1 Midazolam John Schou, M. D. Clinical Use of

1

Midazolam

Midazolam

John Schou, M.D.

Clinical Use of

Alix

2

John Schou, M.D.

County Hospital of Loerrach

Spitalstr. 25

D-79539 Loerrach, Germany

Alix Publishing

Wallbrunnstr. 106E · D-79539 Loerrach · Germany

© 2004

ISBN 3-928811-21-5

3

Midazolam

Clinical Use of

Midazolam

John Schou

4

Midazolam brought benzodiazepines to a prominent place in

anaesthesia and related disciplines. While generally used

according to previous standards, the partially unexpected features

of this drug also helped to define new clinical options. These

greatly enlarged the clinical field that could profit from

utilization of this unique drug.

Having had occasion to work with midazolam right from its

introduction in Germany in 1984, the author actively participated

not only in defining new standards for the use of this drug, but

also in coping with its adverse effects. Any useful drug poses

side-effects, and full comprehension of these are a precondition

for an optimal use. Although all drugs may be claimed to have

negative properties, these may often derive simply from incorrect

usage.

This booklet is not a strict review; it reflects the author’s

experiences which may occasionally appear to be in contrast to

some study results. This depends in part on, how the defined

questions were studied. An effort has been made to avoid obscure

comparisons and to focus instead on practical clinical aspects as

they have appeared in the author’s use of midazolam.

5

Midazolam

Contents

General pharmacology

6

Premedication of children

9

Premedication of adults

11

The paradoxical reaction

12

Sedation for regional anaesthesia

13

Sedation for diagnostic procedures

14

General anaesthesia

15

Intensive care

16

Use as an anticonvulsive

18

Emergency medicine

19

The benzodiazepine-antagonist

21

Synopsis: indications and dosages

22

Conclusion

25

References

27

6

General pharmacology

With its introduction into anaesthesia 20 years ago, midazolam displayed two unique

features, which remain relevant today: it is the shortest-acting and the only water-

soluble benzodiazepine-agonist [BZD] available. It engenders the general BZD effects,

causing anxiolysis, muscular relaxation and amnesia in lower doses and sedation or

hypnosis in higher ones. In addition, it has distinct anticonvulsive properties. This

may be interesting but it does not make clear what are the practical consequences

for the use of this drug.

Water- and lipid-soluble BZD-agonist

Year of synthesis

1976

Elimination half-life

1.5-2.4 h

Active metabolite (

α

-OH-midazolam)

10 %

-  metabolite elimination half-life

0.7-0.8 h

The drug has an elimination half-life of about 2 h (1.5-2.4). Only 10 % is transferred

to one sole active metabolite with lower potency and an even shorter half-life [1].

This is in contrast to the old BZD diazepam with 3 metabolites and a half-life of 48-

72 h. An even longer acting BZD, dipotassium-chlorazepate, has strangely enough

gained favour among other anaesthetists.

The water-solubility of midazolam allows for an unproblematic galenic preparation

for painless injection or mixture with common infusion solutions. It also explains the

fast resorption after IM injection. After 10 min, similar plasma levels are reached as

after IV injection of the same dose, without the latter’s high initial peak levels.

Bioavailibility of an IM dosage amounts to 90%. Crossing the blood-brain barrier

demands lipid solubility, also found in this drug.

Table 1. Key midazolam data

7

Midazolam

Midazolam is not an analgesic; still, it exerts an important effect on central nociception

(perception of pain) which, in clinical use, sometimes makes it difficult to tell the

difference to analgesia. The combination of midazolam with opioids leads to a useful,

but possibly also dangerous synergism - at least then dirigable through repeated

dosage of fractional amounts.

Adverse effects are similar to those of other BZD. Contrary to many BZDs, including

the two mentioned above, midazolam is considered not porphyrrhoigenic. A respiratory

depressant effect is not profound in lightly sedated patients. Strong sedation, in

particular in combination with opioids, may yield problematic respiratory depression

[2]. For the same reason, any BZD should not be used in patients with chronic respiratory

failure and possible hypercapnia.

A few years after its introduction, midazolam experienced an important safety

development with the introduction of the BZD-antagonist flumazenil. In 1977 a specific

BZD-receptor was demonstrated, which is localized in association to the GABA-receptor

[3,4]. BZDs thus augment the inhibitory GABAergic stimulation, which is an indirect

effect and only possible to a certain level. They do not exert any direct inhibitory

action.

Galenically, midazolam is distributed in aqueous solution void of any preservation

agents. It exists in ampoules of

í

 1 ml, 5 mg

(Multiple purposes)

í 

3 ml, 15 mg (see: Role in general anaesthesia)

í 

5 ml, 5 mg

(see: Sedation for regional anaesthesia and

diagnostic procedures)

í 

10 ml, 50 mg (see: Role in intensive care).

