+ All Categories
Home > Documents > Sub Title Author Ito, Hiroshi(Kose, Noriko) 巨勢, 典子 ...

Sub Title Author Ito, Hiroshi(Kose, Noriko) 巨勢, 典子 ...

Date post: 05-Dec-2021
Category:
Upload: others
View: 1 times
Download: 0 times
Share this document with a friend
11
Title Comparative analysis of partition coefficients in linear and nonlinear eliminating organs without measuring tissue concentrations Sub Title Author Ito, Hiroshi(Kose, Noriko) 巨勢, 典子(Nakashima, Emi) Deguchi, Yoshiharu Benet, Leslie Z. 中島, 恵美 Publisher 共立薬科大学 Publication year 2007 Jtitle 共立薬科大学雑誌 (The journal of Kyoritsu University of Pharmacy). Vol.3, (2007. 10) ,p.9- 18 Abstract A new method is derived for the determination of partition coefficients (KP) for well-stirred physiologic models using residence time concepts. The difference in area under the moment curve (AUMC)/area under the curve (AUC) between inflow and outflow plasma or blood concentrations for specific organ is shown to be a function of a tissue distribution volume, blood flow rate, and extraction ratio. The method allows the investigator to calculate KP of a tissue and/or organ independent of the mode and route of administration without actually requiring tissue concentration measurements. Although the KP determination method described here works only for organs in which distribution and elimination follow linear kinetics, nonlinearities observed in other organs will not compromise the accuracy of the determination in an organ following linear processes. It is also shown that addition of a second compound, such as antipyrine, relieves the requirement of determination of organ blood flow. The method was applied to determine the KP of a drug, thiamylal, in the rabbit leg muscle and brain. The calculated KP values were in good agreement with the values of KP at steady-state determined as the ratio of measured tissue to plasma concentrations. Notes 原著論文 Genre Departmental Bulletin Paper URL https://koara.lib.keio.ac.jp/xoonips/modules/xoonips/detail.php?koara_id=jkup2007_3_009 慶應義塾大学学術情報リポジトリ(KOARA)に掲載されているコンテンツの著作権は、それぞれの著作者、学会または出版社/発行者に帰属し、その権利は著作権法によって 保護されています。引用にあたっては、著作権法を遵守してご利用ください。 The copyrights of content available on the KeiO Associated Repository of Academic resources (KOARA) belong to the respective authors, academic societies, or publishers/issuers, and these rights are protected by the Japanese Copyright Act. When quoting the content, please follow the Japanese copyright act. Powered by TCPDF (www.tcpdf.org)
Transcript

Title Comparative analysis of partition coefficients in linear and nonlinear eliminating organs withoutmeasuring tissue concentrations

Sub TitleAuthor Ito, Hiroshi(Kose, Noriko)

巨勢, 典子(Nakashima, Emi)Deguchi, YoshiharuBenet, Leslie Z.中島, 恵美

Publisher 共立薬科大学Publication year 2007

Jtitle 共立薬科大学雑誌 (The journal of Kyoritsu University of Pharmacy). Vol.3, (2007. 10) ,p.9- 18 Abstract A new method is derived for the determination of partition coefficients (KP) for well-stirred

physiologic models using residence time concepts. The difference in area under the momentcurve (AUMC)/area under the curve (AUC) between inflow and outflow plasma or bloodconcentrations for specific organ is shown to be a function of a tissue distribution volume, bloodflow rate, and extraction ratio. The method allows the investigator to calculate KP of a tissueand/or organ independent of the mode and route of administration without actually requiringtissue concentration measurements. Although the KP determination method described hereworks only for organs in which distribution and elimination follow linear kinetics, nonlinearitiesobserved in other organs will not compromise the accuracy of the determination in an organfollowing linear processes. It is also shown that addition of a second compound, such asantipyrine, relieves the requirement of determination of organ blood flow. The method wasapplied to determine the KP of a drug, thiamylal, in the rabbit leg muscle and brain. Thecalculated KP values were in good agreement with the values of KP at steady-state determined asthe ratio of measured tissue to plasma concentrations.

Notes 原著論文Genre Departmental Bulletin PaperURL https://koara.lib.keio.ac.jp/xoonips/modules/xoonips/detail.php?koara_id=jkup2007_3_009

慶應義塾大学学術情報リポジトリ(KOARA)に掲載されているコンテンツの著作権は、それぞれの著作者、学会または出版社/発行者に帰属し、その権利は著作権法によって保護されています。引用にあたっては、著作権法を遵守してご利用ください。

The copyrights of content available on the KeiO Associated Repository of Academic resources (KOARA) belong to the respective authors, academic societies, orpublishers/issuers, and these rights are protected by the Japanese Copyright Act. When quoting the content, please follow the Japanese copyright act.

Powered by TCPDF (www.tcpdf.org)

Original  PaPer

Comparative  Analysis  of Partition  Coefficients  in Linear  and  Nonlinear

     Eliminating  Organs  without  Measuring  Tissue  Concentrations

HirOShi  ItOl),  NOrikO  KOSel),  YOShiharU.  DegUChi2),  LeSlie  Z. Benet3),  Emi  NakaShimal)*

                   1)Department  of Pharmaceutics,  Kyoritsu  University  of Pharmacy,

                       1-5-30Shiba-koen,  Minato-ku,  Tokyo  l O5-8512,  Japan.

