, 12]. Oral
`administration, on the other hand, gives relatively
`rapid and apparently complete absorption [1, 10, 12,
`19]. The present study compared the rate of absorp—
`tion and absolute systemic availability of oral and
`intramuscular chlordiazepoxide administered to a
`series of healthy male and female volunteers. Data is
`also presented on the rate of chlordiazepoxide
`absorption following these two routes of administra—
`tion in a series of patients receiving the drug for
`treatment of acute alcohol withdrawal.
`
`Materials and Methods
`
`Study One
`
`Fourteen healthy volunteers (6 male and 8 female),
`aged 21 to 38 years, participated after giving
`informed consent. All were free of identifiable medi—
`
`cal or psychiatric disease, and were taking no
`psychotropic medications on a regular basis. No
`other drugs were ingested during the study.
`Each subject received single 50—mg oral and
`intramuscular (i. m.) doses of chlordiazepoxide hy-
`drochloride1 (CDX - HCl) in random sequence with
`
`1 Librium®, Roche Laboratories, Nutley, N. J.
`
`003 1—6970/78/0013/0267/$01.60
`
`AstraZeneca Exhibit 2117 p. 1
`InnoPharma Licensing LLC v. AstraZeneca AB IPR2017-00904
`
`
`
`268
`
`at least one week elapsing between trials. Oral CDX
`- HCl was administered as two 25-mg capsules with
`100 ml of water following an overnight fast. CDX -
`HCl was prepared for i. m. injection by addition of
`2 ml of Special Intramuscular Diluent (the customary
`solvent) to 100 mg of sterile powder. One ml of this
`solution (50 mg of CDX - HCl) was administered as
`a single deltoid intramuscular injection, using a 2.5-
`cm gauge—20 hypodermic needle.
`Venous blood samples were drawn into hep-
`arinized tubes from an indwelling Butterfly cannula,
`or by venipuncture, prior to drug administration, and
`at 0.5, 1.0, 1.5, 2.0, 3, 4, 6, 8, 12, 24, 36, 48, 60 and
`72 h after administration. Whole blood samples were
`immediately frozen until the time of assay. Concen-
`trations of chlordiazepoxide (CDX) and its first phar-
`macologically active metabolite, desmethylchlor-
`diazepoxide (DMCDX), were determined by a spec-
`trophotofluorometric method [13, 18, 20], having
`sensitivity limits of approximately 0.1 jig/ml for CDX
`and 0.05 pig/m1 for DMCDX.
`Blood concentrations of CDX for individual sub—
`
`jects, as well as “composite” data points formed by
`the across—subject mean concentrations at corre—
`sponding points in time, were analyzed by iterative
`weighted nonlinear least-squares regression analysis
`[22]. The iterative process was allowed to proceed
`until convergence criteria were met, or a total of 50
`iterative
`steps was
`completed. Following oral
`administration of CDX - HCl, each set of data was
`fitted to the following two functions:
`
`C = —-(A + B)e_kat + Ae‘o‘t + Be‘fit
`C = A(e_’3t - e4“).
`
`(1)
`(2)
`
`In both of these equations, C is the blood CDX con-
`centration at time t after dosage. A and B are hybrid
`intercept coefficients, and a and [3’ are hybrid expo—
`nents representing the “distribution” and “elimina-
`tion” phases of the blood concentration curve [3, 5,
`7, 23]. k3 is the hybrid exponent representing the
`apparent first-order absorption rate constant. It was
`used to calculate the apparent first—order absorption
`half—life (ti/2,) as follows:
`
`mm = (1n 2)/k, = 0.693/ka.
`
`(3)
`
`For each set of data the choice between Equations
`(1) or (2) as functions of “best fit” was determined
`by comparison of the sum of squares of weighted
`residual errors, and by assessment of the randomness
`of “scatter” of actual data points about the fitted
`function [2]. Since CDX concentrations generally fell
`below the limits of reliable quantitation after 48 h,
`only those data points obtained up to 48 h after dos—
`age were used for analysis.
`Following i. m.
`injection of CDX - HCl, data
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`points were again fitted to the above two functions.
`In some cases, however, neither function provided an
`adequate fit, and it was necessary to provide for two
`rates of absorption, termed the “rapid” and “slow”
`rates. The following function was applicable:
`
`C = ~Ae‘klt —- Be‘klt + (A + B)e‘fit.
