258 BIOPHARMACEUTICS AND PHARMACOKINETICS COMPARTMENT MODELING 259

Extrapolation of linear segment to time axis yields x-intercept equal to defme KE of the drug. The absorption rate constant Ka can be estimated
I /KE since when Xu = 0, t= 1/KE (Fig. 10.11). by the method of residuals using the same data i.e. equation I 0.82.
.
Urinary excretion data after oral administration can also be treated.. ...
according to Wagner-Nelson method to calculate Ka by construction of%_·
ARA plots. The method requires urine collection for sufficiently long I

time to ensure accurate estimation of KE but need not be collected to. time
infmity. The equation derived to relate % ARA with urinary excretion .

i rate is:
dX0 /dt + KE Xu
.. o/oARA = 1 - 1-------­ 100 (10.83)
KE Xuoo
.
A semilog plot of % ARA versus t yields a straight line with slope_
-> t
-Ka/2.303.
Fig. 10.11 Regular plot of Xu versus t during Accurate determination of Ka from urinary excretion data is possible
constant rate i. v. infusion only for drugs with slow rate of absorption since for drugs with rapid
Relationships for rate of excretion when the drug is administered e.v. absorption, collection of urine samples at very short intervals of time is
can also be given similarly. Thus: difficult.
dXu
__=Ke X (10.71) MULTICOMPARTMENT MODELS
dt
(Delayed Distribution Models)
For a drug given e.v. and absorbed by a frrst-order process, X is given
The one-compart111ent model adequately describes phar111acokinetics of
as: many drugs. Instantaneous distribution equilibrium is assumed in such
(10.48) cases and decline in the amount of drug in the body with time is ex­
pressed by an equation with a monoexponential term (i.e. elimination).
Substitution of equation 10.48 in equation 10.71 and integration o� the However, instantaneous distribution is not truly possible for an even larger
same yields: number of drugs and drug disposition is not monoexponential but bi- or
multi-exponential. This is because the body is composed of a heteroge-­
Ke Ka F X0 1 KE.e-Kat neous group of tissues each with different degree of blood flow and·
Xu = -+---- (10.80)
KE Ka (KE - Ka) affmity for drug and therefore different rates of equilibration. Ideally, a
true pharmacokinetic model si1ould be the one with a rate constant for
At time infinity, the equation l 0.80 reduces to: each tissue undergoing equilibrium, which is· difficult mathematically..
Ke F X0 Th� best approach ·is therefore to pool together tissues on the basis of
Xuoo = (10.81)
KE similarity in their distribution characteristics. As for one-comparbnent
models, drug disposition in multicompa1 t111ent systems is also assumed to
Substitution;of equation I 0.81 in equation l 0.80 and subsequent rear­ occur by first-order. Multicompartment characteristics of a drug- are best
rangement yields: understood by giving it as i.v. bolus and observing the manner in which
I Xu00 the plasma concentration declines with time. The number of exponentials
ARE = (Xuao- Xu ) = · (Ka e-KEt - KE e-Kat) (10.82) required to describe such a plasma level-time profile deter111ines the num-
(Ka - KE) ..
ber of kinetically homogeneous compartments into which a drug will
A semilog plot of (Xu00 - Xu) versus t results in a biexponential curve distribute.
and if K8 > KE, the slope of the ter111inal linear portion of the curve '":'ill
 261
260 ,BIOPHARMACEUTICS AND PHARMACOKINETICS
C'OMPARTMENT MODELING

TWO-COMPARTMENT OPEN MODEL ence of two disposition processes viz. distribution and elimination. These
two processes are not evident to the eyes in a regular arithmetic plot but
The commonest of multi�ompartment models is two compartment model. when a semilog plot of C . versus t is m,ade, they can be identified .(Fig.
In such a model, the body tissues are broadly classified into 2 categories: l 0.12). Initially, the concentration of drug in the central compartinent
1. Central Compartment or Compartment 1 comprising of blood declines rapidly; this is due to the distribution of drug from the· .ce�tral
and highly perfused tissues like liver, lungs, kidneys, etc. that compartment to the peripheral compartment. The phase during whic�.1this
equilibrate with the drug rapidly. Elimination usually occurs from occurs is therefore called as the distributive phase. After somet¢1e, a ·
this compartt1.1ent. pseudo-distribution equilibrium is achieved between the two c9tn�art­
2. Peripheral or Tissue Compartment or Compartment 2 com­ ments following which the subsequent loss of drug from the/ central
prising of . poorly perfused and slow equilibrating tissues such as compartment is slow and mainly due to elimination. This second, slower
rate process, is called as the post-distributive or elimination phase. In
.·