  2)Department  ofDrug  Disposition&Pharmacokinetics,  School  of Pharmaceutical  Sciences,  Teikyo  University

                  Suarashi  1091-1,Sagamiko,  Sagamihara,  Kanagawa  229-Ol95,  Japan3)Department  of Biopharmaceutical  Sciences

, University  of Califbmia  San  Francisco  533  Pamassus  Avenue,  Room

                           U-68,San  Francisco,  CA  94143-0446,  USA

(Received  July 20,2007;Revised  August  20,2007;Accepted  August  21,2007)

  Anew  method  is derived  fbr  the  determination  of partition  coefficients(Kp)fbr  well-stirred

physiologic  models  using  residence  time  concepts.  The  difference  in area under  the moment  curve

(AしIMC)/area  under  the curve(/望 σq)between  inflow  and  outflow  plasma  or blood  concentrations

fbr specific  organ  is shown  to be  a fUnction  of a tissue  distribution  volume,  blood  flow  rate, and

extraction  ratio.  The  method  allows  the  investigator  to calculateκp  of a tissue  and/or  organ

independent  ofthe  mode  and  route  of administration  without  actually  requiring  tissue  concentration

measurements.  Although  the飾determination  method  described  here  works  only  fbr organs  in

which  distribution  and  elimination  fbllow  linear  kinetics,  nonlinearities  observed  in other  organs

will not compromise  the accuracy  ofthe  determination  in an organ  following  linear  processes.  It is

also  shown  that  addition  of a second  compound,  such  as antipyrine,  relieves  the requirement  of

determination  of  organ  blood且ow.  The  method  was  applied  to determine  the  Kp  of  a drug,

thiamylal,  in the  rabbit  leg muscle  and  brain.  The  calculated飾values  were  in good  agreement

with  the  values  of  Kp  at steady-state  determined  as  the  ratio  of  measured  tissue  to  plasma

COnCentrat10nS.

lNTRODUCTION

Physiologically  based  pharmacokinetic  models

have  gained  increasing  attention(1).  This  is due  in

no  small  degree  to the  capacity  of such  models  to

combine  infbrmation  concerning  the  distribution,

metabolism  and  elimination  of dnlgs  and  metabolites

as a fUnction  of time,  together  with  blood  flows  to

*Correspondence  to Emi  Nakashima, Department  of

Pharmaceutics,  Kyoritsu  University  ofPharmacy,1-5-30

Shiba-koen,  Minato-ku,  Tokyo  l O5-8512,  Japan.

Phone:                                ,

FAX:,

e-mail:

particular  organs  and  tissues.  However,  physiologic

刊ow  models  are easily  simplified  into compartmental

systems  as  demonstrated  by  Bischoff  et al.(2)in

their preliminary  analysis  of methotrexarate.  In this

paper,  we  combine  both  physiologic  and

compartmental  treatments  to  point  out  that  the

steady-state  partition  coef且cients  between  organs

and  blood  or plasma(Kp),  which  are required  as part

of  the  physiologic  model,  can  be  estimated  from

analysis  of concentration-time  data  without  actually

requiring  a  separate  measurement  of  distribution.

Calculations  of  partition  coefficients  in  organs

exhibiting  linear  and  nonlinear  elimination  as well  as

fbr  non-eliminating  organs  are  carried  out  using

J Kyoritsu  Univ  Pharm  2007.  O  Vol.3

ItoHetal.

simulations.  Finally,  experimental  values  fbr  the

partition  of antipyrine  and  thyamylal  into  rabbit  leg

muscle  and  brain  are examined.

THEORY

With  physiological  pharmacokinetics,  a  mass

balance  equation  in each  organ  I can  be  written  by

using  arterial  (Ca)  and  venous  (Ci)  plasma

concentration  as follows:

K…V…Eli/il・

   rRβP・9i・C、 一(RBP・9i+f・ αi。t?・Ci (1)

  where  it is assumed  that  the  anatomical  tissue

volume(玲,  the  organ  blood  flow(gi)the  organ

tissue  to venous  plasma  concentration  ratio(Kpi)and

the organ  intrinsic  plasma  clearance(CLinti)are  organ

specific  parameters  while  the  blood  to  plasma

concentration  ratio(RB1))and  plasma  free  fraction

ωare  organ  independent.  Note  that  up  to  the

present  time,  physiologic  pharmacokinetic  models

almost  exclusively  utilize  the  venous  equilibration

(well-stirred)model  to  express  the  relationship

between  tissue  and  venous  blood  concentrations,

which  is the  approach  we  utilize  here  to  estimate

partition  coefficients.  The  approximations  required

in utilizing  physiologic  pharmacokinetics  probably

negate  any  advantages  related  to the  use  of  other

models  relating  tissue  and  venous  concentrations,  i.e.,

parallel  tUbe,  dispersion  or distributive  models.

Taking  the Laplace  transform  of Eq.  l yields:

s・Kpi・Vi・ αs,i+XiO

-RBP・9i・ α、,。一(RBP・9i+プ ・CZi。ti)・ α、,i (2)

  where  aS,、 and  aS,i  are the  Laplace  transfbrms  of

the  equations  describing  the  concentration  time

course  of drug  respectively,  in arterial  plasma  and  in

venous  plasma  from  compartment  i.  Xio  is the

amount  of  drug  in tissue  i at time  O  and  can  be

assumed  to be zero  fbr an initial dose.