`
`(4)
`
`The quantities C, t, A, B, and fl have the same mean-
`ing as in Equations (1) and (2). k1 and k2 are appa—
`rent
`first-order
`rate constants
`representing the
`“rapid” and “slow” phases of drug absorption. This
`approach has been used previously to describe
`absorption of i. m. chlordiazepoxide [1], phenytoin
`[14], and quinidine [7]; it is consistent with precipita—
`tion of the drug at the injection site, followed by slow
`redissolution.
`
`Thirteen of the fourteen subjects had participated
`in a previous study of intravenously-administered
`CDX ' HCl [9]. In these subjects absolute systemic
`availability of oral and i. m. chlordiazepoxide was
`estimated by comparison of the area under the 48—
`hour CDX blood concentration curve following the
`two extravascular modes of administration to that
`
`observed in the same subject following intravenous
`injection of the same dose. Also assessed was the
`area under the 72—hour blood concentration curve
`
`for DMCDX following oral and i. In. administration
`in comparison with intravenous injection.
`
`Study Two
`
`Twenty—eight male subjects participated after giving
`informed consent. All were chronic alcoholics who
`
`presented at emergency treatment facilities with signs
`and symptoms consistent with acute alcohol with—
`drawal.
`In all subjects administration of chlor-
`diazepoxide as a sedative agent was clinically indi-
`cated. The following patient characteristics were
`recorded: age, body weight, total serum bilirubin,
`serum albumin, serum creatinine, alkaline phospha—
`tase, and serum glutamic oxaloacetic transaminase
`(SGOT).
`Subjects were randomly assigned to one of the
`following four treatment groups:
`a) 100 mg of CDX - HCl by mouth with water;
`b) 100 mg of CDX ' HCl by deltoid i. In. injection;
`c) 50 mg of CDX - HCl by mouth with water;
`d) 50 mg of CDX - HCl by i. m. injection.
`Comparability of the treatment groups with respect
`to the above characteristics was assessed by one-way
`analysis of variance [21].
`into
`drawn
`Venous
`blood
`samples were
`heparinized tubes from an indwelling Butterfly can—
`nula, or by venipuncture, prior to drug administra-
`tion, and every 20 min thereafter until a total of 5 h
`
`AstraZeneca Exhibit 2117 p. 2
`
`
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`269
`
`Table 1. Pharmacokinetics of chlordiazepoxide absorption in study one
`
`Subject
`
`Maximum blood concentration
`(Mg/ml)
`
`Time of maximum concentration
`(h after dose)
`
`First—order absorption half—life
`
`oral
`
`i. m.
`
`oral
`
`i. m.
`
`oral
`(hours)
`
`i. m.
`
`First phase
`(min)
`
`Second (dominant)
`phase (h)
`
`JMK
`GLN
`DJL
`SN
`LDF
`AP
`RIS
`RAD
`KMJ
`LM
`DJG
`RRG
`AJS
`JSH
`Mean
`(:SE)
`
`0.96
`1.39
`1.23
`2.07
`0.66
`1.70
`1.10
`1.92
`1.00
`1.93
`0.82
`1.41
`0.66
`1.58
`0.77
`1.15
`0.48
`1.16
`0.52
`1.80
`1.05
`1.88
`1.13
`1.64
`1.18
`1.90
`0.56
`1.64
`0.87
`1.65
`(i007)
`(i008)
`L____l
`t = 11.30, p < 0.001
`
`12
`2.5
`2.5
`1.0
`8
`1.5
`8
`6
`12
`1.0
`6
`2.5
`6
`1.5
`8
`6
`2.5
`1.5
`8
`1.0
`8
`1.0
`6
`1.0
`8
`1.5
`12
`4
`7.6
`2.3
`(:08)
`(i0.5)
`|—_l
`t = 6.47, p < 0.001
`
`0.64a
`—-
`0.48
`1.18
`0.2b
`0.33
`3.90
`~0b
`0.54
`2.05
`—
`0.35
`4.31
`7.3
`0.42
`1.99
`—
`0.70
`4.59
`13.1
`1.12
`0.80
`—
`3.12
`9.63
`11.2
`0.92
`5.06
`8.5
`0.32
`2.62
`9.6
`0.94
`6.21
`30.9
`0.86
`2.47
`~0b
`0.58
`2.05
`—
`0.69
`3.39
`9.0
`0.71
`(i066)
`(132)
`(i019)
`l—_l
`t = 3.65, p < 0.005
`
`1.46 0.78 1.5 0.64 0bComposite 2.34
`
`
`
`
`
`
`
`
`
`
`
`
`
`a Lag time = 0.85 h prior to the start of absorption
`b
`Iterative analysis yielded a very large rate constant for the first phase of absorption, corresponding to a very small or essentially zero
`half-life value
`
`had elapsed since drug administration. At that point
`the study terminated, and subsequent treatment was
`administered by the attending physician according to
`clinical needs.