In addition, 7.5 mg tablets are available for oral premedication. Rectal and nasal

premedication of children is performed with the ampoule solution (5 mg/ml) while a

syrup for oral (including sublingual/bucal) premedication of children may be prepared

in the local pharmacy (registered in the US).

8

A neglected topic in clinical studies is the relative potency of the BZD on various

BZD-effects. Although animal studies indicate that there are such differencies, the

rationale for claiming some BZDs to be more useful for sedative purposes and others

to be better suited as anticonvulsives, is inconclusive. One difference is found with

the classification of some drugs as partial antagonists (e.g., bromazenil), thus limiting

the sedative component in comparison to the anxiolytic one. In this relation, midazolam

must be considered a strong agonist.

9

Midazolam

Premedication of children

Ideally, the physician specifies what he requires and the substances meeting these

requirements are subsequently found. In real life, the desired drug effects are often

found incidentally while expecting other manifestations with the use of new drugs.

That is what happened to us when we started using midazolam for the premedication

of children.

We had expected a strong sedation, as was required then, giving the anaesthetist an

advantage in achieving sleep induction. At first, we were disappointed, because the

children usually arrived quite awake. It appeared that the smaller the children, the

less sedated they were, even when the dose of midazolam was adjusted to body

surface. Fortunately, it was also noted that, despite the absense of sedation, the

children arrived in the operating room in a cooperative mood. Indeed, they were

often rather euphoric, making a “drunken” impression. Although nobody would claim

analgesic properties from a BZD, school children are often surprisingly calm at the

attempt of establishing an IV line [5,6].

In a rather short time, the desired aim of paediatric premedication changed. Not deep

sedation and reduced combative qualities were now the desired properties but rather

anxiolysis and cooperation. It was also realised that not only drugs but also

organisational measures (e.g. verbal preparation, motivation, cooperation with parents)

were of importance for obtaining these qualities. In contrast, we did not miss the

anticholinergic effects, previously considered a “must“ in premedication. It was only

a somewhat uneasy feeling that in leaving the pharmacological effects of morphine-

scopolamine as the target of any premedication, we were now about to define other

drug effects as a therapeutic aim.

The search for a suitable application and then a relevant dosage soon resulted in the

following recommendations [Table 2]. The intramuscular route was widely abandoned.

The best results were obtained with rectal administration of the concentrated ampoule

solution. Oral administration was hampered by the problem that midazolam has a

disgusting taste which it proved difficult to cover up by taste-modifiers - still, this

succeeded to some better than to others.

10

Larger children will usually take the tablet, under the threat of other applications.

Some may prefer the nasal route (again directly from the ampoule) which, next to

direct intravenous injection, works fastest. It is impossible to predict which route of

administration may be the best in a certain hospital since also non-pharmacological

factors are of importance and all routes have been proven effective [7].

Table 2: Route of administration and dosage of midazolam for premedication of children

Newborns and infants below 6 months are not given any premedication. Some felt

that this border should lie even lower, at 3 months. The question is a double one,

never really answered: is there a need, and is there any effect of midazolam in less

mature brains? Obviously, this is felt by some pediatricians [8] since there is a

recommendation for the use of midazolam in neonates.

A persistant and bigger problem remains the “paradoxical reaction” [PR] of about 5%

of the children, occasionally calling for IM ketamine induction. This failure of effect

differs from the phenomenon of PR defined below in not relating to any interactions

and therefore perhaps calls for another term to be used. Still, to those who remember

what paediatric anaesthesia induction was like previously, introduction of midazolam

represents an enormous advantage.

Route of administration

Age

Dose

Max

min before

(year)

(mg/kg)

(mg)

operation

Intramuscular

0.5-2

0.2

3

     15

or  Nasal

2-6

0.15

5

6-12

0.1

7.5

12-

0.075

7.5

Rectal

0.5-2

0.7

10

     20

2-6

0.6

15

6-12

0.4

15

12-

0.3

15

Oral (mixture)

4-6

0.6

10

     30

6-12

0.4

10

       (tablet)

8-

7.5

11

Midazolam

Premedication of adults

Previous means for premedication mainly included opioids, anticholinergics and

neuroleptic agents. This might produce an outwardly calm patient but some of these

later reported suffering serious panic (“I would have run away but I couldn’t“) [9,10].

Regrettably, the prolonged use of neuroleptic drugs betrays that many anaesthetists

did not routinely interview their patients postoperatively. This phenomenon changed

convincingly with the introduction of midazolam for premedication.