  Rearrangement  of Eq.2,  assuming  X io=0,  yields:

             RBP・9iαs,i=                              'αs,a     s ・KPi・V

,+RBP・9i+!・C乙i。 、i(3)

Solving  fbr area under  the curve(Aひ(=)by  letting

0 JKyoritsu  Univ  Pharm  2007.  O  Vbl.3

s→0

AUCi-lim(、 →o)(a、 ,i)

    -lim(・一・)∫》 ・C・dt

Then  from  Eq.3

 ㏄GF昭

㎞αヴ君溜ニqσ (4)

The  first term  on the right  hand  side  of Eq.4is,  in

fact, one  minus  the plasma  extraction  ratio  from  the

organ(Ei)

AUCi=(1-Ei)・A  UC、 (5)

  As  we  have  previously  described(3),  area  under

the  moment  curve(A  UMC)may  be  determined  as

the derivative  of as ,i  with  respect  to s,(-as,i)', at the

limit when  s apProaches  zero:

AUMCi=lim(s→o)(-as,i)'

      -lim(・一・)∫♂ ・㍉ ・C・dt

Therefore  from  Eq.3:

オ㎝鵡

=1'「n(・一・)[(

、.K

RBP・9,・KPi・V,

十  S・KPi・V,+RBP・9,+f・CLi。 、i

Pi・V,+RBP・e,+!・ α

   RBP・9i

、。tl)・'as・a

Solving  Eq.6as  s→0

AUMCi=RBP・9,・KPi・V,

・AUCa

・(一as,a)]

(RBP・2、+f・CL、 。、、)2

      RBP・9,十                ・AUMCaRBP・9i+!・ αi

。,i

(6)

(7)

Factoring  out  a common  term  from  the right  hand

side  ofEq.7equivalent  to、4ひCi,  as defined  by Eq.4,

yields:

Original  paper

AUMCiニRBP・9,・Auc、

・(

RBP・2,+f・CL,。,i

      KPi・v,

=AUCi・(

RBP・9i+ノD・ αi。,i

        KPi・v,

  AUMC十     a

Auc、

  AUMC十      a

)

)(8)

measurement  of dnlg  and  antipyrine  in arterial  and

venous  plasma  across  an organ  will  yield  the  organ

tissue  to plasma  concentration  ratio  of  the  drug  as

per  Eq.12,  independent  of  blood  flow  and  tissue

VOlUme  meaSUrementS.

RBP・9,+f・CLi。,i Auc、

  Dividing  Eq.8by/A  UCi  and  rearranging  yields  the

mean  transit  time  of drug  in organ  i, that is:

       AUMCi      [(

        .4uclKPi=      AUMC

i    [(

      .4UCi

           AUCaAUMCa                ]drug      )・RBP・

           .4UCiAUC、

           ・4σCa・4しηMCa                ]antipyrine      )・RBP・

           .4ひCiAUCa

(12)

AUMC, AUMCa

.4ひCi AUCa

since丘om  Eq.4:

AUCi

一 κ…v,.Auq

  RBP・2,  A UCa

1

RBP・9i・AUC、   RBP・9i+プ ・αi。 、i

(9)

(4a)

  The    equations    described    above    require

measurements  over  all time(0→ α○)to obtain  accurate

estimates  of  Kpi.  It  is  also  possible  to  use

incremental  A UC  measurements  to  obtain  such

estimates  as  described  in  the  appendix.  However,

simulations  indicated  large  errors  in  such

determinations  as will be described.

  Solving  Eq.9fbr  Kpi  yields  the  general  equation

fbr  predicting  the  organ  tissue  to  plasma

concentration  ratio  without  actually  measuring  organ

COnCentratlOnS.

      AUMC,KPi=(      オ㏄

AUMCa

AUCa )・iiE'Sf'9弓器(1・)

  For  a non-eliminating  organ,  where  there  is no

difference  between/A  UCa(o→ 。。) and  A UCi(o→ 。。), then

Eq.10can  be written  as:

Kpi=(       Auc

,

オしiMq  AUMG      )・AU(1

RBP・9,

塔(ll)

  Thus,  if it is possible  to measure  concentrations

exiting  a particular  organ,  it is theoretically  possible

to determine  the tissue  to plasma  concentration  ratio

in  that  organ  assuming  the  venous  equilibration

model.    Equations  lO  and  ll  require  the

investigator  to determine  the  organ  blood  flow  rate

(2i)and  the  tissue  anatomical  volume(Vli).  The

former  quantity  may  be  especially  difficult  to

determine  accurately  in vivo.  Therefbre,  we  have

chosen  to  carry  out  studies  using  antipyrine  as  a

reference  compound  since  it is known  that  tissue  to

plasma  and  blood  concentration  ratios  fbr  this

compound  are  l  (i.e.,  Kpi,antipyirne  =  1;

RB1)antipyrine=1).  Therefbre,  when  a  drug  is

administered  simultaneously  with  antipyrine,

MATERIALS  AND  MEHODS

Simulations  A    representative    physiologic

pharmacokinetic  model  containing  a  plasma

compartment,  a non-eliminating  organ  and  eliminating

organs  exhibiting  linear  and  nonlinear  disposition  is

depicted  in Fig  l.  In this model  the exit concentration

from  plasma  compartment  is  equivalent  to  the

arterial  concentration  (Ca)  used  in the  theoretical

derivation  and  i varies  from  2 to 4 for the nonlinear                                           ,

linear  and  non-eliminating  organs,  respectively.

Simulated  concentrations  fbr the model  described  in

Fig  l  were  estimated  by  the  Runge-Kutta-Gill

methods  at 28 time  points  over  lOO  min  fbr doses  of

250,2500,12500mg  as depicted  in Fig 2.