`
`Plasma was separated and frozen until the time of
`assay. Plasma concentrations of CDX and DMCDX
`were determined by spectrophotofluorometric assay
`[13, 18, 20].
`Differences between oral and i. m. administration
`
`between subject groups receiving the same dose were
`assessed by calculating across-subject mean plasma
`concentrations of CDX and DMCDX at correspond-
`ing points in time. Differences in plasma levels were
`tested using Student’s unpaired two-tailed t—test [21].
`
`Results
`
`Study one
`
`All subjects complained of moderate to severe local
`pain associated with the intramuscular injection. The
`injection caused elevations in serum creatinine phos-
`phokinase concentrations in the individuals in whom
`this was measured [11].
`
`Highest measured CDX concentrations following
`oral administration averaged 1.65 jig/ml, and were
`attained an average of 2.3 h after dosage (Table 1).
`The blood concentration curves in all subjects were
`adequately described by functions having the form of
`Equation (1) or (2)
`(Fig. 1). The mean apparent
`first—order absorption half-life was 0.71 h (Table 1).
`After
`i. m.
`injection, however, highest measured
`levels averaged only 0.87 jig/ml and were reached at
`an average of 7.6 h after dosing. Both of these values
`were significantly lower and later, respectively, than
`those observed in the same subjects following oral
`administration of the same dose. Two phases of
`absorption were observed in nine of the fourteen
`subjects (Fig. 1); the initial rapid phase in these indi-
`viduals proceeded with a mean apparent half—life of
`9.0 min (Table 1). In the other five subjects, a single
`phase of absorption was evident, although in one
`subject (JMK) a lag time of 0.85 h elapsed prior to
`the start of absorption (Table 1). The mean absorp-
`tion half—life for all 14 subjects following i. m. injec-
`tion (using the second or dominant phase for the 9
`“biphasic” individuals) averaged 3.39 h, which was
`significantly longer than that observed following oral
`administration of the same dose (Table 1). Thus i. m.
`
`AstraZeneca Exhibit 2117 p. 3
`
`
`
`----- —— x = Oral
`
`—-— A = Intramuscular
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`\
`i X\‘x
`K
`‘\ X
`I
`X
`ll
`\~\
`A
`I
`'A__\_~< \E'
`: A/A
`\\\.\‘
`\_
`AA
`\ \
`\-\A
`24
`‘x A
`\ \ \‘ ~
`X
`
`I X
`
`"A
`.
`
`“x
`x
`
`2
`,’ ‘\
`.‘ X X“
`1
`A.
`I
`:
`A
`X
`km~\
`A
`
`‘x\
`
`x‘
`
`A
`
`-
`
`\\
`
`\
`
`\
`
`\.
`
`x
`x “x
`
`A
`
`\~
`
`‘x
`
`\\‘ X
`\
`
`ms.
`
`|
`
`|
`
`1
`
`1
`
`__J
`
`D..J.L.
`.L
`
`L
`
`I
`
`\
`
`A
`
`\'\
`A
`I \ I
`
`*x‘ax‘
`‘ g x
`.A'f‘i‘K‘» A
`
`a4
`.
`
`\- A ““ ~A
`\.
`\.
`
`xx
`
`\
`
`3“
`\
`A
`/A‘ \. A \
`\ \\
`
`X
`‘1
`
`A
`
`270
`
`2 .0
`
`1 .0
`
`2 O _ 5
`E
`B"
`3
`x:
`"1%, 0.2
`E
`§
`1
`8 O _
`:8“ 2 .o
`g
`._
`E: 1
`a
`3
`E
`o
`E o 5
`U '
`
`.
`
`o , 2
`
`O
`
`'
`
`
`
`\X
`\- \\\
`A\
`
`‘\
`
`\\
`
`L I M .