muscles, skin, adipose, etc._ and considered as a hybrid of several

functional physiologic
••
units. contrast to the central compart1nent, the drug concentration in the periph­
eral compart111ent frrst increases and reac�es a maximum. This corresponds
Classification of a particular tissue, for· example brain, into central or
with the distribution phase. Following ··peak, the drug concentration de­
p�ripheral comparttnent depends UI?:2!!_ _the physicochemical properties of
clines which corresponds to the post-distributive phase (Fig. I 0.12).
'
the drug. A highly lipophilic drug can cross the BBB and brain would
· -� then be included in the central compartment. In contrast, a polar . drug -- - ----------- ---------
cannot penetr�te the BBB and brain in this case will be a part of periph­
eral comparttneqt despite the fact that it is a highly P$!rfused organ. \.---------Initial rapid decline
· Depending upon the compart.t11ent from which the drug is eliminated, distributive phase
. the -two-compartment model can be categorized into 3 types:
C
0

·1. Two-comparttnent model with elimination from central compart­

-------------
. -­
ment. I •

.____ Slow terminal decline

2. Two-compartment model with elimination from peripheral com­ elimination phase
part111ent. ee,,"rq/
' '-

3�· Two-compartment model with elimination &6m both the compart-

Ca
m�. ��

""""e ,
In the absence of infor1nation, elimination is. assumed to occur exclu­ ,,

sively from central compart1nent.

'•

- . Two."'.'.Co�p,artment Open Model-Intravenous ->i Time I

Bolus · Administration ,,
Fig. 10.12 Changes in drug concentration in the central (plasma) and the peripheral
. · The model can· be depicted as shown below with elimination from the compartment after i.v. bolus of a drug that fits two-compartment f!10del.
centr.al COlll:part1nent.
Let K 12 and K2 I be the first-order distribution rate constants depicting ,, .

1 ..
' K12 2 drug transfer between the centr�l and the peripheral compart111ents and let · ·.
Central Peripheral subscript c and p define central and peripheral compartment r�spectively.
Compart'l'.'ent
'\
K21 Compartr11ent I
• The rate of change in drug c9ncentration in the central compart111ent is
KE
given b·y:
. ', "'
' ( I 0.84)
After the i.v. bolus of a drug that follows two-compa11t11ent kinetics,
the d�cline in plasma concentration is biexponential ·indic�ting
•
the pres-
'
 . 263
262 BIOPHARMACEUTICS AND PHARMACOKINETICS COMPARTMENT MODELING

Xo K21 - a
- K21- a (10.93)
Extending the relationship X = Vde to the above equation, we have A=
Vc p-a p-a
dCC K21 xp K12 Xe KE Xe
(10.85) · Xo K21 - �
dt VP · Ve Ve K21 - P (10.94)
B=
Ve a-P a-�
where Xe and Xp are the amounts of drug in the central and peripheral
compartments respectively and Vc and V are the apparent volumes of the nc en tra tio n im m ed ia tely aft er i. v. in jec tio n.
where C0 = plasma dr ug co
P
central and th� peripheral compart1nent respectively. The rate of change ne nt ia l di sp os iti on cu rv e ob tai ne d
Method of Residua ls : Th e bi ex po
in. drug concentration in the peripheral compartment is given by: m pa rtm en t m od el ca n be ·re so lv ed
after i.v. bolus of a drug that fits tw o co
dCp po ne nt s by th e m eth od of re sid ua ls. Rewriting the
into its individual ex
= K12 Cc - K 21 Cp (10.86).
dt equation 10.92: t
•