As for the children, the effect of premedication in adults initially gave rise to

disappointment with the expected degree of sedation. But then it was readily

recognized that sedation is not a feature required in premedication - anxiolysis,

preferably without any dangerous sedation, is the superior quality. Slowly emerged

another recognition, in sharp contrast to earlier standards: there is no need for

analgesics (exception: acute or chronic pain), neither are anticholinergics generally

required; indeed, when they are not strictly needed, these two drug compounds may

rather be the reason for adverse effects.

As a consequence of altered clinical options, the dose recommended was altered to

7.5 mg orally, which is equipotent to about 3 mg IM. Even then, the anaesthetist was

repeatedly confronted with amnesia when the patient the following day claimed not

to have experienced anything since leaving the ward. Declaring amnesia to a target

of premedication is, however, not reasonable and represents a repetition of older

mistakes when the pharmacological effects of other drugs were declared the desirable

target of any premedication. A less charming manifestation of amnesia can be

experienced when a patient claims that he did not feel any pain from the needle last

time. Generally, the anaesthetist may then restore his pride when asking the patient

about his experiences the following day.

Even 3.75 mg orally has been shown to blunt the immunological response [11]. A

dose reduction is recommended for old patients. Very old patients and those with

chronic respiratory depression should not receive any premedication at all. Caution is

necessary when opioids are required.

12

The paradoxical reaction

In contrast to the expected sedation, as experienced with higher doses of midazolam,

restlessness and confusion occasionally occurred when used for facilitating regional

anaesthesia. This soon came to be feared as “the paradoxical reaction“ [PR]; and

when titrated during regional anaesthesia, slight restlessness turned worse and

occasionally made a general anaesthesia necessary, simply in order to keep the patient

on the operating table. An adverse effect of the drug, we thought, until we found out

that we were provoking it unknowingly ourselves [12]. We had initially utilized a

traditional premedication with phenothiazines and atropine, fearing that sedation

would complicate the cooperation of the patient (later we learned that it did not).

Under the assumption to antagonize midazolam - before flumazenil was available, -

the patients with a PR were given 0.4-1.2 mg of physostigmine IV, which did not only

calm them down but resulted in what you might well call a “paradoxical sedation“.

From the success of physostigmine, it was evident that the PR was to be considered

an expression of a central anticholinergic syndrome [13] and as such the result of a

drug interaction between the BZD on the one side and an anticholinergic or neuroleptic

drug on the other. So this experience demonstrated both how to treat the PR and how

to prevent it. Besides, using midazolam alone for premedication, we did not experience

such effects.

The term PR has been (ab-)used in a broader sense. It seemed that alcoholics

occasionally reacted to BZD without needing other drugs for creating a PR. Also the

PR found in children seems to arise from another genesis. In such cases, using the

experiences from emergency medicine [14], another synergism can be used: the se-

dative effect of nalbuphine (even when no analgesics were required) with midazolam.

For an effective utilisation of this interaction, it is essential that nalbuphine is given

first (adult dose 20 mg) whereupon midazolam is titrated in mg-doses. That, however,

demands that it is suspected in advance that the patient will cause problems.

Nalbuphine is not a drug for titration; therefore, its application at a later time, after

midazolam has proved insufficient, may lead to an unpredictable sedation.

13

Midazolam

Sedation for regional anaesthesia

The lesson learned from the PR was to titrate midazolam and, perhaps more importantly,

change the options. The fearless and relaxed, not the sleepy patient was our target;

these, at least, were the new options to which the patients also could easily be

persuaded. Occasionally, it proved more difficult to convince the surgeon that this

was feasible.

In order to avoid any PR, the patients were premedicated with midazolam. Many

brave men and women did not want it at all but vasovagal reactions became rare

when you insisted on it. Atropine may be the treatment required when the pulse-rate

suddenly goes down but it has no preventive action, on the contrary. Then, of course,

the regional anaesthesia must be convincing. Only with these preconditions, you can

proceed.

The key to success lies in adding tiny doses of 1 mg (or even 0.5 mg) midazolam in a

titrated fashion. Probably adding to the effect of 7.5 mg orally as a premedication,

patients would generally require 1.5-2 mg. This made another formulation preferable,

that of 5 mg in 5 ml ampoules. The patient would even snore loudly, but assuming

sleep or amnesia in advance would lead to instant disappointment and should therefore

not be a declared therapeutical option. Music on headphones (according to the patient’s

personal choice) can be an excellent supplement, even if the experience is later found

to be buried in amnesia.

In the cases where, for some reason, a sleep seems mandatory, the author prefers the

combination with nalbuphin as described in the previous chapter. Using these

modifications (and acting for consequent prevention and treatment of post-dural

puncture headache), regional anaesthesia could be offered on a larger scale than

previously seemed suitable.