`‘
`8.0
`
`D .J IQ
`‘\
`L
`I
`h
`I
`1
`I \x‘
`8.0
`16.0 24.0 32 .0 40.0 48.0 0
`Hours after dose
`
`1
`l
`16.0 24.0
`
`1
`I
`I
`32 .0 40.0 48.0
`
`Fig. 1. Blood concentrations of
`chlordiazepoxide, together with
`calculated pharmacokinetic func—
`tions of best fit, in 4 subjects fol—
`lowing oral and intramuscular
`administration of 50 mg of chlor-
`diazepoxide hydrochloride
`
`
`
`
`
`Bloodconcentration(pg/ml)
`
`2.0
`
`Chlordiazepoxide
`
`ix
`: XXX
`i
`\x \/Oral
`.'
`“>0
`1
`.A/Aia\‘A<‘\
`5 fig\
`\'\\\
`:f Intramuscular
`';\\
`{a
`\s\
`0.4— g
`1
`\\\.\
`0,3— I
`E
`\s
`
`O
`6.0
`12.0
`18.0
`24.0
`30.0 36.0 42,0 48.0
`
`1 .O r
`
`0-6‘
`
`0.2
`
`\Q
`
`0.6 i—
`
`Q4-
`0.3 "
`O 2 ..
`'
`
`Desmethylchlord iazepoxide
`x——-—><_________ __
`‘91:»-
`
`Oral
`
`{I
`
`./A\
`Intramuscular
`
`N§‘IQ\\,
`“\\“\\I\A
`“x
`
`)IZX
`A/
`,I
`X
`/
`>,<
`.
`
`0.1L 1;: A/l
`1
`1
`I
`J
`|
`l
`I
`6.0
`12.0
`18.0 24.0 30.0
`36.0 42.0 480
`Hours after dose
`
`
`
`CDX ' HCl is absorbed more slowly than after oral
`administration (Fig. 1 and 2).
`In most subjects,
`absorption proceeds as a biphasic process, consistent
`with precipitation at the injection site.
`The absolute systemic availability of oral CDX -
`HCl relative to intravenous administration averaged
`102.4%, which was not significantly different from
`100%. Following i. m. CDX - HCl, systemic avail-
`ability averaged 86.1% of the corresponding intrave—
`nous value; the difference between this value and
`100% availability approached but did not reach
`statistical significance (t = 1.79, p < 0.1) (Fig. 3), as
`did the difference between oral and i. m. CDX - HCl
`
`(paired t = 1.99, p < 0.1). Following oral CDX -
`HCl, the area under the 72-hour blood concentration
`curve for DMCDX averaged 115.7% of the intrave-
`nous value. After i. m. injection, the area under the
`
`Fig. 2. Mean blood concentrations of chlordiazepoxide and
`desmethylchlordiazepoxide
`following oral
`and intramuscular
`administration of 50 mg of chlordiazepoxide hydrochloride. Each
`point is the mean value for all 14 subjects at the corresponding
`time. Also shown are the calculated pharmacokinetic functions of
`best fit for oral and intramuscular chlordiazepoxide
`
`AstraZeneca Exhibit 2117 p. 4
`
`
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`271
`
`8 200.0 ”
`
`Chlordiazepoxide
`
`200.0 »
`
`Desmethylchlordiazepoxide
`x
`A
`
`X
`
`X
`
`x
`
`A
`A
`
`x x X X
`. . . .
`. .
`. . . . . .. . .. . . ... . . . .. . . . . . . ..A.A-A.A . . . . . . ...
`:-
`X x
`A A
`
`"
`><
`A
`x
`
`x
`Oral
`
`A
`A
`A
`Intramuscular
`
`E5
`
`E M
`
`8:
`g
`S
`.
`
`U x
`
`3 E
`
`175.0-
`
`‘
`
`6
`E
`m
`a 125.0—
`
`O _
`
`"
`
`_
`
`(b
`\T 50.0
`:5
`E 25.0 -
`
`
`
`3 NE
`
`< F
`
`O _ —l—l—.
`
`ig. 3. Area under the 48-hour blood concentration curve for
`chlordiazepoxide following oral and intramuscular administration
`of 50 mg of chlordiazepoxide hydrochloride. Each value is ex—
`pressed as a percentage of the corresponding intravenous value in
`the same subject. Also shown are mean ($SE) values for all 13
`subjects in each group
`
`Fig. 4. Area under the 72-hour blood concentration curve for
`desmethylchlordiazepoxide
`following oral and intramuscular
`administration of 50 mg of chlordiazepoxide hydrochloride. Each
`value is expressed as a percentage of the corresponding intrave-
`nous value in the same subject. Also shown are mean ($SE) val-
`ues for all 13 subjects in each group
`
`Table 2. Characteristics of patients in study two
`
`
`
`Mean ($SE) for Each Treatment Group
`
`100 mg
`p.o.
