provide examples of quantitative urine concentra-
tion monitoring, and enable an assessment of the
glomerular filtration rate (GFR). The GFR can
measure renal injury with a greater degree of sen-
sitivity than measuring serum creatinine or urinary
protein excretion.
Cerebrospinal fluid (CSF)
It is a widely held myth that drug in the CSF has
crossed the ‘blood–brain barrier’. This demon-
strates an ignorance of basic anatomy and physiol-
ogy. If a drug has appeared in the CSF it may
have got there by filtration or secretion by the
choroid plexi in the lateral ventricles, or by diffu-
sion from the circulation directly. It is, therefore,
rash to assume thatdrug concentration in the CSFis
a good surrogate for actual brain exposure(Davson,
1967; De Lange and Danhof, 2002).Vice versa,
when a drug is not found in the CSF, to assume that
it is not present in the brain parenchyma presumes
an absence of sequestration. Buprenorphine is a
good example, where absence of detectable drug in
the CSF (and venous blood) correlates with a pro-
longed analgesic effect.
Quite apart from the so-called ‘blood–brain bar-
rier’ (and where it is lacking, e.g. some parts of the
pituitary, hypothalamus and brain stem, i.e. the
chemoreceptor trigger zone), there are many
other factors which govern equilibration of drug
concentration between CSF and the parenchyma of
the brain itself. The differential effects of P-glyco-
protein saturable active transport can govern the
CNS sequestration in a manner that is completely
unrelated to relative lipophilicity or ambient drug
concentration.
Lastly, there are obviously more technical and
clinical obstacles to obtaining CSF for pharmaco-
kinetic purposes than when sampling venous blood
or urine. These illustrate the topographical com-
plexities of measuring CSF concentrations. In
animal studies, after systemic drug administration,
drug concentrations in the CSF can be unequal
between ventricles and the subarachnoid space.
Clinically, CSF from the cisterna magna and
from around the corda equina can also differ in
drug concentration. It is for this reason that pre-
clinical scientists often resort to intracerebral
microdialysis, and this is also why magnetic reso-
nance spectroscopy and positron emission tomo-
graphy are now being pursued more commonly for
drug studies. The scope for useful CSF concentra-
tion monitoring in the ordinary clinical situation
remains vanishingly small.28.5 Summary
In this chapter, the measurement of drug concen-
trations in humans has been placed into the con-
text of ordinary clinical practice, rather than the
research environment. This is, therefore, a chapter
that impinges on product labeling and risk man-
agement plans, and not necessarily pharmacoki-
netics or the quantification of drug interactions in
normal volunteers. The general criteria for when
plasma concentration monitoring may or may not
be worthwhile have been reviewed. The essen-
tially qualitative nature of urine monitoring,
unless measuring GFR, and some fundamentals
about CSF drug concentrations have also been
reviewed.References
Back DJ, Khoo SH, Gibbons SE, Barry MG, Merry C.- Therapeutic drug monitoring of antiretrovirals
in human immunodeficiency virus infection.Ther.
Drug Monit. 22 : 122–126.
Bell R, McLaren A, Galanos J, Copolov D. 1998. The
clinical use of clozapine levels.Aust. NZ J. Psy-
chiatr. 32 : 567–574.
Buchowsky SS, Partovi N, Ensom NH. 2005. Clinical
pharmacokinetic monitoring of itraconazole is war-
ranted in only a subset of patients.Ther. Drug Monit.
27 : 322–333.
Davson H. 1967.Physiology of the Cerebrospinal
Fluid. J. & A. Churchill:London,passim.
De Lange EC, Danhof M. 2002. Considerations in
the use of cerebrospinal fluid pharmacokinetics to
predict brain target concentrations in the clinical
setting: implications of the barriers between blood
and brain.Clin. Pharmacokinet. 41 : 691–703.
REFERENCES 379