Diazepam
A white, almost white, or yellow, almost odourless, crystal-
line powder. solubilities are: very slightly soluble in
water and soluble in alcohol. USP solubilities are: soluble I
in 333 of water. I in 16 of alcohol, I in 2 of chloroform, and
I in 39 of ether.
Store in airtight containers. Protect from light.
Dilution. Great care should be observed on dilution of di-
azepam injections for administration by infusion because of
problems of precipitation. The manufacturer's directions
should be followed regarding diluent and concentration of di-
azepam and all solutions should be freshly prepared.
Incompatibility. Incompatibility has been reported between
diazepam and several other drugs. The manufacturers of di-
azepam injection advise against its admixture with other
drugs.
Sorption. Substantial adsorption of diazepam onto some
plastics may cause problems when administering the drug by
continuous intravenous infusion. More than 50% of diazepam
in solution may be adsorbed onto the walls of polyvinyl chlo-
ride infusion bags and their use should, therefore, be avoided.
Administration sets should contain the minimum amount of
polyvinyl chloride tubing and should not contain a cellulose
propionate volume-control chamber. Suitable materials for
infusion containers, syringes, and administration sets when
administering diazepam include glass, polyolefin. polypro-
pylene, and polyethylene.
Dependence and Withdrawal
The development of dependence is common after
regular use of diazepam or other benzodiazepines.
even in therapeutic doses for short periods. Depend-
ence is particularly likely in patients with a history
of alcohol or drug abuse and in patients with marked
personality disorders. Benzodiazepines should not
therefore be discontinued abruptly after regular use
for even a few weeks, but should be withdrawn by
gradual reduction of the dose over a period of weeks
or months. The extent to which tolerance occurs has
been debated but appears to involve psychomotor
performance more often than anxiolytic effects.
Drug-seeking behaviour is uncommon with thera-
peutic doses of benzodiazepines. High doses of di-
azepam and other benzodiazepines, injected
intravenously, have been abused for their euphoriant
effects.
Bmzodiazepine withdrawal syndrome. Development
of dependence to benzodiazepines cannot be predicted but
risk factors include high dosage, regular continuous use, the
use of benzodiazepines with a short half-life, use in patients
with dependent personality characteristics or a history of drug
or alcohol dependence, and the development of tolerance.
The mechanism of benzodiazepine dependence is unclear.
One possible mechanism is a relative deficiency of functional
aminobutyric acid (GABA) activity resulting from down-reg-
ulation of GABA receptors.
Symptoms of benzodiazepine withdrawal include anxiety, de-
pression, impaired concentration, insomnia, headache, dizzi-
ness, tinnitus, loss of appetite, tremor, perspiration.
irritability, perceptual disturbances such as hypersensitivity to visual and auditory stimuli and abnormal taste, nausea, vom-
iting, abdominal cramps, palpitations, mild systolic hyperten-
sion. tachycardia, and orthostatic hypotension. Rare and more
serious symptoms include muscle twitching, confusional or
paranoid psychosis, convulsions, hallucinations, and a slate
resembling delirium tremens.
Symptoms typical of withdrawal have been observed despite
Continued use of benzodiazepines and have been attributed ei-
ther to the development of tolerance or. as in the case of very
short-acting agents such as triazolam. to rapid benzodi-
azepine elimination. Pseudowithdrawal has been reported in
patients who believed incorrectly that their dose of benzodi-
azepine was being reduced Benzodiazepine withdrawal can
theoretically be distinguished from these reactions and from
rebound phenomena (return of original symptoms at greater
than pretreatment severity) by the differing time course. A
withdrawal syndrome is characterised by its onset, by the de-
velopment of new symptoms, and by a peak in intensity fol-
lowed by resolution. Onset of withdrawal symptoms depends
on the half-life of the drug and its active metabolites. Symp-
toms can begin within a few hours after withdrawal of a short-
acting benzodiazepine, but may not develop for up to 3 weeks
after stopping a longer-acting benzodiazepine. Resolution of
symptoms may take several days or months. The dependence
induced by short- and long-acting benzodiazepines appears to
be qualitatively similar although withdrawal symptoms may
be more severe with short-acting benzodiazepines.
With increased awareness of the problems of benzodiazepine
dependence, emphasis has been placed on prevention by
proper use and careful patient selection. Withdrawal from
long-term benzodiazepine use should generally be encour-
aged. Since abrupt withdrawal of benzodiazepines may result
in severe withdrawal symptoms dosage should be tapered.
The patient should have professional and family support and
behavioral therapy may be helpful. Withdrawal in a special-
ist centre may be required for some patients. There are no
comparative studies of the efficacy of various withdrawal
schedules and in practice the protocol should be titrated
against the response of the patient. Clinicians often favour
transferring the patient to an equivalent dose of diazepam giv-
en at night. Time required for withdrawal can vary from about
4 weeks to a year or longer. The daily dosage of diazepam. for
example, can be reduced in steps of 0.5 to 2.5 mg at fortnight-
ly intervals. If troublesome abstinence effects occur the dose
should be held level for a longer period before further reduc-
tion. increased dosage should be avoided if possible. It is bet-
ter to reduce too slowly than too quickly. In many cases the
rate of withdrawal is best decided by the patient. The follow-
ing approximate dosage equivalents to diazepam 5 mg have
been recommended in the UK to help in a withdrawal pro-
gramme that involves substituting an equivalent dose of di-
azepam: chlordiazepoxide 15 mg, loprazolam 0.5 to I mg.
lorazepam 500 mcg. lormetazepam 0.5 to I mg, nitrazepam
5 mg, oxazepam 15 mg, and temazepam 10 mg.