`
`100 mg
`i.m.
`
`50 mg
`p. o.
`
`6
`37.3 ($4.2)
`72.9 ($5.7)
`
`7
`42.9 (:3.2)
`63.0 (:3.5)
`
`7
`44.4 (26.2)
`56.9 (23.0)
`
`50 mg
`i. m.
`
`8
`38.5 ($3.1)
`70.5 ($4.0)
`
`Value of F from
`One-Way ANOVAa
`
`—
`0.62
`2.50
`
`
`
`Characteristics
`
`Number of patients
`Age (years)
`Weight (kg)
`Serum concentrations:
`
`1.07
`4.2 ($0.3)
`4.0 ($0.2)
`3.6 ($0.3)
`3.6 ($0.2)
`Albumin (g/100 ml)
`0.14
`1.13 ($0.12)
`0.96 ($0.16)
`1.06 ($0.28)
`1.08 ($0.22)
`Total bilirubin (mg/100 ml)
`0.24
`0.88 ($0.04)
`0.94 ($0.02)
`0.90 ($0.11)
`0.88 ($0.04)
`Creatinine (mg/100 ml)
`0.52
`33.1
`($2.9)
`34.6 ($5.2)
`33.6 ($1.4)
`40.0 ($6.6)
`Alkaline phosphatase (IU/ml)
`
`SGOT (IU/ml) 2.14 48.0 ($13.6) 98.6 ($41.5) 27.1 ($44) 37.6 ($6.7)
`
`
`
`
`
`a For all values, p > 0.1
`
`72-hour DMCDX curve averaged 100.7% of the
`intravenous value. Neither of these values was sig—
`nificantly different from 100%, nor from each other
`(Fig. 4). Thus absorption of both oral and i. In. CDX
`was essentially complete on the average.
`
`Study Two
`
`One-way analysis of variance indicated that the four
`treatment groups were comparable with respect to
`the variables under study (Table 2). In all 4 groups,
`mean values of total bilirubin, alkaline phosphatase,
`and SGOT were near or above the usual upper limits
`of normal, consistent with mild alcoholic hepatitis
`(Table 2).
`During the 5—hour interval following the 100—mg
`
`dose of CDX - HCl, mean plasma concentrations of
`CDX after oral dosage were significantly higher at all
`corresponding points in time than the levels mea—
`sured after i. In. injection of the same dose (p < 0.05)
`(Fig. 5). Mean levels of DMCDX also were signifi-
`cantly higher at all corresponding points in time fol-
`lowing oral administration than after i. In. injection.
`The results were similar following the 50—mg dose
`(Fig. 6). Plasma concentrations of CDX were sig-
`nificantly higher at all corresponding times after oral
`dosage than after
`i. In.
`injection. Differences in
`plasma levels of DMCDX at corresponding points in
`time also were higher following oral administration,
`but the differences were of borderline significance
`due to the large individual variation. Thus i. In. injec-
`tion of CDX - HCl during acute alcohol withdrawal
`
`AstraZeneca Exhibit 2117 p. 5
`
`1750*
`
`150.0)-
`
`125.0—
`
`1
`
`
`
`
`
`
`
`Areaunder72-hourcurve(percentofintravenous) 6 O o
`
`
`
`
`
`75-0 *
`
`50.0%
`
`25.0L
`
`OL
`
`x
`
`X Y
`
`Oral
`
`DDD
`
`A A A
`Intramuscular
`
`—._4#~_.—l—_.$
`
`
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`leads to slower absorption of the parent compound
`with correspondingly slower generation of the first
`pharmacologically active metabolite than does oral
`administration of the same dose.
`
`Discussion
`
`Previous reports indicate that oral and intramuscular
`administration of chlordiazepoxide are by no means
`pharmacokinetically comparable [1, 4, 12]. Whereas
`absorption of oral CDX - HCl appears to be rela—
`tively rapid, absorption after i. m. injection is slower.
`Earlier studies were based upon studies of smaller
`numbers of subjects, all of whom were healthy volun-
`teers. The present study expanded the findings to a
`large number of healthy volunteers, and included
`comparison with intravenous administration to pro—
`vide an estimate of absolute systemic availability.
`Furthermore, we assessed the potential clinical sig-
`nificance of the pharmacokinetic difference in a study
`of chlordiazepoxide absorption in patients with acute
`alcohol withdrawal. Intramuscular chlordiazepoxide
`is traditionally prescribed for patients with this dis-
`order.