C = A e-a t
+ B e- J3 (10.92)
_ K12 Xe K21 Xp ex po ne nt ial cu rv e gi ve n in Fi g. l 0. 12 ., th e
(10.87) As apparent from the bi
Ve Vp to di str ib ut io n is m or e ra pi d th an _ th e te1 1n in al de clin e
initial decline due -at
on i.e. th e rate co ns tan t a >> B an d- . he nc e th e ter rn e
Integr�tion of e9u�tions I 0.85 and 10.87 yields equations that describe due to elim in ati
th an do es e- 13t. Th us , eq ua tio n 10 .9 2 re du ce s
the concentration of ,�g in the central and peripheral compartments at approaches zero much faster
any given Jime t: . to:
(.10.95)
­Xo
C = (10.88)
c VC In log fonn, the equation becomes: \

� pt ( 10.96)
Xo K12 -at K12 J3t C = log B- _
log
C = ­ ( )e +( ) e- ( I 0.89) 23
P Vp P-.a a- p
C
where = back extrapo lat ed
nn
pl
in
as
a
m a
lin
co
ea
nc
r
en
ph
tra
as e
tio
of
n va
th e
lu es
cu
.
rv e
A se
ha
m
vi
ilo
ng
g
slo
p lot
pe
where X0 = i. v. bolus dose, a and P are hybrid first-order constants for of C versus t yields the te l
g
la te d to tim e ze ro , yi el ds y- in te rc ep t_ lo
the rapid distribution phase and the s low e limination phase respectively -�/2.303 and w he n back extrapo m
fo r th e e im in at io n ph as e ca n be ob ta in ed fro
which depend entirely upon .the first-order constants K 12, K21 and KE. B (Fig. 10.13 .).The ty 2
l

The constants �12 and K21 that depict reversible transfer of drug equation t1;2 = 0.693/p.
pl as m a co nc en tra tio n va lu es . of th e el im in a­
between compartments are called as microconstants or transfer con­ Subtraction of extrapolated
( ) from th e co rre sp on di ng tru e pl as m a co nc en tra tio n
stants. The mathematical relationships between hybrid and microconstants tion phase equation 10.95 C .
ds a se rie s of re sid ua l co nc en tra tio n va lu es r
are given as: values (equation 10.92) yiel
a + p = K12 + K21 + KE (10.90) = C - C = A
�

e-at (10.97)
Cr
a p = K21 KE (10.91)
In log form, the equation becomes:
Equation 10.88 can be written in simplified fo1m as: at (10.98)
leg Cr == log A - ---
Cc = A e-at + B e-�t (10.92) 2.303
Distribution Elimination A semilog plot of Cr versus t yields a straight line with slope -a/2.303
C = +
c exponent exponent and }'-intercept log A (Fig. l 0.13 ).
. where A and B are also hybrid constants for the two exponents and can
be resolved graphically by the method of residuals.
264 BIOPHARMACEUTICS AND PHARMACOKINETICS COMPARTMENT MODELING 265
A B
,._________ fog c 0 AUC - + (10.103)
Cl

The apparent volume of central compartn1ent Vc is given as:

1.....--------log A
Xo
V = - (10.104)
c Co
L+-�-------· log B
Back extrapolated terminal portion Apparent volume of peripheral compartment can be obtained from
tog C
.....------ �
equation:
of elimination phase (log C value.s)
�
1----True plasma concentration Vp = (10.105)
values (log C values)
Residual Line (log Cr values) •
The apparent volume of distribution at steady-state or equilibrium can
. slope = -� I 2.303 now be defined as:
\ slope = -a I 2.303 (10.106)

-� t .It is also given as:

Fig. 10.13 Resolution of biexponential plasma concentration-time curve by Vd,area = (10.107)
the method of residuals for a drug that follows two-compartment J3 AUC
kinetics on i.v. bolus administration.
Total systemic clearance is given as:
Assessment of Pharmacokinetic Parameters : All the parameters of (10.108)
equation 10.92 can be resolved by the method of residuals as described
above. Other parameters of the model viz. K12, K 2 1, KE, etc. can now be The pharrnacokinetic parameters can also be calculated by using uri­
derived by proper substitution of these values. nary excretion data:
dXu
C0 =A+ B (10.99) --=Ke Ve (10.109)
dt
_ a J3 C0
KE- (10.100) An equation identical to equation 10.92 can be derived for rate of
Af3+Ba excretion of unchanged drug in urine:
K _ A B (J3 - a)2 dXu =
12 - (10.101) Ke A e---at + Ke B .e-Pt (10.110)
C0 (A J3 + B a) dt
Af3+Ba The above equation can be resolved into individual exponents by the
K21 = (10.102) method of residuals as described for plasma concentration-time data.
Co
Renal clearance is given as:
It must be noted that for two-compartment model, KE is the rate
constant for elimination of drug from the central compartment and J3 is the CIR - Ke Ye (10.111)
rate constant for elimin�tion from the entire body. Overall eli�ination t Yi
should therefore be calculated from J3. Two-Compartment Open Model
-Intravenous Infusion
Area under the plasma concentration-time curve can be obtained by
The model can be depicted as shown below with elimination from the
the following equation:
central compart1nent.
 266 BIOPHARMACEUTICS AND PHARMACOKINETICS COMPARTMENT MODELING 267

Ro 1 K12 2 The plasma concentration at any time t is given by eq�ation:

- Central Peripheral a + .
Compartment \
K21 Compartment C =N e-K t L e-ut + M e-�t (10.116)
•

Abso rptio n Distr ibuti on Elimination

KE
C = exponent + exponent + exponent
',,•

The plasma or central ·compartment concentration\ of a drug that fits where Ka, a and � have usual meanings. L, M and N are coefficients.
wo com art ent model when administered as constant rate (zero-order)
� � � � The 3 exponents can be resolved by stepwise application of method of
1. v. 1nfus1on, 1s given. by equation: '.I
residuals assuming Ka > a > � as shown in Fig. I 0.14. The various
Ro KE - K E - a phar1nacokinetic parameters can then be estimated.
p
- ) e-at + ( ) e-Pt
C=-- I +( (10.112)
Vc KE P-a a-p .
At steady-state (i.e. at time infinity), the second and the third tenn in
lo g N
the bracket becomes zero and the equation reduces to:
, True plasma concentration
curve (log C values)
(IO.I13)
log L Back extrapolated distri�ution
t-
Now VcKE = VdB. Substituting this in equation JO.113 , we get : curve (log C -C) val•,es
-
log M-- \'-!,. � Back extr olated elimination
'!';
(10.114) .......�-�-�-� curve (log C values)
'\'- (lo g C r va lu es )
---- Fi rs t re si du al cu rv e 1
�lilllllll �..-.--

The loading dose Xo.L to obtain C 55 immediately at the start of infu­

.
sion can be calculated from equation:
\'
-- Second residual liAe
rption) log Cr values
(abso 2
.slope = -'3 I 2.303
(10.115) log C
.slope=
-a/ 2.303
Two-Compartment Open Model
-Extravascular i\dministration \ slope = -Ka / 2.303
F'irst-Order Absorption
The model can be depicted as follows:
--) t

Fig. 10.14 Seinilog plot of C versus t of a drug with two-compartment

Ka 1 K12 2 characteristics when administered extravascularly. The various
.
Central Peripheral exponents have been resolved by the method of residuals.
Compartment K21 -.Compartment
Besides the method of residuals, Ka can also be estimated by Loo-
KE
Riegelman method for a drug that follows two-cornpart111ent characteristics.
"...
This method is in contrast to the Wagner-Nelson method for deter111ination
� or a drug that enters. the body by a first-order absorption process and
. of Ka of a drug with one-compartment characteristics. The Loo-Riegelman
d1str1buted according to two compartment model, the rate of change in
method requires plasma drug concentration-time data both after oral and
drug concent'.ation in the central compartment is described by 3 expo;ents
i. v. administration of the drug to the same subject at different times in
an �bsorpt1on exponent, and the two usual exponents tl1at describe drug._
. . order to obtain all the necessary kinetic constants. Despite its complexity,
d1spos1t1on.
268 BIOPHARMACEUTICS AND PHARMACOKINETICS 269
COMPARTMENT MODELING
the method can be applied to drugs that distribute in any number of 18. What are the applications and limitations of method of resi duals?
compart111ents. !
19. What is flip-flop phenomenon and when is it observed?
va lue of Ka co mp ute d fro m me thod of
QUESTIONS 20. Under what circu ms tan ce s is the
residuals incorrect?
1. In one-compartment open mod el, what do you infer from plasma being de me rit s of W ag ne r-N els on me tho d in co mp uting
21. What are the m�rits and
called as reference compartment? Ka?
2. In compartment modeling, what d oes the term open mean? 22. Wha t is th� infl ue nc e of Ka an d KE on Cm ax, tm ax an d AUC?