ChemicaIs    Thiamylal    sodium     (Isozole;

Yoshitomi  Pharmaceutical  Co.  Ltd.,  Osaka,  Japan)

and  amobarbital  sodium(Isomytal;Nippon  Shinyaku

Co.  Ltd.,  Kyoto,  Japan)were  used  as  generously

supplied.    Antipyrine  (Wako  Pure  Chemical

Industries  Ltd.,  Osaka,  Japan)  and  all  other

chemicals  were  of reagent  grade  and  used  without

fUrther  purification.

Animals      Male  albino  rabbits(2.5-3.5  kg

from  Sankyo  Laboratory  Company,  Toyama  Japan)

were  used  fbr  intravenous  infusion  stUdies  without

fasting.

  a)Rabbit  hind  leg:The  left femoral  artery  was

cannulated  with  polyethylene  tubing(type  No.15;

J Kyoritsu  Univ  Pharm  2007.  O  Vol.3

ItoHetal.

Plasma  Compartment

     Vl  25・mL

      RBP   1

50 mL/min

N・nlin・ar-elimin・ting・rg・n  10 mUmin

             15 mL

(兀int 2001(10+C2)  mL/min

Lmear-eliminating  organ

       V315・mL

        Kp3  3

CLint 2.O mLlmin

Non-eliminating  organ

     V4200・mL

      Kp4  0・2

Fig  l

A schematic  compartment  model  for drug  and  the  various  elimination  processes. Symbol  are  in the  text.

15mI」1min

25ml/min

                            :i。 。P1-C・mp・ ・tm・nt  1,。  Nonlinea「'elimination o「gan                             目                     (

                            警1・・  .._、5..㎎ 甚1・                             ,9  糟                    )

                             這         as。。¶9       屋                             り                                                   一                             目                                  儒

                             ゆ                                                                       り                                         く                                                            、1

                            5・   es。.。        1                             窪 ・el            δ ・。1                             器                             =

・o。1                       ・ool

                                  O   20   40   60   ヨ0  100       0   20   40   60   E〔】  100

                                        Time(min)                Time(min)

                                   Non-elimination  organ         Linear-elimination  organ

                                                                                                                 臼                           日

                            量  10。                 量  1㎝                             ⇔の                                   ◎9                             =し                               =L

                             )    10                           )

                             目                               目   10                             0                                    0

                             1昌                            眉                             6                                    爵

                             ニロ                                         レ

                            葛  ..                 基                             り                                            リ                             ロ                                  コ

                            δ 。1            δ

                              .OO1                             .01                                 0  20  40  巨o  巳0  100      0  20  40  60  so  loo

                                        Time(min)                 Time(min)

Fig 2

Time  courses ofconcentration  ofthe drug in the fbur compartments.  Three different doses were used in the simulation:250,2500,  and l 2500 mg.

2 JKyoritsu  Univ  Pharm  2007.  O  Vbl.3

Igarashi  Ika  Kogyo  Co.  Ltd.,  Tokyo,  Japan)and  the

right  femoral  vein  was  punctured  with  a 23  G needle

(Terumo  Corporation,  Tokyo,  Japan)  via  the

abdominal  wall.  Thiamylal  sodium(10  mg/kg)and

antipyrine(50  mg/kg)were  simultaneously  infUsed

through  a right  ear vein  fbr lO min.  Blood  samples

(lmL)were  simultaneously  withdrawn  from  the  left

femoral  artery  and  the right  femoral  vein  through  the

cannula  and  needle,  respectively,  at 2,4,7,10,15,20,

30,45,60,90,120min  after  the  start of  infUsion.

Plasma  samples  were  separated  and  frozen  at-80℃

until analyzed.

  b)Rabbit  brain:Catheters  were  placed  in the  left

femoral  artery  and  the left internal  j ugular  vein  fbr

blood  sampling.  PE  l O polyethylene  tubing(Clay

Adams,  Becton  Dickinson  Company,  Parsippany,  NJ)

was  used  fbr  the  catheterization  of  the  intemal

jugular  vein  and  apProximately  O.5  mL  blood

samples  were  collected  at the appropriate  times.  All

other  experimental  conditions  were  identical  to that

described  fbr the hind  leg measurements.

Determination  of  tissue-to-plasma  partition

coefficients  at steady  state(Kpss)

  Arep  The  conventional  method  was  also  used  fbr

the  determination  of Kps、 in rabbit  hind  leg  muscle

and  brain  tissue  in each  of the animals  studied.  The

thiamylal  solution  was  infUsed  at a rate of 9 mg/hr  to

achieve  a steady  state  plasma  concentration  of  lO

μg/mL.  After  4 hr, the  rabbits  were  sacrificed  and

the  tissue  from  the  hind  leg  muscle  and  the  brain

were  quickly  excised,  rinsed  with  saline,  and  blotted.

AnalyticaI  procedures    Hゆ 緬 ㎜ce

liquid  chromatography  (LC-6A  system;Shimadzu

Co叩oration,  Kyoto,  Japan)was  used  to determine  the

concentrations  of  thiamylal  and  antipyrine  in

biological  samples.  For  thiamylal  in  plasma,  a

portion(0.2  mL)of  the plasma  sample  was  added  to l

mL  of  pH  5.0,  I  M  phosphate  buffer  solution

containing  60μg/rnL  of  amobarbirtal  as  an  internal

standard  and  vigorously  mixed  with  2 mL  of ethyl

acetate.  The  mixture  was  spun  in a cenUifUge  fbr 5

min  at  3500叩m.  A  l.5  mL  portion  of  the

supemLatant  organic  layer  was  separated  and

evaporated  to  dryness  under  nitrogen.  The  residue

was  reconstitUted  with  mobile  phase,  and  a 20  FL

aliquot  was  i功ected  onto  the  HPLC  colu㎜

Original  paper

(Shim-pack  CLC-ODS,6mm×150  mm;Shimadzu).

