Acta Poloniae Pharmaceutica Drug Research, Vol. 74 No. 5 pp.

15271531, 2017 ISSN 0001-6837

Polish Pharmaceutical Society

APPLICATION OF LAPLACE TRANSFORMS TO A PHARMACOKINETIC

OPEN TWO-COMPARTMENT BODY MODEL
TADEUSZ WADYSAW HERMANN

Department of Physical Pharmacy and Pharmacokinetics, Pozna University of Medical Sciences,

6 wicickiego St., 60-781 Pozna, Poland

Abstract: The anti-Laplace of complicated transforms for two-compartment differential equations may be
found only in an extensive table of Laplace transforms which usually are not available. Therefore, a general par-
tial fraction theorem was used for obtaining their inverse Laplace transforms. First, the disposition function of
the central compartment in a linear two-compartment mammillary model, ds,C must be written down. Second,
except for that disposition function appropriate intake functions were considered with intravenous instanta-
neous bolus injection, constant zero-order rate infusion, and intramuscular first-order single dose injection of
indobufen (Ibustrin). The product of the disposition function and the appropriate input function yields the
experimental Laplace transforms for the amount (concentrations) of drug in the central compartment, as,C (XC).
The input functions for the above mentioned routes of drug administration are equal: X0 (the dose), k0 (1
eTs)/S, and ka F X0/ (s + ka), respectively. The equations derived are also illustrated in four figures.

Keywords: two-compartment body model, Laplace transforms, the inverse Laplace, partial fraction theorem,
disposition function, different intake function: bolus injection, infusion, intramuscular injection

The use of Laplace transforms of a function, A two-compartment model with intravenous

LF(t), and the inverse operation, L-1 , may be applied bolus injection
to obtain solutions to the systems of linear differen- The experimental Laplace transform for the
tial equations of the first-order. However, sufficient disposition function of the central compartment, ds,C
matrix algebra is needed to use the above transforms in a linear two-compartment mammillary model
(1). That operation transforms differentiation into where elimination of drug from central compartment
multiplication as well as integration into division occurs has been provided (2-4):
(1). The Laplace transform enables complex rate N

expressions to be manipulated easily by convention- (s + Ei)

i=2
ds,C = (1)
al algebraic techniques (2). However, it is true with N
(s + i)
regards to a one-compartment body model. The anti- i=1

Laplace of the resulting complicated transforms may where: E1 = k10+k12 and E2 = k21, and i is a disposition
be found only in an extensive table of Laplace trans- rate constant, which may be expressed in terms of
forms if two-compartment body model is concerned. the above individual intercompartmental transfer
Therefore, according to Gibaldi and Perrier (2), and rate constants and elimination rate constants,
Benet (3, 4), it is easier to use a general partial frac- continued product, s the Laplace operator which
tion theorem for obtaining inverse Laplace trans- replaces the time domain of a rate equation.
forms (4). To illustrate the application of this The model concerned is presented below:
approach for solving linear differential equations of k12
the first-order a two-compartment pharmacokinetic Xc XT
model with the first- or zero-, or bolus (instanta- k21
neous) input will be employed. It should be men- k10
tioned that the application of the Laplace transforms where XC and XT are the amounts of a drug as a func-
to a one-compartment body model is much easier tion of time in central and tissue compartments, and
and has been explained quite simply (2, 5, 6). k12, k21, and k10 are the first-order rate constants of

* Corresponding author: e-mail: [email protected]; phone: +48 61 693 101 555

1527
1528 TADEUSZ WADYSAW HERMANN

transfer a drug from the central compartment to the P(s) N

P(i)
L-1 = e t (5)
i

tissue compartment and vice versa as well as elimi- Q(s) i=1 Qi(i)
nation rate constant from the central compartment, The roots of the polynomial, Q(s) are 1 = and 2
respectively. The product of the input, ins, and dis- = . The term Qi(i) may be defined as follows.
position functions yields the Laplace transform for When
the amount of drug in the central compartment, as.c i = 1 Qi(i) = (i + ) = ( )
as,C = ins ds,C (2) i = 2 Qi(i) = (2 + ) = ( ) (6)
where ins = X0 (a single dose injected) for the above Hence the solution for the amount of drug in the
model. central compartment, XC, applying the general par-
Therefore tial fraction equation, is obtained
N
N
(s + Ei) (Ei l)
N
i=2
as,C = X0 (3) XC = X0
i=2
N
e t =
i

(i l)
N l=1
(s + i) i=1
i=1
X0(K21 ) X0(K21 )
The above equation may be rewritten for the = e t + e t
( ) ( )
central compartment amounts of drug following X0
intravenous bolus injection XC = [( k21) e t + (k21 ) e t

X0 (s + E2) P(s)
as,C = = (4) X0( k21) X0(k21 )
(s + 1) (s + 2) Q(s) when = A i = B (7)

The anti-Laplace of the resulting transform, L-1 may t t
XC = A e + B e (8)
be obtained by use of a general partial fraction theo-
rem. If the quotient of two polynomials P(s)/Q(s) is A plot of the logarithms of drug plasma con-
given as above, then centrations (C = XC/Vd) versus time according to the

Figure 1. Semilog plot of plasma levels vs. time after intravenous bolus administration of the iodine salt to a healthy volunteer described
by biexponential equation and the method of residuals (2, 6)
 Application of Laplace transforms to a pharmacokinetic... 1529

Figure 2. Serum concentrations of (-)-R-indobufen enantiomer vs. time after single 200 mg intramuscular dose administration of racemic
indobufen (Ibustrin) to a healthy volunteer described by TopFit 2.0 computer program for an open two-compartment body model (6, 7)