14

Sedation for diagnostic procedures

Sedation for endoscopy shows much resemblance to the challenge of adjoining regi-

onal analgesia. Preferable seems the IM premedication with 5 mg, but some physicians

are not afraid to meet the tablet again in the stomach (this encounter was never

reported to me). In most cases, an IV line is then not required. It is, of course, possible

to perform gastroscopy without any sedation, but patients clearly prefer midazolam

to nothing [15]. For a day-case patient, early ambulation is an essential quality;

however, midazolam should not be given to patients who, in spite of all warnings,

insist on driving by themselves after the examination. The doctor should refuse to

take any co-responsibility for the traffic events that day.

Midazolam certainly acts with before-mentioned BZD-effects: anxiolysis, sedation

and (at least to a certain degree) amnesia. It does, however, also contribute to a

reflex depression of the upper airways [16,17].

When a more sophisticated gastroscopy is required (for banding of oesophagical

varices), the patient is often of a certain constitution that makes a deeper sedation

necessary. This is occasionally also necessary for facilitating coloscopy, often a rather

painful matter. Here, again, the above-mentioned sedative synergism between an

opioid and midazolam has proved useful. Certainly, such strong sedation is not quite

suitable for day-case patients, but neither are bleeding oesophagel varices, and it

still proved possible to avoid a general anaesthesia in most of these cases.

For other diagnostic measures, when sedation is indeed needed (e.g. in children), it is

advisable to use premedication as described above and/or in a titrated fashion of mgs

of midazolam IV.

In case of the need for stronger sedation, the combination of low doses of midazolam

and propofol may make a regimen with spontaneous ventilation (under oxymetry

monitoring) possible through favourable synergism [18] (compare with co-induction

in general anaesthesia).

15

Midazolam

General anaesthesia

It is possible to induce hypnosis with 0.15-0.2 mg/kg midazolam IV; however, what is

possible is not always preferable. Generally, a rather awake patient is preferred soon

after surgery and using a full induction dose of midazolam is not the correct way to

achieve this. Personally, I have used the high dose only in patients where post-opera-

tive ventilation was expected.

Why, then, use it in anaesthesia at all? The answer is a philosophical one [19], therefore,

not accessible to all. While the reader’s patients may be heavily “asleep“ (some

comatised condition commonly understood as anaesthesia), the victims of the author’s

techniques might have been influenced by verbal stimuli. They did not complain of it

afterwards but I knew it even then; in the meantime, there is solid evidence for the

existence of cerebral activity in response particularly to auditory stimuli (speech).

The “awareness in anaesthesia“, as represented by the patient who then sues the

anaesthetist for a wakeful condition when everybody assumed her or him to be asleep,

is just an extreme, a rare event that is frequently obscured by amnesia.

Practically, the use of midazolam can be facilitated with “co-induction“: Using 0.05

mg/kg of midazolam with 1.5 mg/kg of thiopental (a third of the usual dosage) produces

a favourable synergism, by which the adverse effects of thiopental are reduced when

midazolam is added at a crucial time. Somewhat disturbing, it has not been possible

to measure the effect of midazolam when given in this context. Its use remains based

upon philosophical considerations. Co-induction also make sense with other hypnotics

[20].

In day-case patients, co-induction is not recommended. After all, these patients should

regain vigilance more readily. This is then best assured by total intravenous anaesthesia

based upon propofol.

When using ketamine, it has been recommended to add a BZD for suppressing the

hallucinogenic effect. This effect has been reduced but not eliminated with the

introduction of the racemate (s-ketamine), still the combination is recommended.

For this type of anaesthesia, midazolam has proved preferable to diazepam [21,22].

16

Intensive care

We had already gained important experiences with midazolam when the drug was

finally introduced in the intensive care unit [ICU]. Nonetheless, other anaesthetists

were still making the same mistakes there as described in other fields. Midazolam

was expected to produce a long-term anaesthesia which would then cover up some

of the mess typically found within the ICU. Patients are often struggling with life-

threatening conditions and have extreme pain and discomfort to cope with as well.

This is then amplified through various organizational failures, e.g. an eternal

disturbance through alarms, a special jargon adopted by the staff towards their

speechless patients and more. Not surprising to lay people, this atmosphere more

than occasionally results in serious psychic disturbances.

This seems to touch one of the frequent challenges met at ICUs - delirant conditions

[23,24]. The occurrence of that is then often seen by the staff as a confirmation that

the patient has an alcohol problem (which may, of course, be counted to the etiologies).

Who is to blame? That the ICU had any co-responsibility in producing this condition

is rarely recognized.