The  HPLC  conditions  used  were:flow  rate,1.O

mL/min;wavelength  UV  230  nm:mobile  phase:60%

methanol  in  O.OI  M  NaH2PO4.  Peak  area

measurements(model  CR-3A  recorder;Shimadzu)

yielded  good  linearity  over  the  concentration  range

l-50μg/rnL(r=0.999).

  For  measurement  of thiamylal  in muscle  tissue,  a

modification  of the method  of Stout  and  De  Vane(4)

was  utilized.  A  l g portion  of the tissue  sample  was

homogenized  with  O.34  M  perchloric  acid(4  mL).

AO.5  mL  portion  of the  homogenized  sample  was

mixed  with  l mL  of pH  l l.4 phosphate  buffer,0.15

mL  of O.34  M  perchloric  acid,0.5  mL  of the intemal

standard  solution,  and  4 mL  of ethyl  acetate.  The

organic  layer(3  mL)was  evaporated  to dryness  and

then  reconstituted  with  mobile  phase(acetonitrile:

0.4mM  KH2PO4=2:3)and  inj ected  onto  an  ODS

colu㎜.  The  HPLC  conditions  were:flow  rate,  l

mL/min;UV  wavelength  254  nm.  Good  linearity

was  obtained  over  the  concentration  range  l.56-25

μg/mL(r=0.997).

Antipyrine  plasma  concentrations  were  determined

by  a modification  of the  method  of Teunissen  et al.

(5).AO.l  mL  portion  of  the  plasma  sample  was

vigorously  mixed  fbr  l5-30  second  with  lOμL  of

the  intemal  standard  solution,0.4  mg/rnL  phenacetin

in ethanol,  and  2 mL  of dichloromethane:n-pentane

(1:1).After  centrifUgation  at 4000  rpm  for 5 min,

the  organic  layer(1.5  mL)was  evaporated  under

nitrogen  fbr  lO  min  at  room  temperatUre.  The

residue  was  reconstituted  with  mobile  phase

(acetonitrile:0.05  M  phosphate  buffer=1:3)and  a

20μLaliquot  was  inj ected  onto  an  ODS  HPLC

colu㎜.  The  HPLC  conditions  were:刊ow  rate  l.O

mL/min;wavelength  UV  245  nm.  Good  linearity  was

obtained  over  the  concentration  range  3.125  - 100

μ9/mL(r=0.999).

RESULTS

  Calculated  Kpi  values  from  Eqs.10  and  l l fbr the

simulations  using  the  model  depicted  in Fig  l are

given  in Table  l fbr the  3 different  doses  utilized.

Excellent

estimates  of the  Kp  values  fbr  the  non-eliminating

(compartment  3)were  obtained.  The  estimate  ofKp

f()rthe  nonlinear  eliminating  organ(compartment  2)

was  within  2.2%of  the  theoretical  value  fbr the  low

J Kyoritsu  Univ  Pharm  2007.  O  Vol.3 3

ItoHetal.

Table  l Estimation  ofKp  values  for compartments  2,3and  40f  the  model  depicted  in Fig  l.

κ

Dose(mg)

Compartment 250 2500 12500 Theoretical

2(Nonlinear)

3(Linear)

4(Noneliminating)

4.89

3.30

0.200

3.34

3.30

0.201

2.45

3.30

0.201

5.0

3.0

0.200

dose(250  mg)simulation.  Significant  error  in the

Kp  estimate  for this compartment  was  observed  as the

dose  increased(Table  l).  Note  in  Fig  2 that  the

nonlinearity  resulting  from   compartment  2

elimination  at  the  higher  doses,  causes  the

concentration  time  curves  fbr  each  of  the  fbur

compartments  to exhibit  nonlinearities,  yet  accurate

estimates  ofKp  for compartments  3 and  4 can  still be

obtained.  Reasonable  Kpi  estimates  also  were

obtained  using  the  incremental  area  equations

presented  in the Appendix  but  only  fbr concentration

measurements  obtained  very  early  in the  sampling

schedule  when  arterial-venous  di fferences  were

relatively  large.  For  the 250  mg  dose,  estimates  of

Kp  exhibited  less than  5%error  using  measurements

up  to 20  min  post  dosing  fbr compartments  3 and  4,

while  up  to  5  min  fbr  the  nonlinear  eliminating

compartment  2.  In contrast  fbr the  l 2500  mg  dose,

estimates  of  Kp  exhibiting  less  than  5%  error

necessitated  use  of  measurements  within  5 min  of

dosing  fbr  compartments  3  and  4,  while  only

measurements  in the  first 3 min  post  dosing  yielded

accurate  estimates  ofKp  for compartment  2.

