Blood gases indicate a mixed picture:
- pH is low, and HCO 3 is low, due to the direct
acidotic challenge of the toxin. - pCO 2 is low due to hyperventilation (Kuss-
maul respiration, and possibly also a direct
effect of salicylate).
–pO 2 is typically normal.
Excretion of salicylate is principally in the urine
and can be enhanced by alkalinization because:
- Salicylate is filtered in the glomerulus and this
is pH independent. - Resorption of salicylate in the nephron is
substantial because of the concentration of
the toxin. - The nephron is a lipid membrane and in an
alkaline environment more salicylate is
ionized.
Lastly, remember that basic drugs are excreted
more vigorously in a urine that is acidified with
oral ammonium sulfate. A common and popular
drug of abuse at the moment in California is me-
thamphetamine; the suffix again tells us it is a basic
drug, and the same logic applies, but in reverse.
28.3 When is plasma
concentration monitoring
irrational?Drugs for which concentration assays are clearly
unsuited include acute therapies (i.e. not used
at steady state), those with extraordinarily short
half-times (e.g. injected or intranasal polypep-
tides) and those for which either treatment is
indicated regardless (late acetaminophen/parace-
tamol overdoses, see above), or when adverse
events are almost automatic and should be mon-
itored in other ways, for example CNS toxicity
with salicylates (see above) or liability to bone
marrow suppression with cytotoxic agents.
Furthermore, the efficacy and tolerability of
some drugs are known to be unrelated to circulat-
ing concentrations (e.g. penicillin anaphylaxis),
which also makes plasma concentration monitor-
ing pointless.
It is also important to remember that just
because a clinical laboratory offers a plasma
level for a particular drug, this does not mean
that it will be universally useful. For example,
plasma levels of antiviral drugs used in the treat-
ment of HIV infection are now inferior to CD4
lymphocyte counts and RNA measures of viral
load when monitoring for efficacy (e.g. Back
et al., 2000). Another good example is the case
of many antidepressant drugs. If the patient is also
using a potentially interacting, concomitant ther-
apy, if there is doubt as to treatment compliance or
if there is some other special clinical feature, then
a plasma level of, say, amitriptyline can be very
useful. A plasma level might also be very useful in
the emergency department for the diagnosis of an
intoxicated patient. But studying lower concentra-
tions of amitriptyline when treating depression is
of almost no practical value in predicting efficacy
(e.g. Ursolak, 1989). Although much is known
about the pharmacokinetic interactions of selec-
tive serotonin reuptake inhibitors (and indeed
some are even used as probes in clinical pharma-
cology studies), this does not extrapolate to their
routine plasma level monitoring in the clinic
(Sprouleet al., 1997). Thus, routine venesection
of patients in psychiatric clinics is likely to have a
relatively low yield of useful information and may
not be cost-effective.28.4 Concentration monitoring
in other biological fluidsUrine
Urine is commonly screened for evidence of illicit
drug use or alcohol consumption. This may be
viewed as a drug concentration monitoring proce-
dure, even if only of a qualitative type. It remains
controversial whether poppy-seed bagels can lead
to positive urine screens for opioids!
Materials such as radioactive sodium iothala-
mate or inulin can be regarded as drugs. These378 CH28 MONITORING DRUG CONCENTRATIONS IN CLINICAL PRACTICE