Midazolam is not an antipsychotic drug and such are necessary in the treatment of

delirant conditions. It may, however, be useful in the prevention of psychotic reactions

- not to cover up unpleasant circumstances in the ICU but along with organizational

reshapening. Such non-pharmacological measures include a respectful behaviour to

the patients, also apparently confused and unresponsive ones, a restriction of auditory

disturbances and measures in support of a diurnal rhythm - let the patient, when

possible, sleep without interruption from drug injection and other manipulations at

short intervals. Other pharmacological interventions include pain treatment with

systemic or regional measures.

Patients judged to be in need of prophylactic midazolam at the ICU [25,26] are given,

for example, a continuous infusion of 0.2-1.0 mg/h from 04 am - 10 pm (anxiolytic

dose) and 1-2 mg from 10 pm to 04 am (hypnotic dose). In the evening, the altered

infusion rate is not producing higher plasma levels at once, therefore a bolus of 2.5 is

added; similarly, changing the infusion rate in the morning does not instantly alter

17

Midazolam

the plasma level, but here you can calculate 2 hours for the adjustment. Certain

measures, carried out thrice a day, are then confined to the hours 6 am, 2 pm and 10

pm, while the sleeping time is kept as free as possible from any intrusion.

The failure of midazolam to sedate a patient should not be understood as an adverse

effect of the drug itself but indeed as a symptom of severe psychotic disturbances.

Similarly, the failure of moderate doses of antipsychotic drugs (e.g. haloperidol) to

influence the patients must be understood as a symptom in itself, not as a drug

failure. Here, anaesthetists obviously have a lot to learn from psychiatrists.

A different, possibly more frequent use of midazolam is found in the production of

strong analgosedation of ventilated patients. Here doses of 2-5 mg per hour are

regularly used. Again, it is recommended to use small boluses (~2.5 mg) when an

increased infusion rate is required, whereas a complete intermission for 1-2 hours

will show the effect of a dose-reduction more rapidly.

For analgosedation (or sedoanalgesia, a suitable clinical concept), practically all opioids

can be considered without caring for any possible synergistical respiratory depression

[27], once the patient is anyhow ventilated.

Used under these criteria, it is possible to avoid the application of continuous

anaesthesia (propofol infusion or combination with ketamine) in most cases but not

all. The indiscriminate use of propofol for all ventilated patients can be understood as

an indication that some organisational measures, as indicated above, need to be

taken. Just calling it a “sedation“ does not remove the adverse effects of anaesthesia.

Midazolam has now been used for many years in the ICU and has stod this test of

time brilliantly. A recent European survey revealed 63% use of midazolam for sedation

vs. 35% propofol and 2% others [28]. It needs attention that the pharmacokinetics of

many drugs, including midazolam [29], is altered in intensive care patients. It may

thus rarely be necessary to use the BZD-antagonist flumazenil. Personally, I under-

stand this as only a theoretical option when using midazolam whereas I have

experienced some “surprises“ in patients who had been given chordiazepoxid, diazepam

or flunitrazepam for premedication even days before.

18

Use as an anticonvulsive

All BZD act as anticonvulsives and some are used for long-term preventive purposes.

Midazolam is only to be considered for treatment of convulsive states and immediate

prophylaxis thereafter (exceptions to be found in the ICU [30]). What then makes the

drug a logical choice in emergency use is its fast action after IM application - it is

difficult to establish an IV line on a jerking arm - and its short duration of action,

making EEG examinations possible the following day.

The advantage of midazolam for emergency use was recognized shortly after its

introduction [31], and its therapeutical use as an anticonvulsive has been confirmed

in other studies [32]. It thus eliminates the need for any other BZD in the emergency

equipment. The use includes a prophylactical IM application after the convulsions

have stopped. The actual need of it has been challenged with the arguments that

most convulsions are singular, and long-acting BZD may disturb diagnosis for a

considerable time. All BZD may induce a beta-rhythm in EEG; it is therefore advisable

to postpone electroencephalographic diagnostics to at least the following day - which

would still not suffice for other BZD, requiring an interval of 3-7 days. Arguments for

the prophylactic IM application are: 1) this makes the emergency physician ready for

another prehospital mission and 2) it becomes unnecessary to search at length for an

IV access in an infant (typically after febrile convulsions) which has in the meantime

woken up.

The IM application had indeed a problem of acceptance among physicians who believed

rectal diazepam should remain the method of choice. Recent studies have confirmed

that also midazolam 0.2 mg/kg intranasal [33] or buccal/sublingual 0.3 mg/kg [34,35]

are effective (though hardly advantageous) against seizures. A precondition for the

use of midazolam as the sole BZD in emergency medicine is that it also acts as an

effectful anticonvulsive.

19

Midazolam

Emergency Medicine

The use as an anticonvulsant is perhaps a minor topic in emergency medicine. If,

however, that purpose can be fulfilled with a drug predominantly used for other

indications, this is advantageous, given the limited drug supply.