  Typical  femoral  artery  and  venous  plasma

concentration  profiles  of  thiamylal  fbllowing  a  l O

(日目\。ω二言

。遍

ち5

。り

儒目

の三自

100

1・筋

1

Artgrv

Ψ口舶 囎

 ,f

   O          60          120

             Time(min)Fig  3

Typical  femoral  artery  (口)and  venous(■)plasma

concentration  profiles  ofthiamylal  in a rabbit(2.7  kg)following

intravenous  infusion  of thiamylal(10mg/kg)and  antipyrine(50

mg/kg)for  lOmin.

min  intravenous  infUsion  to a rabbit  are shown  in Fig

3.The  arterial  concentrations  fbr the  first 20  min

were  much  higher  than  the  arterial  levels.  These

results  indicate  that thiamyhlal  is rapidly  taken  up by

rabbit  hind  leg  muscle  tissue  and  that  the  drug

exhibits  marked  arterial-venous  plasma  concentration

differences.  Fig  4 illustrates  typical  femoral  artery

and  venous  plasma  profiles  of antipyrine  also  shows

marked  arterial-venous  plasma  concentration

differences  in rabbit  hind  leg.

  Figs  5 and  6  exhibit  typical  femoral  artery  and

intemal  jugular  vein  concentration  profiles  of

thiamylal  and  antipyrine,  respectively.  In  contrast

to  the  results  shown  in Figs  3  and  4, no  marked

arterial-venous  plasma  concentration  differences

exist,  indicating  that  pseudo-steady-state  is rapidly

achieved  in brain  tissue.

  The  experimental  parameters  fbr  thiamylal  and

antipyrine  necessary  fbr  the  determination  of  Kp

using  Eq.12,  as well  as the caluculatedκp  value  and

measuredκp  value  at steady  state  fbr  thiamylal  in

rabbit  hind  leg  are  listed  in  Table  2.  The  mean

values  fbr.∠ ひCa/.4σCi  of thiamylal  and  antipyrine

were  slightly  lower  than  unity  although  not

statistically  different  from  unit》1.  Mean  transit  times

(目目\⇔。ユ言

。嘱驚

ち唱8

9

儒目

の儒=

1000

100

10

彊日■■

Arieiy

Venovs

  1

    0          60          120

             Time(min)

Fig  4

Typical  femoral  artery  (口)and  venous(■)plasma

concentration  profiles  of antipyrine  in the  same  rabbit  studied  in

Fig.3.

JKyoritsu  Univ  Pharm  2007.  O  Vbl.3

Original  paper

(目目\b幻ユ)昌o欝

』省

8

唱oり

儒∈

器=

100

10

1

Artery

距 船 雌

  ¶1

    0          60          12a

             Time(min)

Fig  5

Typical  femoral  artery(□)and  intemal  jugular  venous(■)

plasma  concentration  profiles  of  thiamylal  in a rabbit(2.8  kg)

f()llowing  intravenous  infusion  of  thiamylal(10  mg/kg)and

antipyrine(50  mg/kg)for  l O min.

(目目\⇔⑳ユ)唱o眉捻』芒

8

9

帽目量

1000

100

10

Artar)t

Venovs

  1

   0          60          120

             Time(min)

Fig  6

Typical  femoral  artery(口)and  intemal  jugular  venous(■)

plasma  concentration  profiles  of antipyrine  in the  same  rabbit

(2.8kg)studied  in Fig.4.

in hind  leg  were  21.7±4.5  min  fbr  thiamylal  and

24.1±2.7min  fbr  antipyrine.  The  Kp  value

estimated  using  Eq.12(0.62±0.02)was  slightly

larger  than  the measured  KpsS  value(0.52±0.ll),  but

there  is no  significant  difference  between  them(p<

0.05)fbr  this limited  set of experiments.

Table  31ists  the variable  moment  parameters  and  the

estimated  and  observed  Kp  values  fbr thiamylal  and

antipyrine  in brain  tissue.  Mean  transit  time  in brain

tissue  was  6.32±2.35  min  fbr thiamylal  and  7.13±

3.41min  fbr  antipyrine.  These  values  are  much

smaller  than  those  fbund  in hind  leg.  The  mean

estimated  Kp  value  using  Eq.12(0.96±0.43)was  in

good  agreement  with  the  mean  measured  Kpss  value

(0.94±0.23),however,  the  mean  value  exhibited  a

larger  coe箭cient  ofvariation(45%vs.24%).

DISCUSSION

In  physiological  pharmacokinetic  analysis,  the

estimation  of  Kp  values  is usually  based  on  tissue

sampling  at steady-state  in various  animal  species.

Since  the sampling  of human  tissue  is limited,  animal

scale-up  to  humans  is  the  most  popular  method

utilized  to predict  Kp  values  in man.  Experimentally,

Kp  is determined  by  measuring  tissue  and  plasma

concentrations  at steady-state  (6,7)and  also  under

non-steady-state  conditions(8).  Gallo  etα1.(9)

demonstrated  that.4ひC  measurements  of  tissue

concentrations  may  provide  a reasonable  estimate  of

the  partition  coefficient.  However,  in all of these

methods(6-9)an  animal  must  be  sacrificed  fbr each

tissue-sampling  time  point.  Moreover,  many

difHculties   arise   in   correctly   assaying   drug

concentrations  in  various  tissue  samples.  In  the

Table  2 Estimationκb  values  using  Eq.12  qnd  Qbserved  KpsS  values  at . steady  state. for thiamylal  in rabbit  hind  legs,  together  withthe moment  parameters  of thiamylal  afid antipyrine  used  for the estimation  ofKp  values.