Adjuvant therapy should generally be avoided although a beta
blocker may be given for prominent sympathetic over activity
and an antidepressant for clinical depression. Antipsycholic
drugs may aggravate symptoms.
Symptoms gradually improve after withdrawal but postwith-
drawal syndromes lasting for several weeks or months have
been described. Continued support may be required for the
first year after withdrawal to prevent relapse.
Adverse Effects
Drowsiness. sedation, and ataxia are the most fre-
quent adverse effects of diazepam use. They gener-
ally decrease on continued administration and are a
consequence of CNS depression. Less frequent ef-
fects include vertigo, headache. confusion, mental
depression (but see Effects on Mental Function, be-
low). slurred speech or dysarthria, changes in libido,
tremor, visual disturbances, urinary retention or in-
continence, gastro-intestinal disturbances, changes
in salivation, and amnesia. Some patients may expe-
rience a paradoxical excitation which may lead to
hostility, aggression, and disinhibition. Jaundice,
blood disorders, and hypersensitivity reactions have
been reported rarely. Respiratory depression and hy-
potension occasionally occur with high dosage and
parenteral administration.
Pain and thrombophlebitis may occur with some in-
travenous formulations of diazepam.
Overdosage can produce CNS depression and coma
or paradoxical excitation. However, fatalities are
rare when taken alone.
Use of diazepam in the first trimester of pregnancy
has occasionally been associated with congenital
malformations in the infant but no clear relationship
has been established. This topic is reviewed undei
Pregnancy and Breast Feeding, below. Administra-
tion of diazepam in late pregnancy has been associ-
ated with intoxication of the neonate.
In a report of 2 patients who developed rhabdomyolysis sec-
ondary to hyponatraemia it was suggested that the use of ben-
zoidiazepines might have contributed to the rhabdomyolysis.
These 2 patients brought the number of reported cases of
rhabdomyolysis associated with hyponatraemia to 8, and of
these, 5 had received benzodiazepines.
Carcinogenicity. In an evaluation of the carcinogenic po-
tential of several benzodiazepines the International Agency
for Research on Cancer concluded that there was sufficient
evidence from human studies that diazepam did not produce
breast cancer and that there was inadequate data to support its potential carcinogenicity at other sites. For most other benzodiazepines the lack of human studies meant that the carcinogenic risk to humans was not classifiable. However, there appeared to be sufficient evidence of carcinogenicity in animal studies for oxazepam to be classified as possibly carcinogenic in humans. There was also limited evidence of
carcinogenicity in animal studies for doxefazepam.
Effects on body temperature. Studies in healthy
subjects indicate that benzodiazepines can reduce body
temperature. After a single dose of diazepam 10 mg by mouth
in 11 subjects body temperature on exposure to cold fell to a
mean of 36.93°, compared with 37.08° on exposure without
the drug. An 86-year-old woman developed hypothermia af-
ter administration of nitrazepam 5 mg. After recovery she
was mistakenly given another 5-mg dose of nitrazepam and
again developed hypothermia.
Hypothermia has been reported in the neonates of mothers
given benzodiazepines during the late stages of pregnancy.
Effects on endocrine function. Galactorrhoea with nor-
mal serum-prolactin concentrations has been noted in 4 wom-
en taking benzodiazepines. Gynaecomastia has been
reported in one man taking up to 140 mg diazepam daily and
in 5 men taking diazepam in doses of up to 30 mg daily.'
Again, there was no change in prolactin concentrations but
serum-oestradiol concentrations were raised in the group of 5
men.
Raised plasma-testosterone concentrations were observed in
men taking diazepam 10 to 20 mg daily for 2 weeks.
Effects on the eyes. Brown opacification of the lens oc-
curred in a married couple who took diazepam 5 mg or more
daily by mouth over several years.
Effects on the liver. Reports of cholestatic jaundice in one
patient and focal hepatic necrosis with intracellular
cholestasis in another associated with the administration of
diazepam.
Effects on mental function. The effects of benzodi-
azepines on psychomotor performance accumulated from
laboratory tests have been reviewed in detail by Woods el al.
The results of such tests are not easily extrapolated to the clinical situation.
Concern has been expressed as lo the possible effects of long-
term benzodiazepine use on the brain. Of the many studies of
possible impairment of psychological function few have clin-
ical significance. Golombok et al - have carried out a detailed
study using a variety of neuropsychological tests on 50 chron-
ic benzodiazepine users. 34 chronic users who had withdrawn
from medication for at least 6 months, and 61 control sub-
jects. They found that patients taking high doses of benzodi-
azepines for long periods of lime perform poorly on tasks
involving visual-spatial ability and sustained attention. Then
was no evidence of impairment in global measures of intellec
tual functioning such as memory, flexibility, and simple reac
tion time. The authors could draw no conclusions as to the
effect of benzodiazepine withdrawal on these changes. A few
studies have examined the effect of benzodiazepine use on
brain morphology. A study of 17 long-term users indicated a
dose-dependent increase in ventricle size.
Sexual fantasies have been reported in women sedated with
intravenous diazepam or midazolam. These appear to be
dose-related.