Hyperventilation attacks can sometimes not be clearly distingushed from convulsions.

A standard dose of 2.5 mg IV or 5 mg IM will generally solve the problem more rapidly

than rebreathing measures and many comforting words. Except for this condition,

the use as an anxiolytic should be restricted for patients anyhow to be admitted,

since the brief action makes midazolam unsuitable for longer treatments and this is,

in any case, beyond the scope of an emergency physician.

Midazolam is used very frequently along with an opioid in so-called “analgosedation“

(“sedoanalgesia“). The rationale is that a rather small dose of midazolam (1.5-2.5 mg)

through its central effect on nociception potentiates any opioid; in addition, it adds

to the sedative effects of 

κ

-stimulating opiods, e.g. nalbuphin. The surprise of a sudden

emergency gives anxiety a dominant place in generation of pain, explaining the success

of this combination exactly prehospitally beyond a general synergism with the opioid.

The need for prehospital sedation without intubation is a problem of its own. It

generally obliges continued monitoring during transport where the physician might

have been ready for another mission. Not always successful is the use in psychotic

and violently aggressive patients, who are therefore initially given nalbuphine 20-40

mg and, when everything has failed, ketamine thereafter. The experiences from pediatric

premedication may in a few cases be adopted for facilitating an IV line in children 5-

10 min after IM application.

Particularly prehospitally, midazolam may be used as an adjunct for ketamine

anaesthesia [36]. Although intubation is generally called for in prehospital anaesthesia,

a few cases, predominantly children, may not need ventilation.

20

An alternative principle for prehospital anaesthesia involves the use of midazolam

only after successful intubation [37]. It is recommended to carry out the intubation

exclusively with etomidate and a suitable adjunct, e.g. nalbuphine. With the opioid

already given (and/or possibly added later), midazolam is then titrated according to

the demand. This demand is signified by movements or wakeful appearences, a primi-

tive principle for primitive surroundings and thus highly effective.

A personal testimonial for the success of this techniqe may be worth considering: the

author’s preferred intubation technique (blind nasotracheal) was dependent upon

persistent breathing, excluding midazolam and most opioids along with etomidate.

Possibly facilitated through the route of intubation (an orotracheal tube produces a

persistant noxious reflex stimulation), it was probably easier to keep the patients

sedated.

The dosage used varies according to the general and haemodynamic condition. Patients

in shock need much less and comatose trauma patients more than the average

requirements. The dosages of midazolam are shown in Table 3.

Table 3: Dosage of midazolam after intubation until admission of 162 consecutive patients.

Primary indication for endotracheal

n

Midazolam, mg

intubation

(average)

Comatose, trauma

53

12.4

Comatose, non-trauma

37

8.1

Shock, trauma

26

7.7

Shock, non-trauma

26

 2.8

No shock, all

110

10.8

No of Pat.s (n)/average

162

9.1

Taking all indications together, midazolam became the most frequently used drug in

the author’s prehospital emergency kit.

21

Midazolam

The benzodiazepine-antagonist

Flumazenil is the first and so far only clinically available BZD-antagonist, acting by a

strong affinity and negligible intrinsic action at the benzodiazepine receptor. It does,

however, produce weak anticonvulsive properties [38], which may help explain why

convulsions are extremely rare after reversal of benzodiazepine sedation with this

drug. Other agonistic activities of flumazenil (all weak and clinically irrelevant) include

anxiolytic activity in stimulated stress [39]. The elimination half-life of 53 min is

shorter than any benzodiazepine agonists, making rebound sedation a possibility after

long-acting BZD. Like midazolam, it is a water-soluble imidazo-benzodiazepine and

can therefore be administered IM or mixed in an infusion. In order to avoid sudden

withdrawal, the drug is titrated with 0.1 mg each min until desirable effect or a

maximum 1.0 mg is reached. It is recommended to add any excess to the infusion

solution in order to provide an increased duration of action. It may produce some

awakening effect in alcohol intoxication [40] and hepatic coma [41] but complete

awakening can be regarded as a diagnostic (and therapeutic!) proof of BZD overdose.

I have used flumazenil quite a number of times, also prehospitally [42], but very

seldom towards midazolam. Still, I consider the mere availibility of this antagonist

also of importance for midazolam, since it enables the reversal of any excess

(undesirable) sedation or respiratory depression in unexpected synergism [43] or

presence of other disease conditions resulting in awkward effects of the drug. Thereby,

flumazenil adds substantially to the safety of midazolam - provided it is available

among your drugs.

A less complete and unspecific reversal of midazolam sedation is achieved with

physostigmine [44] and aminophylline [45].