3. Disposition of a drug that follows one-compartment kinetics is a monoexponential for ob tai nin g a va li uri na ry ex cre ti on data?
23. What criteria are necessa ry d
process. Explain. ges an isa va nta ge s of su ch a me tho d in assessment
What are the ad vanta d d d

4. Why do first-order rate equations require logarithmic transformations? of pharmacokinetic parameters?

ca cu lat ng KE fro m uri na ry ex cre tio n data?
5. With examples, explain what you understand by primary and secondary 24. What are the tw o me tho d s for l i

parameters. Compare the ir merits and demerits.

6. The expression Vd = XJC0 can only be used to estimate the. apparent 25. Why are urine samples required to be collected for at least 7 biolog� cal h�lf­
volume of distribution of a d rug that follows one-compartment kinet ics lives when using sigma-mi nus method whereas only two consecutive urine
when ad mi nistered as i.v. bolus. Why? samples are sufficient in case of rate of excreti on method?
7. Clearance is a more important parameter than half-life or elimination rate 26. Determination of Ka using ur inary excretion data is not suitable for rapidly
constant in expressing elimination. characteristics of a drug. Explain. absorbed drugs. Why?
es cla ssi fie in a tw o-c om pa rtm en t. _m od el? De pe nd i� g
8. Deflne clearance, total body clearance and organ clearance. What are the 27. How are body tissu d
ity , in wh ch co mp art me nt bra i' n · . sh ou ld be in-
advantages of expressing cleara�ce at an individual organ level? upon the d rug 's li po ph ilic i

9. Define and explain extraction ratio. How is it related to oral availability of cluded?
rap i ec lin e an of su bs eq ue nt slo we r decline
a drug? What is the influence of blood flow rate and protein binding on 28. What is the cause of ini tia l d d d

els aft er a m nis tra tio n of a d rug tha t follow s tw o-c om pa rtm ent
total clearance of drugs with high and those with low ER values? in plasma lev d i

10. What are the rate-limiting steps in the hepatic clearance of drugs? Based on kinetics?
m ea ni ng s fo r on e- an d tw o- co m pa rtm en t
ERH values, to which drugs do they apply? 29. The parameter KE has different
11. What are the advantages of ad ministering a d rug by constant rate i. v. models. Explai n.
nc en tra tio n of vio my c, n aft er i .v. bo lus ad mi nis tra tio n wa s
infusion? 30. If the plasma co i
l at 2 an d 4 ho urs res pe cti ve ly, ass um ing
found to be 10.0 and 5.5 mc g/m
12. Compute mathematically the number of half-lives required to attain 90% of
steady-state concentration after i. v. i nfusion (Hint : use equation I 0.41 ). one-compartment kinetics, calculate:

13. Prove mathematically that when an i.v. loading dose is followed immedi­ a. the half-li fe of the drug
ately by a constant rate infusion, the plasma concentration remains steady as Answ�r : 2.3 hours.
long as the infusion· is continued. b. the concentrat ion of drug in plasma at time zero
14. On what parameters does the time to reach steady-state depend? Answer : 18.2 mcg/ml.
15. How can the steady-state be attained rapidly? When is it advisable? Give c. the Vd i f dose administered was 30_0 mg
expressions for calculating such doses for a drug that fits one-compartment
model. Answer : 16. 5_ liters.
16. Compare t� absorption characteristics of drugs absorbed by zero-order with d. the total systemic clearance·
a
those absorbed by first-order process after e�.admi ni stration. Answer : 82.2 ml/min.
17. Show that after e.v. administration. the Cmax that can be attained is no more e. the renal clearance if fraction excreted unchanged in urine is 0.8
than 3 7°/o of maximum level attainable with the same dose given as i. v. I