`
`In the 14 healthy volunteers subjects, peak blood
`CDX concentrations after i. m. CDX - HCl were sig-
`nificantly lower and reached later after the dose than
`with oral administration of the same dose. Absorp-
`tion of oral CDX - HCl, when taken in the fasting
`state, proceeded as an apparent first-order process in
`all subjects, with a mean apparent absorption half-
`life of about 43 min. In only two of the fourteen
`subjects did the absorption half—life exceed 60 min.
`Following i. m. injection, however, a biphasic rate of
`absorption was evident in the majority of subjects.
`This would be consistent with rapid drug precipita-
`tion at the injection site followed by slow redissolu—
`tion, and has been previously suggested as a possibil-
`ity with chlordiazepoxide [1], as well as with pheny-
`toin [14] and quinidine [7]. This property of CDX
`could be predicted on the basis of the physicochemi-
`cal characteristics of the drug [1, 15, 16]. As a weak
`organic base with a pK of about 4.8, chlordiazepox—
`ide for parenteral injection is supplied in a solvent
`vehicle having a pH of approximately 3.0, in which
`the drug exists predominantly in the water—soluble
`hydrochloride salt form. Following i. m.
`injection,
`buffering of the solution to physiologic pH causes the
`water-insoluble free base form to predominate, thus
`favoring precipitation at the injection site. The acidic
`pH of the injection solvent, together with its high
`concentrations of propylene glycol, probably explain
`the
`local pain and muscle damage
`following
`intramuscular injection [11].
`
`AstraZeneca Exhibit 2117 p. 6
`
`272
`
`3.5 r
`
`3.0—
`
`2
`
`X\
`’E‘ 2.0 — j
`g
`.'
`v 1.5 —.‘
`c
`l
`o
`1
`
`
`
`.
`I
`ChlordIazepOXIde
`x,
`
`x,
`\(Oral
`X‘~x
`
`\\XI
`
`,x\
`
`,’
`
`\\
`X‘\
`
`’X‘
`
`x’
`
`‘x_-x
`
`’6‘):
`,’
`II
`[I
`l
`‘x'
`
`A___A.,A\ /A‘—A\.A,A/A\A
`‘A
`
`A. A.\A_’A/
`Intramuscular
`
`:
`I:
`l
`g 1 0—:
`8
`I {8/
`° 0 54
`m
`.
`u
`.
`I, /
`
`V
`E
`OK
`I
`I
`l
`I
`‘8
`E
`O
`1.0
`2.0
`3.0
`5.0
`
`4.0
`
`1 -0
`
`Desmethylchlordlazepoxlde
`x--x*‘X—-X“X’—x
`
`0. 5
`
`oralfi—Xr‘x'w'x
`,x—-x"
`Intramuscular
`
`x‘“
`A_.A_.A._.A—4—‘A—‘A_‘A_'A—.A
`oaéA-—A"IA‘A—'A"'I
`4L
`.1
`_I
`1 .O
`2.0
`3.0
`4.0
`5.0
`O
`Hours after lOO-mg close
`
`Fig. 5. Blood concentrations of chlordiazepoxide and desmethyl-
`chlordiazepoxide for patients in Study Two receiving a lOO—mg
`dose of chlordiazepoxide hydrochloride orally or by i. m. injection
`(Table 2). Each point is the mean for all subjects at the corre-
`sponding time
`
`2-5
`
`2-0
`
`A
`
`Chlordiazepoxide
`X
`I \ ,vxr-x—.X
`X
`/~X"X~\x
`Oral
`
`II,
`2‘
`
`IIX\\
`
`“x
`
`x-~x--x
`
`E 1.5
`0"
`:l.
`V
`
`r
`I
`I
`I
`
`g 1 O ,'
`f!
`[I
`'\ ./
`4"
`g 0.5 , A/
`A
`l
`O
`'
`I
`O
`
`I
`I
`_J.
`I
`_l
`on
`0
`1.0
`2.0
`3.0
`4.0
`5.0
`
`UC
`
`E
`ECL
`
`Intramuscular
`A/A. A¢A ’A‘\A-—a —A’A ‘A"‘A ~A
`
`0 5 r
`Desmethylchlordlazepoxide
`O/ral
`Intramuscular
`X,-x—-X_-x_-x-_x,-x--><'-><“X
`
`-><-—><-‘X"
`____A__A__.A._.A—-A—‘A"“A
`Ozark—TX ~ we‘d—'Q’AT‘A l
`l
`_l
`1 .O
`2.0
`3.0
`4.0
`5.0
`O
`Hours after 50- mg dose
`
`Fig. 6. Blood concentrations of chlordiazepoxide and desmethyl-
`chlordiazepoxide for patients in Study Two receiving a 50-mg dose
`of chlordiazepoxide hydrochloride orally or by i. m.