22

Synopsis: Indications and Dosages

Table 4: Synopsis of the text. Titr.'>Titr.: repetitive titrated intravenous doses of 0.5-1.0 mg midazolam

until desired effect (or a maximum of 5 mg) is obtained. IV application if not specified.

Indication

Dosage and Application

Premedication

Children

Rectal, p.o., nasal, IM

(See table 2)

Adults

7.5 mg p.o. or 3-5 mg IM

Regional anaesthesia

1-2 mg initially + 0.5-1.0 mg IV Titr. (max. 5 mg)

Analgosedation

Opiod, then midazolam Titr. as above

Strong sedation

Nalbuphine 20 mg first, then IV Titr. as above

Sedation for endoscopy

7.5 mg p.o. or 5 mg IM + 1-2 mg IV + Titr.

General anesthesia

5 mg 1 min IV before hypnotic in reduced dosage

or

0.15-0.2 mg/kg with opioid agonist

(0.3 mg/kg without premedication)

Intensive care

Anxiolysis & hypnosis

04-22 h: 0.2-1 mg/h

22-04 h: 1-2 mg/h

22 h: 2.5 mg + Titr.

Sedation by ventilatory support

2-5 mg/h + opioid by infusion

Convulsive status

2.5 - 15 mg IV Titr., followed by 10-15 mg/h

Emergency medicine

Analgosedation

Nalbuphine 20 mg first, then 2.5 mg IV + Titr.

or

opioid agonist first, then 2.5 mg IV + Titr.

Seizures, therapy

5 mg IM + Titr. IV

Seizures, prophylaxis

0.2 mg/kg IM

(infants)

or

0.3 mg/kg sublingual

Hyperventilation

2.5-5mg IM/IV

„Anaesthesia“

2.5-5 mg IV + Titr. (see guidance in Tab. 3)

after intubation

Antagonism

Flumazenil 0.2-0.5 mg Titr.

23

Midazolam

Comments

Therapeutic option:

cooperation and anxiolysis, not sedation

(degree of sedation inversely proportional to age)

Option: anxiolysis, amnesia

may occur (unpredictable)

Music on headphones

The opiod should be given first, midazolam then titrated

When strong sedation, not analgesia is dominant demand

Cautious use for outpatients (no driving home afterwards)

Co-induction, bringing anxiolysis into general anaesthesia

Patients more sedated postoperatively

Plus organizational measures, preferably with sedation monitoring

Possible alternative: propofol (then better call it anaesthesia)

Ventilatory support should prevail at such doses

Nalbuphine is not useful for titration purposes

Ventilatory monitoring (oxymetry) mandatory

IV line after success from IM dose (avoiding IV line after febrile

consulsions)

Higher dose in hospital

Caution in shock and hypovolaemia

Good to have just in case...

24

Abbreviations:

BZD

Benzodiazepine

ICU

Intensive Care Unit

IM

Intramuscular

IV

Intravenous

p.o.

Peroral (by mouth)

PR

Paradoxical Reaction

Titr.

Titrated, i.e.: repetitive small IV doses until desired effect

Caution is required for the use of midazolam

í

  with chronic respiratory failure

í

  in old age

í

  with hypovolaemia or shock

í

  by interaction in particular with anticholinergics and neuroleptics (PR)

í

  by interaction with other sedatives

í

  by interaction with opioids

Table 5: Conditions where adverse effects are prone to occur.

25

Midazolam

Conclusion

With midazolam, not just another BZD was introduced into clinical use. The drug is

better titrated than any other BZD and its short duration of action turned out to be

important for its use in anaesthesia and related fields. The principles for premedication,

both paediatric and adult, were changed as a consequence of its introduction (it may

now be difficult for younger anaesthetists to imagine what premedication was like

before). No other BZD had been used the way, midazolam now gained introduction

into anaesthesia, intensive care and emergency medicine.

Along with the use, also the options for the expected BZD-effects changed. It was

realized that a cautious and small dosage was often effective in removing anxiety

without necessarily demonstrating sedation. Should it not suffice, you could add a

mg or two, whereas you could hardly draw back 5 mg that had already been injected.

Even as it became possible to antagonise the drug with flumazenil, you would certainly

prefer to avoid any excess effect, leaving the antagonist a potential rescue medication

“better not used“.

We can now look back at 20 years of clinical use of Midazolam - surprising adverse

effects of major importance are not apt to occur. Still, some interactions, indicated in

this description, deserve studies. Uncritical use is, of course, not to be advocated, but

for anaesthetists, intensive care and emergency physicians, it has become a drug

worthy of being known by heart. It remains an important resource in the clinical

areas here described.

26

27

Midazolam

References

1

Amrein R, Hetzel W. Pharmacology of Dormicum (midazolam) and Anexate

(flumazenil). Acta Anaesthesiol Scand 1990;92(Suppl.):6-15.