2
above equation will yield a biexponential curve ka F X0 (s + Ei)
i=2
(Fig. 1). as,C = (11)
2
The disposition constants and may be (s + ka) (s + i)
i=1
obtained applying the method of residuals (Fig. 1).
Solving for the amount of drug in the central
A two-compartment model with a first-order compartment by taking anti-Laplace yields
input process 2
(Ei ka)
For a drug that enters the body by an apparent i=2
XC ka F X0 eka t +
2
first-order absorption process (generally via the oral (i ka)
i=1
or intramuscular routes) and distributes in the body 2

according to a two-compartment model: 2 (Ei l)

+ ka F X0
i=2
el t (12)
k12 i=1 2
(ka l) (i l)
k
Xa a XC XT i=1

k21 k21 ka
XC = ka F X0 eka t +
k10 (1 ka) (2 ka)
k21 1
+ ka F X0 e1 t +
where: Xa amount of drug at an extravascular site (ka 1) (1 2)
of absorption as a function of time , ka the first- k21 2
order rate constant of absorption. + ka F X0 e2 t
(ka 2) (2 1)
The disposition function for the central com-
partment is identical to the disposition function for When the hybrid rate constants and (1) as
an intravenous bolus injection
N
well as the other constants B, A, and C0 the corre-
(s + Ei) sponding zero-time intercepts obtained by the
i=2
ds,C =
N
(9) method of residuals, are substituted, respectively,
(s + i) the equation may be transformed to a simpler form:
i=1
However, the input function is different to describe k21 ka
XC ka F X0 [ eka t +
first-order absorption (2, 5, 6) ( ka) ( ka)
ka F X0 k21 k21
ins = (10) + e t + e t
s + ka (ka ) ( ) (ka ) ( )
where F determines the fraction of drug absorbed. XC = B e t + A e t C0 eka t (13)
The Laplace transform for the amount of drug
in the central compartment equals the product of the The plots of the above equation presented on
disposition and input functions both a graph and on a semilog papers are very char-
1530 TADEUSZ WADYSAW HERMANN

Figure 3. Semilog plot of serum (-)-R-indobufen enantiomer vs. time after single 200 mg intramuscular dose administration of racemic
indobufen (Ibustrin) to a healthy volunteer described by TopFit 2.0 computer program for an open two-compartment body model (6, 7)

Figure 4. Semilog plot of plasma levels vs. time of a drug that confers two-compartment model characteristics, following constant rate
intravenous infusion to steady-state ( ___ ) and following the rapid intravenous injection of a dose that gives an initial drug concentra-
tion equal to the steady-state concentration (_ _ _) (2)
 Application of Laplace transforms to a pharmacokinetic... 1531

acteristic for a two-compartment model (Fig. 2 and It should be noted that the above single equa-
3). tion describes the amount of drug in the central com-
The experimental data presented in Fig. 2 and partment as a function of time while the infusion is
3 have been excerpted from a published article (7). being carried out and after infusion stops. While the
infusion is continuing, T = t and varies with time
A two-compartment model for an intravenous k0 ( k21) (e-t e-2t)
Xc = +
infusion ( )
k12
k
0 XC XT k0 ( k21) (e-t e-2t)
+ (18)
k21 ( )
k10 However, when infusion ceases, time t
where: k0 the zero-order rate of drug infusion (con- becomes a constant corresponding to T the dura-
stant). tion of the infusion. It should be apparent that upon
The disposition function for the central com- stopping the infusion, drug concentrations in the
partment is identical to the disposition function for plasma decline in a biexponential manner (Fig. 4).
an intravenous bolus injection. It can be seen from Fig. 4 that the biexponen-
s + E2 tial characteristics of the drug is more evident fol-
ds,C = (14) lowing the bolus injection than after terminating the
(s + 1) (s + 2)
Multiplication of this disposition function by infusion.
the input function for an intravenous infusion begin-
ning at time zero (i.e., ins = k0 (1 e-Ts)/s (2, 5, 6) Acknowledgment
yields the following Laplace transform for the
amount of drug in the central compartment This publication has been written to give
k0 (s + E2) (1 e-Ts) thanks to the Almighty God for my 80th birthday
as,C = (15) anniversary which hopefully I am supposed to sur-
s (s + 1) (s + 2)
where T is the duration of an infusion. vive on June 13, 2017.
The above two polynomials fulfill the require-
ments for the use of a partial fraction theorem for REFERENCES
obtaining inverse Laplace transforms. Hence the
solution for the amount of drug in the central com- 1. Wagner J.G., Pernarowski M.: Biopharma-
partment XC as a function of time may be written ceutics and relevant pharmacokinetics, Drug
k0 (E2 1) (1 e-lT) Intelligence Publications, Hamilton, Illinois
Xc = el t + 1971.
1 (2 1)
2. Gibaldi M., Perrier D.: Pharmacokinetics, 2nd
k0 (E2 2) (1 e-2T) edn., Marcel Dekker, New York and Basel
+ e2 t (16)
2 (1 2) 1982.
The roots of the polynomial, Q(s) were speci- 3. Benet L.Z.: J. Pharm. Sci. 60, 1593 (1971).
fied previously, and the equation may be rewritten 4. Benet L.Z.: J. Pharm. Sci. 61, 536 (1972).
k0 ( k21) (1 e-lT) 5. Hermann T.W.: Farm. Pol. 71, 289 (2015).
Xc = et +
( ) 6. Hermann T.W.: Pharmacokinetics. Theory and
k0 ( k21) (1 e-T) practice. PZWL, Warszawa 2002.
+ et(17)
( ) 7. Gwka F.K.: Chirality 12, 38 (2000).
where: k0 is the zero-order rate of the infusion.
Received: 9. 09. 2016