Parameters 1

Rabbit  Number

     2 3 mean  土  SD

Thiamylal

R、BP

AしwrCi/AひCi-A  UMC、IA  UC、(min)

AひC。/AUCi

0.651

16.8

1.05

0.684

25.5

0.841

0.801

22.8

0.863

0.712土0.079

 21.7土4.46

0.918±0.116

Antipyrine

AしiMCi/A  UCi一4  UMCa/A  UCa(min)

AUC。/A  UCi

gi略(min'1)

21.6

0.879

0.0527

23.6

1.Ol

O.0420

27.0

0.911

0.0407

  24.1土2.71

 0.932±0.066

0.0451±0.0066

KPss

0.603

0.619

0.617

0.400

0.641

0.536

0.620土0.019

0516±0.107

J Kyoritsu  Univ  Pharm  2007.  O  Vol.3

ItoHetal.

Table  3 Estimated飾values  using  Eq.12  and  observed  Kpss                                           value  at steady  state  fbr thiamylal  in rabbit  brain  tissue,  and  moment

parameters  ofthiamylal  and  antipy'rine'used  for the estimation  ofKp  values.

Parameters 4

Rabbit  Number

      5 6 mean  土  SD

Thiamylal

RBP

A UMCi/A  UCi-AUMC。IA  UC。(min)

AUC。/A  UCi

0.929

8.99

0.992

0.87

4.56

0.995

0.720

5.41

0.962

0.841±0.108

 6.32土2.35

0.983±0.Ol9

Antipyrine

A UMCi/A  UCi-AUMC、/A  UC、(min)

AUC。/A  UCi

gi/Vi(min-1)

8.03

0.878

0.142

3.36

0.952

0.313

10.0

0.814

0.123

 7.13土3.41

0.881±0.069

0.192土0.105

 

1.18

0.674

1.24

1.04

α

1

0.959±0.434

0.940±0.234

method  presented  here,  tissue-sampling  is  not

necessary  and  thus  the development  of an appropriate

tissue  sampling  technique  and  the  corresponding

standard  curve  is not necessa「y.

  Kety  and  Schmidt(10)determined  cerebral  blood

flow  rates  in man  using  nitrous  oxide.  However,  this

determination  required  the fbllowing  assumptions:a)

the  brain  is a homogeneous  tissue;b)Kp  of nitrous

oxide  between  tissue  and  blood  is unity,  and  c)

nitrous  oxide  is taken  by  the tissue  via  a blood  flow

limited  process.  Using  Eq.13  developed  by  Kety

and  Schmidt,  Kp  values  in particular  organ  could  be

determined  independent  of  nonlinear  disposition

processes  which  may  occur  in  other  organs  of the

body.

     RBP・9、 ・(AUC。(。 →,)-AσC、(。 →,))KPi=                V

i・Ci(13)

Where  Ci  is the  measured  concentration  of  the

drug  in the  venous  outflow  from  organ  i at time  t.

Since  small  Ci value  may  result  in large  errors  in Kp

determination,  Eq.13  is most  accurate  during  the

initial sampling  Phase.

  Statistical  moment  theory  has  been  used  fbr many

years  to determine  tissue  transit  times(ll).  Kakutani

et al.(12)applied  moment  theory  to determine  Kp  for

the in situ perfused  hind  leg of rabbits  using  outflow

measurements  fbllowing  single  pass  drug  perfUsion

of the  leg.  In  the  present  study,  we  demonstrate

that Kp  values  can  be  determined  independent  of the

mode  of administration  using  Eq.12,  as long  as no

drug  is present  in the  system  from  a previous  dose.

Lassen  and  Perl(ll)defined  the"mean  transit  time

of system"as  given  in Eq.14  when  both  inflow  and

outflow  for the system  can  be measured  over  time.

mean  transit time  of system=∬q。・4弔c諺

熊 ㎝4'熊 ・4'

(14)

  We  previously  showed  in a reversible  mammillary

model  with  a single  input  site, but independent  of the

site of  input  or the  type  of input  process,  that  the

difference  in.4σMCZ4ひC  ratios  between  the central

and  a peripheral  compartment  will  yield  a measure  of

the  sum  of  exit  rate  constants  from  that  peripheral

compartment(3).    For  example,  consider  the

model  depicted  in Scheme  l from  reference  3.

  For  such  mammillary  models,  where  there  are no

restrictions  on site of input  or exit, then:

1

-

AUMC,  A UMC,

Auc, Auc, (15)

  where  i is a peripheral  compartment  into  which

input  does  not  occur(i.e.1<i≠pin  Scheme  l)and

thus  the mean  transit  time  of system(Eqs.9and  l4)

in the physiologic  model  is equivalent  to the  sum  of

the  exit  rate  constants(Eq.15)in  a compartmental

model.

Although  the  method  proposed  here  to determine

Kp  values  has  many  advantages  over  previous  direct

(measure  Kpss  as in reference  6 and  7)or  indirect

JKyoritsu  Univ  Pharm  2007.  O  Vbl.3

Original  paper

Input  into  Central

Compartment(i)

Input  into  Peripheral

Compartment(P)

/

.-,■■■■

/Scheme  l

methods  (Eq.  13  and  references  8-10),  the

calculations  are  subj ect  to  the  errors  inherent  in

determining  .4 UMC  (13).      That  is,  the

extrapolation  ofA  UMC  to time  infinity  is much  more

sensitive  to  errors  in  calculating  the  terminal

disposition  rate  constant  as  compared  to  the  AUC

extrapolation.  However,  it is our  belief(although

not  exhaustively  tested  as  yet)  that  calculations

involving  a  difference  in  AUMC/A  UC  measures

versus  a single  measure  as used  by  Kakutani(12)

may  be  less  influenced  by  errors  in the terminal  rate

constant.