`injection
`(Table 2). Each point is the mean for all subjects at the corre-
`sponding time
`
`
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`The absolute systemic availability of i. m. CDX -
`HCl based upon 48 h of sampling averaged 86%,
`which was not significantly different from 100%. A
`longer period of sampling with more sensitive assay
`techniques probably would bring the average closer
`to 100%. Systemic availability of oral CDX - HCl
`averaged approximately 100% over the 48-hour
`sampling period.
`Extension of our study to patients with acute
`alcohol withdrawal yielded consistent
`findings.
`Plasma concentrations of CDX during the first 5 h
`after dosage were lower after i. m.
`injection than
`after oral administration of the same dose, whether
`
`50 mg or 100 mg. The rate and extent of appearance
`of DMCDX,
`the first pharmacologically active
`metabolite of CDX, after i. m.
`injection was also
`reduced in comparison with oral administration.
`Since DMCDX is formed by hepatic demethylation
`of CDX, the presence of alcoholic hepatitis might
`influence the rate of formation of DMCDX from
`
`CDX. However, indices of hepatic function (or dys-
`function) did not differ between groups, making it
`unlikely that
`this influenced differences between
`routes of administration. The slow absorption of i. m.
`CDX - HCl may explain its poor effectiveness in a
`recent clinical study in patients with alcohol with-
`drawal [17].
`Thus intramuscular injection of chlordiazepoxide,
`like that of many other drugs [6], may not be an
`optimal mode of administration, and certainly does
`not lead to prompt achievement of high blood con—
`centrations. Chlordiazepoxide should be adminis—
`tered by slow, well-controlled intravenous infusion in
`situations
`requiring parenteral
`therapy. When
`intravenous administration is not
`feasible, oral
`administration may be preferable to intramuscular
`injection.
`
`Acknowledgements. We are grateful for the assist-
`ance of Paul Anderson, Marcia Allen, Jerold S. Har—
`matz, Dr. Dean S. MacLaughlin, and the nursing staff
`of the Clinical Research Unit and Emergency Room
`of the Clinical Institute, Addiction Research Founda-
`tion, Toronto. Dr. MacLaughlin is supported by
`USPHS Grant GM—23430 to the Boston Collabora-
`
`tive Drug Surveillance Program.
`The authors wish to thank Dr. Jan Koch—Weser
`for his advice and collaboration.
`
`References
`
`1. Boxenbaum, H. G., Geitner, K. A., Jack, M. L., Dixon, W. R.,
`Spiegel, H. E., Symington, J., Christian, R., Moore, J.D.,
`
`273
`
`Weissman, L., Kaplan, S. A.: Pharmacokinetic and biophar-
`maceutic profile of chlordiazepoxide HCl in healthy subjects:
`single—dose studies by the intravenous, intramuscular, and oral
`routes. J. Pharmacokinet. Biopharm. 5, 3—23 (1977)
`. Boxenbaum, H. G., Riegelman, S., Elashoff, R. M.: Statistical
`estimations in pharmacokinetics. J. Pharmacokinet. Biopharm.
`2, 123—148 (1974)
`. Gibaldi, M., Perrier, D.: Pharmacokinetics. New York: Marcel
`Dekker, Inc. 1975
`.Gottschalk, L. A., Biener, R., Dinovo, E. C.: Effects of oral
`and intramuscular routes of administration on serum chlor—
`diazepoxide levels and the prediction of these levels from pre—
`drug fasting serum glucose concentrations. Res. Commun.
`Chem. Pathol. Pharmacol. 8, 697—702 (1974)
`. Greenblatt, D.J., Koch-Weser, J .: Clinical pharrnacokinetics.
`N. Engl. J. Med. 293, 702—705, 964—970 (1975)
`. Greenblatt, D.J., Koch—Weser, J .: Intramuscular injection of
`drugs. N. Engl. J. Med. 295, 542—546 (1976)
`. Greenblatt, D.J., Pfeifer, H. J., Ochs, H. R., Franke, K., Mac-
`Laughlin, D.S., Smith, T. W., Koch-Weser, J.: Pharmacoki-
`netics of quinidine in humans after intravenous, intramuscular
`and oral administration.