Mental depression is generally considered to be one of the
less frequent adverse effects of benzodiazepines. However
Patten and Love consider that there is no good support for the
view that benzodiazepines can cause depression.
Adverse effects of alprazolam on behaviour have been re-
viewed by Cole and Kaiido.
Effects on the nervous system. There are a few isolated
reports of extra pyramidal symptoms in patients taking benzo-
diazepines. Benzodiazepines have been used with limited
success to treat such symptoms induced by antipsychotics.
ENCEPHALOPATHY. Prolonged use of midazolam with fentanyl
has been associated with encephalopathy in infants sedated
under intensive care.
Effects on sexual function. Benzodiazepines have no di-
rect effects on erection or ejaculation but their sedative effect
may reduce sexual arousal and lead to impotence in some pa-
tients. Conversely sexual performance may be improved
where it is impaired by anxiety. Increased libido and orgas-
mic function has been reported in 2 women after withdrawal
of long-term benzodiazepine us.
Effects on the skin. Analysis by the Boston Collaborative
Drug Surveillance Program of data on 15438 patients hospi-
talised between 1975 and 1982 detected 2 allergic skin reac-
tions attributed to diazepam among 4707 recipients of the
drug. A reaction rate of 0.4 per 1000 recipients was calculat-
ed from these figures.
Hypersensitivity, Hypersensitivity reactions including ana-
phylaxis are very rare following administration of diazepam
Reactions have been attributed to the polyethoxylated castor
oil vehicle used for some parenteral formulations.
There is also a report of a type I hypersensitivity reaction to
lipid emulsion formulation of diazepam.
Local reactions: Ischaemia and gangrene have been report
ed after accidental intra-arterial injection of diazepam.
Clinical signs may not occur until several days after the even
Pain and thrombophlebitis after intravenous administration
may be similarly delayed. Local reactions after intravenous
injection have been attributed to the vehicle, and have bee
observed more often when diazepam is given as a solution in
propylene glycol than in polyethoxylated castor oil. A
emulsion of diazepam in Soya oil and water has been associ
ated with a lower incidence of local reactions' Pain and phle
bitis may also be caused by precipitation of diazepam at the
site of infusion Arterial spasm experienced by one patient
given diazepam intravenously was probably due to pressure
from a cuff on the arm being inflated causing extravasation of
diazepam out of the vein and into the radial artry.
Local irritation has also been observed after rectal administration of diazepam.
Overdosage. Impairment of consciousness is fairly rapid in
poisoning by benzodiazepines. Deep coma or other manifes-
tations of severe depression of brainstem vital functions are
rare. more common is a sleep-like state from which the pa-
tient can be temporarily raised by appropriate stimuli. There
is usually little or no respiratory depression, and cardiac rate
and rhythm remain normal in the absence of anoxia or severe
hypotension. Since tolerance to benzodiazepines develops
rapidly consciousness is often regained while concentrations
of drug in the blood are higher than those which induced co-
ma. Anxiety and insomnia can occur during recovery from
acute Overdosage while a full-blown withdrawal syndrome.
possibly with major convulsions, can occur in patients who
have previously been chronic users.
During the years 1980 to 1989, 1576 fatal poisonings in Brit-
ain were attributed to benzodiazepines. Of these, 891 were
linked to overdosage with benzodiazepines alone and another
591 to overdosage combined with alcohol. A comparison of
these mortality statistics with prescribing data for the same
period, to calculate a toxicity index of deaths per million pre-
scriptions. suggested that (here were differences between the
relative toxicities of individual benzodiazepines in overdos-
age. A later study of another 303 cases of benzodiazepine
poisonings supported these findings of differences in toxicity
as well as pointing to the relative safety of the benzodi-
azepines in overdosage.
Treatment of Adverse Effects
Following recent ingestion of an overdose of di-
azepam the stomach may be emptied by lavage.
Treatment is generally symptomatic and supportive.
The specific benzodiazepine antagonist, flumazenil,
may be used in the differential diagnosis of unclear
cases of overdose but expert advice is essential since
serious adverse effects may occur in patients de-
pendent on benzodiazepines.
Precautions
Diazepam should be avoided in patients with pre-
existing C~JS depression or coma, acute pulmonary
insufficiency, or sleep apnoea, and used with care in
those with chronic pulmonary insufficiency. Di-
azepam should be given with care to elderly or de
bilitated patients who may be more prone to adverse
effects. Caution is required in patients with muscle
weakness, or impaired liver or kidney function. The
sedative effects of diazepam are most marked during
the first few days of administration; affected patient
should not drive or operate machinery. Monitoring
of cardio respiratory function is generally recom
mended when benzodiazepines are used for deep se-
dation.
Diazepam is not appropriate for the treatment of
chronic psychosis or for phobic or obsessional
states. Diazepam-induced disinhibition may preci
pitate suicide or aggressive behaviour and it should
not. therefore, be used alone to treat depression
anxiety associated with depression: it should also be
used with care in patients with personality disorder
Caution is required in patients with organic brain
changes particularly arteriosclerosis. In cases of be
reavement. psychological adjustment may be inhib
ited by diazepam.
Many manufacturers of diazepam and other benzo-
diazepines advise against their use in patients with
glaucoma, but the rationale for this contra-indica-
tion is unclear.
For warnings on benzodiazepines during pregnancy
and breast feeding, see below.
Dependence characterised by a withdrawal syn-
drome may develop after regular use of diazepam,
even in therapeutic doses for short periods (see
above); because of its dependence liability, di-
azepam should be used with caution in patients with
a history of alcohol or drug addiction.
