and acquired factors, notably disease of the organs responsible
for drug metabolism and excretion. Pharmacokinetic modelling
is crucial in drug development to plan a rational therapeutic
regime, and understanding pharmacokinetics is also import-
ant for prescribers individualizing therapy for a particular
patient. Pharmacokinetic principles are described in Chapter 3
from the point of view of the prescriber. Genetic influences on
pharmacodynamics and pharmacokinetics (pharmacogenet-
ics) are discussed in Chapter 14 and effects of disease are
addressed in Chapter 7, and the use of drugs in pregnancy
and at extremes of age is discussed in Chapters 9–11.
There are no good animal models of many important human
diseases. The only way to ensure that a drug with promising
pharmacological actions is effective in treating or preventing
disease is to perform a specific kind of human experiment,
called a clinical trial. Prescribing doctors must understand the
strengths and limitations of such trials, the principles of which
are described in Chapter 15, if they are to evaluate the litera-
ture on drugs introduced during their professional lifetimes.
Ignorance leaves the physician at the mercy of sources of infor-
mation that are biased by commercial interests. Sources of
unbiased drug information include Dollery’s encyclopaedic
Therapeutic drugs, 2nd edn (published by Churchill Livingstone
in 1999), which is an invaluable source of reference. Publications
such as the Adverse Reaction Bulletin, Prescribers Journaland
the succinctly argued Drug and Therapeutics Bulletinprovide
up-to-date discussions of therapeutic issues of current
importance.
FURTHER READING
Dukes MNG, Swartz B. Responsibility for drug-induced injury.
Amsterdam: Elsevier, 1988.
Weatherall DJ. Scientific medicine and the art of healing. In: Warrell
DA, Cox TM, Firth JD, Benz EJ (eds), Oxford textbook of medicine,4th
edn. Oxford: Oxford University Press, 2005.SCIENTIFICBASIS OFUSE OFDRUGS INHUMANS 5Key points- Drugs are prescribed by physicians of all specialties.
- This carries risks as well as benefits.
- Therapy is optimized by combining general knowledge
of drugs with knowledge of an individual patient. - Evidence of efficacy is based on clinical trials.
- Adverse drug effects may be seen in clinical trials, but
the drug side effect profile becomes clearer only when
widely prescribed. - Rational prescribing is encouraged by local formularies.
Case history
A general practitioner reviews the medication of an
86-year-old woman with hypertension and multi-infarct
dementia, who is living in a nursing home. Her family used
to visit daily, but she no longer recognizes them, and needs
help with dressing, washing and feeding. Drugs include
bendroflumethiazide, atenolol, atorvastatin, aspirin, haloperi-
dol,imipramine, lactulose and senna. On examination, she
smells of urine and has several bruises on her head, but
otherwise seems well cared for. She is calm, but looks pale
and bewildered, and has a pulse of 48 beats/min regular,
and blood pressure 162/96 mmHg lying and 122/76 mmHg
standing, during which she becomes sweaty and distressed.
Her rectum is loaded with hard stool. Imipramine was started
three years previously. Urine culture showed only a light
mixed growth. All of the medications were stopped and
manual evacuation of faeces performed. Stool was nega-
tive for occult blood and the full blood count was normal.
Two weeks later, the patient was brighter and more mobile.
She remained incontinent of urine at night, but no longer
during the day, her heart rate was 76 beats/min and her
blood pressure was 208/108 mmHg lying and standing.
Comment
It is seldom helpful to give drugs in order to prevent some-
thing that has already happened (in this case multi-infarct
dementia), and any benefit in preventing further ischaemic
events has to be balanced against the harm done by the
polypharmacy. In this case, drug-related problems probably
include postural hypotension (due to imipramine, ben-
droflumethiazide and haloperidol), reduced mobility (due to
haloperidol), constipation (due to imipramine and haloperi-
dol), urinary incontinence (worsened by bendroflumethi-
azide and drugs causing constipation) and bradycardia (due
to atenolol). Drug-induced torsades de pointes (a form of
ventricular tachycardia, see Chapter 32) is another issue.
Despite her pallor, the patient was not bleeding into the
gastro-intestinal tract, but aspirin could have caused this.