efficacious and that it has only a low risk of serious
toxicity.
The target value for the monitoring marker
depends largely upon the seriousness of the illness
being treated. The more serious the disease, in
general, the higher the target value. The physician
and patient both want a high probability of effective
therapy and, usually, both are willing to accept a
higher risk of adverse effects. For the drug depicted
in Figure 12.10, a target value of 85 µg/ml, the
upper end of the therapeutic range, may be selected
in the face of a life-threatening arrhythmia. Indeed,
if the desired therapeutic effect is not achieved even
when the monitoring marker is at the upper end of
the therapeutic range, the drug dose may be
increased until toxicity intervenes or until the risk of
a serious adverse event is just too great.
Dose adjustment.
Dose adjustments are based
upon the assumption of linear drug kinetics. Linear
kinetics means that plasma drug concentrations are
directly proportional to drug dose. Using the target
value selected for the monitoring marker,
adjusted D
m
=
current D
m
target marker value
measured marker value
For example, if the steady-state trough plasma con-
centration for a drug is found to be 400 ng/ml and
the target trough concentration is 700 ng/ml, a
maintenance dose of 300 mg/ day should be adjusted
to 500 mg/day,
adjusted dose
=
300
mg
/
day
700
400
=
525
mg
/
day
given that the drug formulation is available in 100
mg gradations. A conservative approach would be
to increase the dose incrementally, first trying a dose
of 400 mg/day then, if indicated, increasing the dose
to 500 mg/day. In this way, if toxicity is ex-
perienced as a consequence of the increase in dose,
it may manifest itself at the lower, less toxic, dose.
The devil is in the details.
There are practical
concerns that have a tremendous impact on the
utility of therapeutic drug monitoring. First among
these is patient compliance with the drug regimen.
Rigorous adherence to a regular schedule of drug
taking is impossible, even for the most well-
intentioned patient. There are doses that are missed
and doses that are taken at irregular times. If this
happens in the period preceding the time of speci-
men collection for therapeutic drug monitoring, the
value for the monitoring marker will not be the true
steady-state for the marker and any adjustment made
in the dose will be ill-founded. To limit occurrences
of this sort, it is highly desirable to have the patient
keep a diary of his or her drug intake for the day or
two before the monitoring specimen is obtained.
For some patients it is even a good idea to witness
the drug administration that precedes the monitoring
specimen to confirm the timing of the specimen
relative to that administration.
Practical considerations arise with inpatient drug
therapeutic monitoring also. The most notable prob-
lem in this setting is spuriously high marker values
that result from taking the monitoring specimen
through the intravenous catheter being used for drug
administration (Murphy 1995). The physician must
always be vigilant for this and other potential
sources of measurement error. The need for such
vigilance is nicely demonstrated in the following
vignette related by a clinical pharmacologist/labora-
torian (Kumor 1985).
I recall another afternoon, . . . nice Dr. L.P.
was on the phone with me. He was very
upset. He was really angry. It seemed that
my laboratory did not care about patient care
and that nobody in the lab was doing his job.
He was taking care of a child recovering from
Haemophilus influenzae
meningitis. He had
sent a sample for chloramphenicol determina-
tion and needed the result desperately because
the previous levels had been low and the dose
had been increased. The determination had
not been done in the 3 days since sending the
specimen, and as it was Friday, it would not
be performed until Monday.
I was at a loss because the laboratory should
get the results out much quicker than that. I
apologized to Dr. L.P., normally a nice guy.
I wandered into the laboratory and asked some
casual questions, like “What the heck is going
on with the chloro levels!” (Reader, please
note: I was in error in being disrespectful to
the technologists, but I must tell the story the
way it occurred). The technologists explained
the problem. The assay requires an enzyme
and our batch was bad. The assay had been
attempted on the Wednesday run, but the
controls were out of acceptable limits. There
was a problem again on Thursday. Friday
they thawed some stored enzyme, but it didn’t
work either. We eventually decided to send
the specimen to a reference laboratory.
Drug Therapy
12-14