Since hypotension and apnoea may occur when di-
azepam is given intravenously it has been recom-
mended that this route should only be used when
facilities for reversing respiratory depression with
mechanical ventilation are available. Patients should
remain supine and under medical supervision for at
least one hour'after intravenous injection. Intrave-
nous infusion should only be undertaken in special-
ist centres with Intensive care facilities where close
and constant supervision can be undertaken.
Administration. INTRAVENOUS. A warning' that prolonged
use of high-dose intravenous infusions of diazepam prepara-
tions containing benzyl alcohol can result in benzyl alcohol
poisoning.
Cardiovascular disorders. See under Respiratory System
Disorders, below.
Driving. Experimental studies of simulated driving and actu-
al driving behaviour have indicated that single or repeated therapeutic doses of most benzodiazepines tested may adversely affect parameters of performance in healthy subjects.
Epidemiological studies have not provided a clear indication
as to whether, or to what extent, benzodiazepines contribute
to the risk of driving accidents. However, recent studies
have found an increased risk. A large case-control cohort
rtudy2 in elderly drivers suggested that the risk of accidents
was increased in those who took longer-acting benzodi-
azepines. Another study suggested that, for users of benzo-
diazepines or zopiclone as a group, the increased risk was
greatest in younger drivers and was increased by alcohol con-
sumption. Patients affected by drowsiness while taking ben-
zodiazepines should not drive or operate machinery.
Drowsiness often becomes less troublesome with continued
use of these drugs.
The elderly. As for many other drugs, old age may alter the
distribution, elimination, and clearance of benzodi-
azepines. Metabolic clearance appears to be reduced in the
elderly for benzodiazepines metabolised principally by oxi-
dation but not for those biotransformed by glucuronide con-
jugation or nitro reduction. Prolonged half-life in the elderly may be a result of this decrease in clearance or of an increase in the volume of distribution. The clinical consequence of these changes depends on factors such as dosage schedule and extent of first-pass extraction by the liver.
Irrespective of pharmacokinetic changes, old people may ex-
hibit increased sensitivity to acute doses of benzodi-
azepines, Impairment of memory, cognitive function, and
psychomotor performance and behaviour disinhibition may
be more common than with younger patients.* Long-term use
commonly exacerbates underlying dementia in elderly pa-
tients.*
The upshot of the pharmacokinetic and pharmacodynamic
changes of benzodiazepines in the elderly is that adverse ef
fects may be more frequent in these patients and lower doses
are commonly required. An epidemiological study of persons
65 years and older found an increased rate of hip fracture
among current users of long-acting benzodiazepines (chlo-
rdiazepoxide, clorazepate, diazepam, and flurazepam) but not
among users of short-acting drugs (alprazolam, bromazepam,
lorazepam, oxazepam, and triazolam). However, preliminary
results of a later meta-analysis of studies in the literature
found no association between the use of benzodiazepines or
other sedatives or hypnotics and falls or fractures in the elderly. Nonetheless, if administration of a benzodiazepine is
considered necessary in elderly patients, a short-acting drug is to be preferred. It should also be remembered that old people are at increased risk of sleep-related breathing disorders, such sleep apnoea and the use of hypnotics such as benzodiazepines should be avoided in these patients (see Respiratory System Disorders, below).
Gout. A 40-year-old Chinese man with a history of gout had
acute attacks on 5 occasions after taking nitrazepam 10 mg.
attacks were also precipitated by diazepam 5 mg. chlo-
rdiazepoxide 15 mg. and by methaqualone with diphenhy-
dramine.
Hepatic impairment. Studies of the pharmacokinetics of
diazepam in patients with liver disorders suggest that oxida
tive metabolism is impaired resulting in a prolonged half-life
and reduced clearance. A reduction in dosage was general-
ly required.
High-altitude disorders. Sleep may be impaired at high
altitude due lo frequent arousals associated with pronounced
oxygen desaturation and periodic breathing. Traditional ad-
vice has been that sedatives should not be given at high alti-
tude. It has been argued that since diazepam. and possibly
other sedatives blunt the hypoxic ventilatory response, sleep
hypoxaemia might be exacerbated. A recent small study has
suggested that small doses of a short-acting benzodiazepine,
such as temazepam might actually improve the subjective
quality of sleep and reduce episodes of arterial desaturation
without changing mean oxygen saturation. However the pos-
sibility of an interaction between acetazolamide taken for
prophylaxis or treatment of acute mountain sickness and the
benzodiazepine should be borne in mind: ventilatory depres-
sion in a mountain climber with acute mountain sickness was
considered lo he due to the potentiation of triazolam by aceta-
zolamide.
Nervous system disorders. Benzodiazepines can reduce
cerebral perfusion pressure and blood oxygenation and may
produce irreversible neurological damage in patients with
head injuries and should therefore be administered with great
care. Their use should be avoided for the control of seizures
in patients with head injuries or other acute neurological le-
sions as these patients can be managed effectively with
phenytoin.
EPILEPSY : There have been rare reports of benzodiazepines pro-
ducing paradoxical exacerbation of seizures in patients with
epilepsy.
Porphyria. Diazepam was considered to be unsafe in pa-
tients with acute porphyria although there is conflicting evi-
dence of porphyrinogenicity. Intravenous diazepam had been
used successfully to control status epilepticus occurring after
the acute porphyric attack. For a discussion of the manage-
ment of seizures associated with acute porphyric attacks.