  In summary,  the  Kp  value  in a physiologic  model

employing  the  traditional  well-stirred  assumptions

can  be calculated  using  Eq.12.  A great  advantage  of

use  of Eq.12  is the Kp  values  can  be then  calculated

independent  of mode  of administration.

AUC。(。 →、)RBP・9、+!・ α 、。,、

オ ㏄ 、(。→、) RBP・9,(A2)

Appendix  A=Derivation  of Eq.13

  1ntegrating  Eq.  l between  times  O and  tl(where  Ci

=Oat  time=Oand  Cl>Oat  time=のto  obtain

incremental  areas  under  the  curve  from  O to time  tl,

AUC(o→tl)・yields:

KPi・V,・Ci-RBP・9i・A  UC、(o→t1)

       一(RBP・9i+!・ α}inti)・A UCi(o→9)(A  l)

Since  Cl  is also  zero  at t=oO,  it fbllows  that:

  Dividing  both  sides  of  Eq.  Al  by  Vi・Ci  and

substituting  the  ratio  of areas  over  all time  fbr  the

extraction  ratio related  te㎜(Eq.  A2)yields

  K,1ニ                V,・Cl・!1ひq(σ→。。)

  For  a non-eliminating  organ  where  A UCa(o→ 。。)

=AUCi(o →。。), Eq. A3  reduces  to Eq.13.

RBP●(鉛(A  uc,(o -…)' A uc・(o→・「AUCa(o-…)●Auc,(〔b・))(A3)

ACKNOWLEDGEMENT

  This  study  was  supported  in part by  a Grant-in-Aid

fbr  Scientific  Research  from  the  Ministry  of

Education,  CultUre,  Sports,  Science  and  Technology

(MEXT),  Japan,  the  Science  Research  Promotion

Fund  from  the  Promotion  and  Mutual  Aid

Corporation  fbr Private  Schools  of Japan.

REFERNCES

l.Himmelstein  KJ,  Lutz  RJ.  A  review  of  the

   apPlications     of     physiologically     based

   pharmacokinetic  modeling.  」  -Pharmacoin'n

   Biopharm  7,127-145(1979).

2.Bischoff  KB,  Dedrick  RL,  Zaharko  DS,

   Longstreth  JA.  Methotrexate  pharmacokinetics.  J

   I)harm  Sci 60,  l l 28-1133(1971).

3.Nakashima  E, Benet  LZ.  An  integrated  approach  to

   pharmacokinetic  analysis  for  linear  mammillary

   systems  in  which  input  and  exit  may  occur

J Kyoritsu  Univ  Pharm  2007.  O  Vol.3 7

ItoHetal.

   in/from  any  compartment.  J  Pharmacokinet

   Bi(4pharm  l 7,673-686(1989).

4.Stout  SA,  DeVane  CL.  Tissue  assay  of

   phenobarbital,  phenytoin  and.ρ 一hydroxyphenytoin

   by  high-perfbrmance  liquid  chromatograph》 へJ

   Chromatogr  285,500-508(1984).

5.Teunissen  MW,  Meerburg-van  der  Torren  JE,

   Vermeulen  NP,  Breimer  DD.  Automated

   high-performance  liquid  chromatographic

   determination  of  antipyrine  and  its  main

   metabolites  in plasma,  saliva  and  urine,  including

   4,4Ldihydroxyantipyrine.  J   Chro〃zatogr  278,

   367-378(1983).

6.Lin  JH,  SugiyamaヱAwazu  S, Hamano  M.  In

   vi〃o and  in vivo  evaluation  of the tissue-to-blood

   partition     coefficient     fbr     physiological

   pharmacokinetic  models.  」  1)harmacokinet

   Bioρhar〃z  lO,637-647(1982).

7.King  Fq  Dedrick  RL.  Physiologic  model  fbr the

   pharmacokinetics  of  2,deoxycoformycin  in

   normal  and  leukemic  mice.  J  Pharmacokinet

   Bioρhar〃z  9,519-534(1981).

8.Chen   HS,  Gross  JF.  Estimation    of

   tissue-to-plasma  partition  coefficients  used  in

   physiological  pharmacokinetic  models.」

   1)harmacokinet.Bioρhar〃17,  l l 7-125(1979).

9.Gallo  JM,  Lam  FC,  Perrier  DG  Area  method  fbr

   the  estimation  of  partition  coefficients  fbr

   physiological  pharmacokinetic  models.  J

   Pharmacokinet  Bi(4フhar〃z  l 5,271-280(1987).

10.Kety  SS,  Schmidt  CF.  The  determination  of

   cerebral  blood  flow  in man  by  the use  of nitrous

   oxide  in low  concentrations.!抽 」1)hysiol  l 43,

   53-66(1945).

ll.Lassen  NA,  Perl  W.  Tracer  kinetic  methods  in

   medical  physiology.  Raven  Press,  New  Y6rk

   l979.

12.Kakutani  T, Yamaoka  K, Hashida  M,  Sezaki  H. A

   new  method  fbr assessment  of drug  disposition  in

   muscle:apPlication  of  statistical  moment  theory

   to  local  perfUsion  systems.  J  Pharmacokinet

   Biophar〃z  13,609-631(1985).

13.Watari  N,  Benet  LZ.  Determination  of mean  input

   time,  mean  residence  time,  and  steady-state

   volume  of distribution  with  multiple  drug  inputs.

   」1)har〃zacokinet  Bi()pharm  l 7,593-599(1989).

8 JKyoritsu  Univ  Pharm  2007.  O  Vbl.3


Recommended