If the drug has a slow disposition phase and
plasma drug concentrations higher then the steady-
state values are to be avoided, the loading dose is
individualized based on the initial volume of distri-
bution. If plasma drug concentrations higher than
the steady-state values can be tolerated, the beta
volume of distribution is used to calculate the
loading dose,
D
l
=
C
ss
,
avg
V
F
Utilizing the fact that
V
=
Cl
yields,
V
,
indiv
V
,
avg
=
Cl
indiv
Cl
avg
avg
indiv
where
β
avg
and
β
indiv
are the population average and
individual values of
β
, respectively.
β
is determined
in part by the slow distribution process and in part
by the clearance rate. Individualizing for clearance
rate, the ratio
β
avg
/
β
indiv
will take on a value between
Cl
avg
/Cl
indiv
and 1 depending upon the magnitude of
the slow distribution process. Taking into account
the maximum clearance rate effect gives,
individualized D
l
=
usual D
l
F
avg
F
indiv
Cl
indiv
Cl
avg
Calculation of V
o,indiv
/V
o,avg
.
The only predictor
of the initial volume of distribution routinely availa-
ble clinically is body weight, so dose individualiza-
tion is based on weight measurement. The usual
clinical practice is to calculate the individualized
dose as
dose
unit body weight individual body weight
or
target C
(
0
)
V
o
,
indiv
rather than individualizing a usual dose.
As an example of the use of the target C(0)
formula, consider the computation of a loading dose
of coagulant factor VIII in a 15-year-old child with
factor VIII deficiency who requires replacement
therapy because of an uncomplicated hemarthrosis.
Because it binds to von Willebrand factor, which is
almost entirely intravascular, factor VIII rapidly
distributes into a volume of distribution equal to the
plasma volume. In nonobese individuals, the plasma
volume is approximately equal to 40 ml/kg body
weight. If the patient weighs 60 kg, his volume of
distribution is, therefore, estimated to be 2400 ml..
For a target initial plasma factor VIII concentration
of 0.4 U/ml,
individualized D
l
=
0.4
U
ml
%
2400
ml
=
960
U
Calculation of Cl
indiv
/Cl
avg
.
For a drug elimi-
nated predominantly by the kidneys, the ratio
Cl
indiv
/Cl
avg
is well approximated by the ratio
GFR
indiv
/GFR
avg
even if the drug is cleared by both
glomerular and tubular mechanisms. An individ-
ual’s GFR can be determined by measuring the
urinary clearance rate of an endogenous substance
eliminated solely by glomerular clearance. Then,
GFR
indiv
=
urine excretion rate
plasma concentration
Creatinine is the endogenous substance used clini-
cally. It is not a perfect glomerular clearance
marker substance in that a small fraction is elimi-
nated extrarenally and some of its renal elimination
is due to tubular secretion. However, extensive
clinical experience has shown that the creatinine
clearance rate is a reliable measure of GFR in both
healthy individuals and in individuals with kidney
disease (Giovanetti and Barsotti 1991
)
.
The difficulty with the direct measurement of
creatinine clearance rate is the necessity for an
accurately timed and complete urine collection on
which to base the measurement of the urine excre-
tion rate. This problem can be circumvented by
employing an alternative expression for clearance
rate,
clearance rate
=
synthesis rate
plasma concentration
and using an estimate of the synthesis rate for creati-
nine based upon gender, body size, and age.
Formulas for calculating creatinine clearance rate,
and thereby GFR, based upon this clearance rate
expression are presented in Table 12.1. There are
numerous similar formulas and nomograms in the
literature (Lam
et al.
1997). The formulas given in
Table 12.1 have been found to be clinically reliable
(Schwartz
et al.
1987, Luke
et al.
1990).
Consider, for example, the calculation of an
individualized maintenance dose of amikacin in a
patient being treated for
E. coli
bacteremia. The
patient is a 50 year old male who weighs 60 kg and
has a plasma creatinine concentration of 2 mg/dl.
Drug Therapy
12-8