Pregnancy and breast feeding. Benzodiazepines have
been widely used in pregnant patients and the reports and
studies of such use have been reviewed. The early reports,
usually involving diazepam and chlordiazepoxide. of some
fetal malformations including facial clefts and cardiac abnor-
malities, were not supported by later reports. Use of benzodi-
azepines in the third trimester and during labour seems lo be
associated in some infants with the neonatal withdrawal
symptoms or the floppy infant syndrome. Also a small
number exposed in utero to benzodiazepines have shown
slow development in the early years but by 4 years of age
most had developed normally, and for those that had not it
was not possible to prove a cause-effect relationship with ben-
zodiazepine exposure. In a recent meta-analysis of benzodi-
azepine use during the first trimester of pregnancy pooled
data from cohort studies showed no apparent association be-
tween benzodiazepine use and the risk of major malforma-
tions or oral cleft alone. There was. however, a small but
significantly increased risk of oral cleft according to data
from case-control studies. Although benzodiazepines did not
appear lo be a major human teratogen use of ultrasonography
was advised lo rule out visible forms of cleft lip. The Com-
mittee on Safety of Medicines (CSM) in the UK has
recommended that if benzodiazepines are prescribed to
women of childbearing potential they should be advised to
contact the physician regarding discontinuation of the drug if
they intend to become, or suspect that they are, pregnant. In
the reviewer's opinion the limited distribution into breast
milk did not constitute a hazard to the breast-feeding infant
but the infant should be monitored for sedation and the inabil-
ity to suckle. It has been suggested that if a benzodiazepine
must be used during breast feeding it would be preferable to
use a short-acting drug with minimal distribution into breast
milk and inactive metabolite, oxazepam, lorazepam, alpra-
zolam, or midazolam might be suitable. However the Ameri-
can Academy of Pediatrics Committee considered that use by
nursing mothers for long periods was a cause for concern and
the CSM has recommended that benzodiazepines should not
be given lo lactating mothers.
Renal impairment. Although plasma protein binding of di-
azepam was reduced in 8 patients with advanced renal failure,
the clearance of unbound drug after a single intravenous dose
was not significantly changed.
Respiratory system disorders. Benzodiazepines may af-
fect the control of ventilation during sleep and may worsen
sleep apnoea or other sleep-related breathing disorders espe-
cially in patients with chronic obstructive pulmonary disease
or cardiac failure. Risk factors for sleep apnoea. which often
goes undiagnosed, include old age, obesity, male sex. post-
menopausal status in women, and a history of heavy snoring.
Although benzodiazepines may reduce sleep fragmentation.
their long-term use may result in conversion from partial to
complete obstructive sleep apnoea in heavy snorers or in short
repetitive central sleep apnoea in patients with recent myocar-
dial infarction.
Interactions
Enhanced sedation or respiratory and cardiovascular
depression may occur if diazepam or other benzodi-
azepines are given with other drugs that have CNS-
depressant properties: these include alcohol, antide-
pressants. antihistamines, antipsychotics. general
anaesthetics, other hypnotics or sedatives, and opio-
id analgesics. The sedative effect of benzodi-
azepines may also be enhanced by cisapride. Similar
effects may be produced by concomitant administra-
tion with drugs which interfere with the metabolism
of benzodiazepines. Drugs which have been report-
ed to alter the pharmacokinetics of benzodiazepines
are discussed in detail below but few of these inter-
actions are likely to be of clinical significance, Ben-
zodiazepines such as diazepam which are
melabolised primarily by hepatic microsomal oxi-
dation may be more susceptible to pharmacokinetic
changes than those eliminated primarily by glucuro-
nide conjugation.
Analgesics. The peak plasma concentration of oxazepam
was significantly decreased during treatment with diflunisal
in 6 healthy subjects, while the renal clearance of the glucuro-
nide metabolite was reduced and its mean elimination half-
life increased from 10 to 13 hours. Diflunisal also displaced
oxazepam from plasma protein binding sites in vitro. Aspirin
.shortened the time to induce anaesthesia with midazolam in
78 patients also possibly due to competition for plasma pro-
tein binding sites. Paracetamol produced no significant
change in plasma concentrations of diazepam or its major me-
tabolite and only marginal changes in urine concentrations in
4 healthy subjects.
Benzodiazepines such as diazepam, lorazepam, and mida-
zolam may be used with opioid analgesics in anaesthetic or
analgesic regimens. An additive sedative effect is to be
expected but there are also reports of severe respiratory de-
pression with midazolam and fentanyl or sudden hypoten-
sion with midazolam and fentanyl or sufentanil. The
clearance of midazolam appears to be reduced by fentanyl,
possibly as a result of competitive inhibition of metabolism
by the cytochrome P450 isozyme CYP3A. Careful monitor-
ing is therefore required during concomitant administration
of midazolam with these opioids and the dose of both drugs
may need to be reduced. Synergistic potentiation of the induc-
tion of anaesthesia has been reported between midazolam and
fmtanyl.9 but one study has suggested that midazolam can di-
minish the analgesic effects of sufentanil. Pretreatment with
morphine or pethidine has decreased the rate of oral absorp-
tion of diazepam. This has been attributed to the effect of opi
oid analgesics on gastro-intestinal motility.
Dextropropoxyphene prolonged the half-life and reduced the
clearance of alprazolam but not diazepam or lorazepam in
healthy subjects.