Answer : 65.8 ml/min. ", ·

bolus. Why is this so?
270 BIOPHARMACEUTICS AND PHARMACOKINETICS COMPARTMENT MODELING 271

31. � 70 Kg patient is to be given oubain by i.v. infusion. The drug has a half­ e. When should the next dose be administered if the drug becomes ineffective
life of 22 hou �s, apparent Vd 15.7 liters and the desired steady-state plasma when the plasma level falls below 50 mcg/ml?
.
concentration 1s 0.0002 rncg/ml. Assuming one-compartment kinetics, cal­ Answer : after 1.2 hours.
culate:
f. What will be-the therapeutic index of the drug if the therapeutic range is 50
a. the time required to reach 90% of C55 to 500 mcg/ml?
Answer : 73 hours. Answer : TI = 10.
b. the infusion rate to achieve the desired C55 g. How much dose should be administered to attain an instantaneous plasma
Answer : 0.1 mcg/hour. concentration of 500 mcg/ml?
c. the loading dose to attain C55 rapidly Answer : 7000 mg.
Answer : 3.14 mcg. • h. For how long will the plasma level lie in the therapeutic window if the
above dose is given as i.v. bolus?
d. the concentration of drug in plasma after 48 hours from the start of infusion
Answer : 2.6 hours.
Answer : 0.00014 mcg/ml and when loading dose is given, 0.00018 mcg/ml.
34. The amount of drug excreted in urine after an i.v. dose of 400 mg of
32. The equation that best fits the pharmacokinetics of paracetamol after oral
norfloxacin was as follows:
administration of 500 mg dose is:
t (hours) Xu (mg) dX,ldt (mg/H) t*
C = 1.18 (e-0.24t _ e-l.6t)
0 0 \
Assuming one-compartment kinetics, determine-
2 49.05
a. peak time
4 80.0
Answer : 1.4 hours.
b. peak plasma concentration After completing the table and using rate of excretion method, determine-
Answer : 0.717 mcg/ml. a. the elimination rate constant and half-life of th� drug.
c. elimination half-life of the drug Answer : KE = 0.2303/hour and ty1 = 3 hours.
Answer : 2.88 hours. b. urinary exr,retion rate constant

d. apparent Vd if fraction bioavailable is 0.4 Answer : Kc = 0.077/hour.

Answer : 200 liters. c. (raction of drug excreted unchanged in urine

e. concentration of drug in plasma after 3 hours of administration Answer : 33.4°/o
Answer : 0 .' 565 mcg/ml. . d. whether the drug can be used for urinary tract infections?

33. The equation that best fits the plasma level time curve of azlocillin after an 35. The half-life of propranolol in a 60 Kg patient is 4 hours and Vd is
i.v. bolus dose of 2000 mg (assuming one-compartment kinetics) is: 5.5· liter/Kg.

C = 143 e--0.&7t a. Determine the total systemic clearance of the drug

a. What is the apparent Vd? Answer : CIT = 953 ml/min.
Answer : 14 liters. b. What will be its renal clearance if fraction excreted un�hanged in urine is
0.047?
b. Whar is the elimination t a;2 of the drug?
Answer : CIR = 44.8 ml/min._
l:�s-wer : 0.8 Hours.
r

c. Comment on the mechanism of renal excretion of drug.

c. What will be the plasma drug concentration after 6 hours?
d. If the drug is. eliminated only by hepatic and renal routes, what will be the
Answer : 0.77 i:ncg/ml. hepatic extraction ratio if blood flow to the liver is 1.5 liter/min?
d. How much of the drug will be left in the body ·after 6 hours? Answer : ERtt = 0.6054.
Answer : 10.8 mg.