A Textbook of Clinical Pharmacology and Therapeutics

harm (Figure 13.2), and where special caution is required with
concurrent therapy. These include:

• warfarin and other anticoagulants;


• anticonvulsants;


• cytotoxic drugs;


• drugs for HIV/AIDS;


• immunosuppressants;


• digoxin and other anti-dysrhythmic drugs;


• oral hypoglycaemic agents;


• xanthine alkaloids (e.g. theophylline);


• monoamine oxidase inhibitors.

The frequency and consequences of an adverse interaction
when two drugs are used together are seldom known pre-
cisely. Every individual has a peculiar set of characteristics
that determine their response to therapy.

RISK OF ADVERSE DRUG INTERACTIONS

In the Boston Collaborative Drug Surveillance Program, 234 of
3600 (about 7%) adverse drug reactions in acute-care hospitals
were identified as being due to drug interactions. In a smaller
study in a chronic-care setting, the prevalence of adverse
interactions was much higher (22%), probably because of
the more frequent use of multiple drugs in elderly patients
with multiple pathologies. The same problems exist for the
detection of adverse drug interactions as for adverse drug
reactions (Chapter 12). The frequency of such interactions will
be underestimated by attribution of poor therapeutic outcome
to an underlying disease. For example, graft rejection follow-
ing renal transplantation is not uncommon. Historically, it
took several years for nephrologists to appreciate that epilep-
tic patients suffered much greater rejection rates than did non-
epileptic subjects. These adverse events proved to be due to an
interaction between anticonvulsant medication and immuno-
suppressant cortico-steroid therapy, which was rendered inef-
fective because of increased drug metabolism. In future, a
better understanding of the potential mechanisms of such
interactions should lead to their prediction and prevention by
study in early-phase drug evaluation.

SEVERITY OF ADVERSE DRUG INTERACTIONS

Adverse drug interactions are diverse, including unwanted
pregnancy (from failure of the contraceptive pill due to con-
comitant medication), hypertensive stroke (from hypertensive
crisis in patients on monoamine oxidase inhibitors), gastro-
intestinal or cerebral haemorrhage (in patients receiving war-
farin), cardiac arrhythmias (e.g. secondary to interactions
leading to electrolyte disturbance or prolongation of the QTc)
and blood dyscrasias (e.g. from interactions between allopuri-
nolandazathioprine). Adverse interactions can be severe. In
one study, nine of 27 fatal drug reactions were caused by drug
interactions.

ADVERSE INTERACTIONS GROUPED BY

MECHANISM

PHARMACEUTICAL INTERACTIONS

Inactivation can occur when drugs (e.g. heparinwithgentam-

icin) are mixed. Examples are listed in Table 13.1. Drugs may

also interact in the lumen of the gut (e.g. tetracyclinewith

iron, and colestyraminewithdigoxin).

PHARMACODYNAMIC INTERACTIONS

These are common. Most have a simple mechanism consisting

of summation or opposition of the effects of drugs with,

respectively, similar or opposing actions. Since this type of

interaction depends broadly on the effect of a drug, rather

than on its specific chemical structure, such interactions are

non-specific. Drowsiness caused by an H 1 -blocking antihista-

mine and by alcohol provides an example. It occurs to a

greater or lesser degree with all H 1 -blockers irrespective of the

chemical structure of the particular drug used. Patients must

be warned of the dangers of consuming alcohol concurrently

when such antihistamines are prescribed, especially if they

drive or operate machinery. Non-steroidal anti-inflammatory

agents and antihypertensive drugs provide another clinically

important example. Antihypertensive drugs are rendered less

effective by concurrent use of non-steroidal anti-inflammatory

drugs, irrespective of the chemical group to which they

belong, because of inhibition of biosynthesis of vasodilator

prostaglandins in the kidney (Chapter 26).

74 DRUG INTERACTIONS

Key points

• Drug interactions may be clinically useful, trivial or
  adverse.


• Useful interactions include those that enable efficacy to
  be maximized, such as the addition of an angiotensin
  converting enzyme inhibitor to a thiazide diuretic in a
  patient with hypertension inadequately controlled on
  diuretic alone (see Chapter 28). They may also enable
  toxic effects to be minimized, as in the use of
  pyridoxine to prevent neuropathy in malnourished
  patients treated with isoniazid for tuberculosis, and
  may prevent the emergence of resistant organisms
  (e.g. multi-drug regimens for treating tuberculosis, see
  Chapter 44).


• Many interactions that occur in vitro (e.g. competition
  for albumin) are unimportant in vivo because
  displacement of drug from binding sites leads to
  increased elimination by metabolism or excretion and
  hence to a new steady state where the total
  concentration of displaced drug in plasma is reduced,
  but the concentration of active, free (unbound) drug is
  the same as before the interacting drug was
  introduced. Interactions involving drugs with a wide
  safety margin (e.g. penicillin) are also seldom clinically
  important.


• Adverse drug interactions are not uncommon, and can
  have profound consequences, including death from
  hyperkalaemia and other causes of cardiac dysrhythmia,
  unwanted pregnancy, transplanted organ rejection, etc.