Antiarrhythmics. An interaction between clonazepam and
existing therapy with amiodarone was suspected in a 78-year-
old man who experienced symptoms of benzodiazepine tox-
icity 2 months after starting with clonazepam 0.5 mg given at
bedtime for restless leg syndrome: symptoms resolved on
withdrawal of clonazepam.
Antibacterials. Both erythromycin' and
triacetyloleandomycin have been reported to inhibit the he-
patic metabolism of triazolam in healthy subjects. Peak plas-
ma-triazolam concentrations were increased, half-life
prolonged, and clearance reduced. Triacetyloleandomycin
prolonged the psychomotor impairment and amnesia pro-
duced by triazolam. Loss of consciousness following eryth-
romycin infusion in a child premedicated with midazolam
was attributed to a similar interaction, and increases in peak
plasma concentrations of midazolam with profound and pro-
longed sedation have been reported following administration
of erythromycin Concomitant administration of midazolam
and erythromycin should be avoided or the dose of mida-
zolam reduced by 50 to 75%. Roxithromycin has also been
reported to have some effects on the pharmacokinetics and
pharmacodynamics of midazolam but it was considered that
these changes were probably not clinically relevant. However,
it was recommended that as a precaution the lowest possible
effective dose of midazolam should be used when given with
roxithromycin. In another study' azithromycin did not appear
to have any effect on the metabolism or psychomotor effects
of midazolam.
There is an isolated report of significant rises in steady-state blood-midazolam concentration coinciding with administra-
tion of ciprofloxacin. Also ciprofloxacin has been reported to
reduce diazepam clearance and prolong its terminal half-life.
although psychometric tests did not show any changes in di-
azepam's pharmacodynamics. However, ciprofloxacin ap-
pears to have no effect on the pharmacokinetics or
pharmacodynamics of temazepam.
Isoniazid has been reported to increase the half-life of a single
dose of diazepam and triazolam but not of oxazepam in
healthy subjects. In contrast, rifampicin has decreased the
half-life of diazepam" and midazolam" while ethambutul
has no effect on diazepam pharmacokinetics. In patients re-
ceiving therapy for tuberculosis with a combination of isoni-
azid, rifampicin. and ethambutol the half-life of a single
diazepam dose was shortened and its clearance increased.
Thus the enzyme-inducing effect of rifampicin appears to
predominate over the enzyme-inhibiting effect of Isoniazid.
Anticoagulants. Reduced plasma binding of diazepam and
desmethyldiazepam and increases in the free concentrations
without changes in the total blood or plasma concentrations
occurred immediately following heparin intravenously.
Benzodiazepines do not usually interact with oral anticoagu-
lants although there have been rare reports of altered antico-
agulant activity.
Antidepressants. It has been recommended that the dosage
of alprazolam should be reduced when given with fluvoxam-
ine as concomitant administration has resulted in doubling of
plasma-alprazolam concentrations. Since plasma concentra-
tions of bromazepam" and of diazepam also appear to be af-
fected by fluvoxamine it has been suggested that patients
taking fluvoxamine who require a benzodiazepine should
preferentially receive a benzodiazepine such as lorazepam
which has a different metabolic pathway. Small studies sug-
gest that fluoxetine can also increase plasma concentrations of
alprazolam. Fluoxetine appears to have a similar effect on
diazepam but plasma concentrations of diazepam's active me-
tabolite desmethyidiazepam are reduced and it is considered
that the overall effect is likely to be minor. The potential for
a clinically significant interaction with sertraline, paroxetine
or citalopram is considered to be less.
The manufacturers have reported that alprazolam may in-
crease the steady-state plasma concentralions of imipramine
and desipramine. although the clinical significance of such
changes is unknown. For a suggestion that benzodiazepines
may increase the oxidation of amineptine to a toxic metabo-
lite. see under Effects on the Liver in Adverse Effects of Am-
itriptyline.
Concomitant administration of nefazodone has been reported
to result in raised concentrations of alprazolam and triazolam.
increased sedation, and impairment of psychomotor perfonn-
ance. Nefazodone may inhibit the oxidative metabolism of
alprazolam and triazolam. No interaction was reported with
lorazepam which is primarily eliminated by conjunction.
There have been an isolated report of hypothermia after ad-
ministration of diazepam and lithium.
There have been occasional reports of sexual disinhibition in
patients taking tryptophan with benzodiazepines.
Antiepileptics : Carbamazepine, phenobarbitone, and
phenytoin are all inducers of hepatic drug-metabolising en-
zymes. Therefore, in patients receiving long-term therapy
with these drugs the metabolism of benzodiazepines may be
enhanced.
Sodium valproate has been reported to displace diazepam
from plasma-protein binding sites. Sporadic reports exist of
adverse effects when valproate is given with clonazepam"
with the development of drowsiness and. more seriously, ab-
sence status epilepticus, but the existence of an interaction is
considered to be unproven. Drowsiness has also been report-
ed when valproate was given with nitrazepam. Concomitant
administration of semisodium valproate and lorazepam has
resulted in raised concentrations of lorazepam due to inhibi-
tion of glucuronidation of lorazepam.
Antifungals. Both a single dose and multiple doses of keto-
conazole decreased the clearance of a single intravenous in-
jection of chlordiazepoxide.' Ketoconazole was considered to
inhibit at least one subset of the hepatic mixed-function oxi-
dase system. Studies have shown that ketoconazole and
itraconazole can produce marked pharmacokinetic interac-
tions with midazolam or triazolam and greatly increase the
intensity and duration of action of these benzodiazepines.