`J. Pharmacol. Exp. Ther. 202,
`365—378 (1977)
`.Greenblatt, D. J., Shader, R. I.: Benzodiazepines in clinical
`practice. New York: Raven Press 1974
`.Greenblatt, D. J., Shader, R. I., Franke, K., MacLaughlin,
`D.S., Ransil, B.J., Koch-Weser, J.: Kinetics of intravenous
`chlordiazepoxide: sex differences in drug distribution. Clin.
`Pharmacol. Ther. 22, 893—903 (1977)
`Greenblatt, D.J., Shader, R.I., Harmatz, J .S., Franke, K.,
`Koch—Weser, J.: Influence of magnesium and aluminium hy-
`droxide mixture on chlordiazepoxide absorption. Clin. Phar-
`macol. Ther. 19, 234—239 (1976)
`Greenblatt, D.J., Shader, R. 1., Koch-Weser, J .: Serum
`creatine phosphokinase concentrations after intramuscular
`chlordiazepoxide and its solvent. J. Clin. Pharmacol. 16,
`118—121 (1976)
`Greenblatt, D. J., Shader, R. I., Koch-Weser, J., Franke, K.:
`Slow absorption of intramuscular chlordiazepoxide. N. Engl. J.
`Med. 291, 1116—1118 (1974)
`Greenblatt, D.J., Shader, R. I., Koch-Weser, J., Franke, K.:
`Clinical pharmacokinetics of chlordiazepoxide.
`In: Phar-
`macokinetics of psychoactive drugs: Blood levels and clinical
`response, pp. 127—139.
`(ed. L. A. Gottschalk, S. Merlis)
`Flushing, NY: Spectrum Publications 1976
`Kostenbauder, H.B., Rapp, R. P., McGovren, J. P., Foster,
`T. S., Perrier, D.G., Blacker, H.M., Hulon, W. C., Kinkel,
`A. W.: Bioavailability and single—dose pharmacokinetics of
`intramuscular phenytoin. Clin. Pharmacol. Ther. 18, 449456
`(1975)
`MacDonald, A., Michaclis, A. F., Senkowski, B.Z.: Chlor—
`diazepoxide, in K. Florey, editor: Analytical profiles of drug
`substances, vol. 1, pp. 15—38. New York: Academic Press
`1972
`
`10.
`
`11.
`
`12.
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`MacDonald, A., Michaelis, A. F., Senkowski, B.Z.: Chlor-
`diazepoxide hydrochloride,
`in K. Florey, editor: Analytical
`profiles of drug substances, vol. 1, pp. 39—52. New York:
`Academic Press 1972
`Palestine, M. L., Alatorre, E.: Control of acute alcoholic With-
`drawal symptoms: a comparative study of haloperidol and
`chlordiazepoxide. Curr. Ther. Res. 20, 289—299 (1976)
`Schwartz, M. A., Postma, E.: Metabolic N-demethylation of
`chlordiazepoxide. J. Pharm. Sci. 55, 1358—1362 (1966)
`Shader, R.I., Greenblatt, D.J., Harmatz, J.S., Franke, K.,
`Koch-Weser, J.: Absorption and disposition of chlordiazepox-
`
`AstraZeneca Exhibit 2117 p. 7
`
`
`
`274
`
`D. J. Greenblatt et al.: Absorption of Chlordiazepoxide
`
`ide in young and elderly male volunteers. J. Clin. Pharmacol.
`17, 709—718 (1977)
`20. Smith, T. C., Moyer, C. E.: Bioavalability of two preparations
`of chlordiazepoxide. Curr. Ther. Res. 20, 204—210 (1976)
`21. Snedecor, G. W., Cochran, W. G.: Statistical methods. Ames,
`Iowa: University of Iowa Press 1967
`22. Usanis, R. A.: NLIN-—Nonlinear least squares estimation of
`parameters (Library Services Series Document No. LSR—089—
`1). Research Triangle Park, NC, 1972, Triangle Universities
`Computation Center
`
`23. Wagner, J. G.: Fundamentals of clinical pharmacokinetics.
`Hamilton, Illinois: Drug Intelligence Publications 1975
`
`Received: October 24, I 977, in revised form: January 5, I978
`accepted: February 17, 1978
`
`Dr. David J. Greenblatt
`Clinical Pharmacology Unit
`Massachusetts General Hospital
`Boston, MA 02114, USA
`
`AstraZeneca Exhibit 2117 p. 8
`
`