One study indicated that the risk of interaction persists for
several days after cessation of itraconazole therapy. It is rec-
ommended that the concomitant use of these antifungals and
benzodiazepines should be avoided or that the dose of the
benzodiazepine should be greatly reduced. One group of
workers' found that the area under the plasma concentration-
time curve for midazolam was increased by 15 times by keto-
conazole and by 10 times by itraconazole while peak plasma
concentrations of midazolam were increased fourfold and
threefold respectively. They also found' that the area under
the curve for triazolam was increased by 22 times by ketoco-
nazole and by 27 times by itraconazole: peak plasma concen-
trations of triazolam were increased about threefold by both
antifungals. A similar but less pronounced interaction occurs
between fiucona~ote and midazolam6 or triazolam:7 nonethe-
less the dosage of the benzodiazepine should be reduced dur-
ing concomitant use.
Antihistamines. A suggestion' that a reduction in
temazepam metabolism caused by diphenhydramine may
have contributed to perinatal death after ingestion of these
drugs by the mother.
Antivirals. Delavirdine and HIV-protease inhibitdrs such as
indinavir, nelfinavir, and ritonavir may inhibit the hepatic mi-
crosomal systems involved in the metabolism of some benzo-
diazepines. Concomitant administration requires monitoring
and dosage adjustments for the benzodiazepme or should be
avoided. Benzodiazepines which should not be used concom-
itantly with HIV-protease inhibitors include alprazolam.
clorazepate, diazepam, estazolam, flurazepam, midazolam.
and triazolam.
Beta blockers. A clear pattern of interactions between ben-
zodiazepines and beta blockers has not emerged. Propranolol
may inhibit the metabolism of diazepam and bromazepam,'
and metoprolol inhibit the metabolism of diazepam or
bromazepam to some extent although in many cases the ef-
fect on pharmacokinelics and pharmacodynamics is unlikely
to be of clinical significance. No significant phannacokinetic
interaction has been observed between propranolol and alpra-
zolam, lorazepam or oxazepam, although the rate of alpra-
zolam absorption may be decreased. Similarly no
pharmacokinetic interaction has been observed between aten-
olol and diazepam, labetalol and oxazepam or metoprolol
and lorazepam.
Calcium-channel blockers. Peak plasma concentrations
of midazolam were doubled and the elimination half-life of
midazolam prolonged when midazolam was given to healthy
subjects receiving diltiazem or verapamil. A similar interac-
tion has been demonstrated between diltiazem and tria-
zolam. Concomitant administration should be avoided or
the dose of these benzodiazepines reduced.
Clonidine. Anxiety was reduced and sedation was enhanced
when clonidine was given with flunitrazepam for premedica-
tion.'
Cyclosporin. In-vitro studies suggested that cyclosporin
could inhibit the metabolism of midazolam. However, blood-
cyclosporin concentrations in patients given cyclosporin to
prevent graft rejection were considered too low to result in
such an interaction.
Digoxin. For the effects of alprazolam and diazepam on dig-
oxin pharmacokinetics, see digoxin record.
Disulfiram. Evidence from healthy and alcoholic subjects
suggests that chronic treatment with disulfiram can inhibit the
metabolism of chlordiazepoxide and diazepam leading to a
prolonged half-life and reduced clearance: there was little ef-
fect on the disposition of oxazepam. ' No significant pharma-
cokinetic interaction was observed between disulfiram and
alprazolam in alcoholic patients' Temazepam toxicity, attrib-
uted to concomitant administration of disulfiram and
temazepam has been reported.
Gastro-intestinal drugs. Antacids have variable effects on
the absorption of benzodiazepines but any resulting inter-
action is unlikely to be of major clinical significance.
Several studies, usually involving single doses of diazepam
given to healthy subjects, have demonstrated that cimetidine
can inhibit the hepatic metabolism of diazepam. The clear-
ance of diazepam has generally been decreased and the half-
life prolonged. Some studies have also demonstrated im-
paired metabolic clearance of the major metabolite, desmeth-
yldiazepam (nordazepam). Cimetidine has also been reported
to inhibit the metabolism of other benzodiazepines (generally
those metabolised by oxidation including adinazolam, al-
prazolam, bromazepam, chlordiazepoxide, clobazam,
flurazepam, midazolam, nitrazepam, and
triazolam. Cimetidine does not appear to inhibit the he-
patic metabolism of lorazepam, oxazepam or
temazepam. The clinical significance of these interactions
between cimetidine and benzodiazepines remains dubious.
Studies which have assessed the effect of cimetidine on the
pharmacology as well as the pharmacokinetics of benzodi-
azepines have found little effect on cognitive function or de-
gree of sedation.
Most studies have failed to demonstrate an effect of ranitidine
on the hepatic metabolism of diazepam although Fee et
al. reported an increase in the bioavailability of a single
dose of midazolam given by mouth, and considered that an
effect on hepatic clearance was more likely than an effect on
absorption. The results of Fee et at. were consistent with
those of another study which demonstrated an enhanced sed-
ative effect of midazolam in patients pretreated with raniti
dine. Ranitidine has been reported to have no effect on the
pharmacokinetics of lorazepam or on the sedative effect of
temazepam- but has increased the bioavailability of tris-
zolam.
Famotidine or nizatidine do not appear to inhibit the he-
patic metabolism of diazepam.
Oral diazepam was absorbed more rapidly after the intrave
nous administration of metoclopramide· Enhanced motility
of the gastro-intestinal tract was implicated. Cisapride may
also accelerate the absorption of diazepam.
Studies of continuous omeprazole administration on the phar-
macokinetics of a single intravenous dose of diazepam in
healthy subjects indicate inhibition of diazepam metabolism
in a similar manner to cimetidine. Omeprazole decreases
the clearance and prolongs the elimination half-life of di-
azepam; in addition both the formation and elimination of
desmethyidiazepam appear to be decreased. The effects may
be greater in rapid than in slow metabolisers of omeprazole
and vary between ethnic groups. The clinical significance of
the interaction remains lo be established. Lansoprazole and
pantoprazole have been reported not to affect the pharma-
cokinetics of diazepam.
General anaesthetics : A synergistic interaction has been
demonstrated for the hypnotic effects of midazolam and lhio-
penronc. Although midazolam failed to produce anaesthesia
at the doses used. the drug caused a twofold increase in the
anaesthetic potency of thiopentone. Similar synergistic inter-
actions have been observed between midazolam and both
methohexitone and propofol The interaction between mi-
dazolam and propofol could not be explained solely by alter-
ation in free-plasma concentration of either drug.
Midazolam has also been reported to be able to produce a
marked reduction in the concentration of halothane required
for anaesthesia.
Kava. A patient whose medication included alprazolam, ci-
metidine and terazosin became lethargic and disoriented af-
ter starting to take kava. An interaction between kava and the
benzodiazepine was suspected.
Levodopa : Reversible deterioration of parkinsonism has
been reported in patients receiving levodopa who were given
benzodiazepines such as diazepam, nitrazepam or chlordiazepoxide. In one case parkinsonism symptoms resolved without alteration in
the medication.
Neuromuscular blocker. Conflicting reports are there including of potentiation,
Antagonism and of no interference of action of benzodiazepines.
Oral contraceptives. Some studies with alprazolam,'
chlordiazepoxide.' and diazepam' have supported sugges-
tions that oral contraceptives may inhibit the biotransforma-
tion of benzodiazepines metabolised by oxidation, although
no significant pharmacokinetic alterations have been ob-
served with clotiazepam, or triazolam. The biotransforma-
tion of benzodiazepines metabolised by conjugation, such as
lorazepam, oxazepam. or temazepam. may be enhanced or
unchanged. No consistent correlation has been observed be-
lween the above pharmacokinetic changes and clinical ef-
fects. Ellinwood et a1. observed that psychomotor
impairment due to oral diazepam was greater during the men-
strual pause than during the 21-daily oral contraceptive cycle.
This may have been due to an effect of oral contraceptives on
diazepam absorption. As part of their pharmacokinetic study.
Kroboth et al noted that women taking oral contraceptives
appeared to be more sensitive to psychomotor impairment
following single oral doses of alprazolam. lorazepam. or tria-
zolam than controls. The effects of temazepam were minimal
in both groups. Alterations in sedative or amnesic effect could
not be established with any certainty.
Penicillamine. Phlebitis associated with intravenous di-
azepam resolved with local heat but recurred on two separate
occasions after oral penicillamine.
Probenecid. Probenecid increased the half-life of intrave-
nous lorazepam in 9 healthy subjects' Probenecid was con-
sidered to impair glucuronide formation selectively and thus
the clearance of drug like lorazepam. Probenecid has also
shortened the time to induce anaesthesia with midazolam in
46 patients. The effect was considered to be due to competi-
tion for plasma protein binding sites. It has been suggested
that the initial dose of adinazolam should be reduced and sub-
sequent dosage monitored when it is administered with
probenecid. A study in 16 subjects had indicated that
probenecid could potentiate the psychomotor effects of adina
zolam. This appeared to be mainly due to inhibition of the
renal clearance of N-desmethyladinazolam. Probenecid has
also been reported lo reduce the clearance of nitrazepam but
not of temazepam.
Smooth muscle relaxants. Intracavernosal papaverine
produced prolonged erection in 2 patients who had been given
intravenous diazepam as an anxiolytic before the papaverine
injection.'
Tobacco smoking. The Boston Collaborative Drug Surveil-
lance Program reported drowsiness as a side-effect of di-
azepam or chlordiazepoxide less frequently in smokers than
non smokers. Pharmacokinetic studies have. however, been
divided between those indicating that smoking induces the
hepatic metabolism of benzodiazepines and those showing no
effect on benzodiazepine pharmacokinetics.? Hence, dimin-
ished end-organ responsiveness may in part account for the
observed clinical effects. Concomitant consumption of large
amounts of xanthine-containing beverages may decrease any
enzyme-inducing effects of smoking.
Xanthine. There are reports of aminophylline given intra-
venously reversing the sedation from intravenous di-
azepam, although not always completely nor as effectively
as flumazenil. Blockade of adenosine receptors by amino-
phylline has been postulated as the mechanism of this inter-
action.
Xanthine-containing beverages may be expected to decrease
the incidence of benzodiazepine-induced drowsiness because
of their CNS-stimulating effects and their ability to induce
hepatic drug-metabolising enzymes. However, decreased
drowsiness has only sometimes been observed and the actions
of xanthines may themselves be decreased by concomitant
heavy tobacco smoking.