2

Gross JB, Blouin RT, Zandsberg S, Conard PF, Haussler J. Effect of flumazenil on

ventilatory drive during sedation with midazolam and alfentanil. Anesthesiology

1996;85:713-20.

3

Mohler H, Okada T. Benzodiazepine receptor: demonstration in the central nervous

system. Science 1977;198:849-51.

4

Squires RF, Braestrup C. Benzodiazepine receptors in rat brain. Nature

1977:266,732-4.

5

Schou J, Atanassov P. Prämedikation mit Midazolam in der Kinderanästhesie.

Kinderarzt 1986;17:326-9.

6

McErlean M, Bartfield JM, Karunakar TA, Whitman MC, Turley DM. Midazolam

syrup as a premedication to reduce the discomfort associated with pediatric

intravenous catheter insertion. J Pediatr 2003;142:429-30.

7

Kogan A, Katz J, Efrat R, Eidelman LA. Premedication with midazolam in young

children: a comparison of four routes of administration. Paediatr Anaesth

2002;12:685-9.

8

Arya V, Ramji S. Midazolam sedation in mechanically ventilated newborns: a

double blind randomized placebo controlled trial. Indian Pediatr. 2001;38:967-72.

9

Tolksdorf W, Berlin J, Bathke U, Nieder G. Psychische und somatische Auswirkun-

gen der Prämedikation mit Rohypnol, Thalamonal und Placebo in Kombination

 mit Atropin. Anästh Intenivther Notfallmed 1981;16:1-4.

10 Seibert W. Thalamonal-Prämedikation als Auslöser extremer Angst und die post-

operativen Folgen. Anästhesist 1987;36:662-3.

11 Heine GH, Weindler J, Gabriel HH, Kindermann W, Ruprecht KW. Oral

premedication with low dose midazolam modifies the immunological stress

reaction after the setting of retrobulbar anaesthesia. Br J Ophthalmol

2003;87:1020-4.

12 Knaack-Steinegger R, Schou J. Therapie und Prophylaxe der paradoxen Reaktion

nach Midazolam zur Regionalanästhesie, Anaesthesist 1987;36:143-6.

28

13 Ruprecht J, Dworacek B. Central anticholinergic syndrome in anaesthetic practice.

Acta Anaesth Belg 1976;27:45-60.

14 Schou J. Prehospital Emergency Medicine - Challenges and Options in Rescue

Services, 2nd Edition. Harwood Academic Publ., Amsterdam B.V., 448 pp, 1997.

15 Bonta PI, Kok MF, Bergman JJ, Van den Brink GR, Lemkes JS, Tytgat GN, Fockens

P. Conscious sedation for EUS of the esophagus and stomach: a double-blind,

randomized, controlled trial comparing midazolam with placebo. Gastrointest

Endosc 2003;57:842-7.

16 Murphy PJ, Erskine R, Langton JA. The effect of intravenously administered

diazepam, midazolam and flumazenil on the sensitivity of upper airway reflexes.

Anaesthesia 1994;49:105-10.

17 D’Honneur G, Rimaniol JM, el Sayed A, Lambert Y, Duvaldestin P. Midazolam/

propofol but not propofol alone reversibly depress the swallowing reflex. Acta

Anaesthesiol Scand 1994;38:244-7

18 Paspatis GA, Manolaraki M, Xirouchakis G, Papanikolaou N, Chlouverakis G, Gritzali

A. Synergistic sedation with midazolam and propofol versus midazolam and

pethidine in colonoscopies: a prospective, randomized study. Am J Gastroenterol

2002;97:1963-7.

19 Schou J. A Philosophical Approach to Anaesthesia. Alix Publ., 104 pp, 1994.

20 Adachi YU, Watanabe K, Higuchi H, Satoh T. A small dose of midazolam decreases

the time to achieve hypnosis without delaying emergence during short-term

propofol anesthesia. J Clin Anesth 2001;13:277-80.

21 Cartwright PD, Pingel SM. Midazolam and diazepam in ketamine anaesthesia.

Anaesthesia  1984;59:439-2.

22 Toft P, Romer U. Comparative evaluation of midazolam and diazepam to

supplement total intravenous anaesthesia with ketamine vor endoscopy. Can J

Anaesth 1987;34:466-9.

23 Wilson L. Intensive care delirium. Arch Intern Med 1972;130:225-6.

24 Hemmingsen R, Kramp P, Rafaelsen OJ. Delirium tremens and related clinical

states. Acta Psychiatr Scand 1979; 59:337-69.

25 Schou J, Kübler J, Scherb M. Induktives Monitoring zur Beurteilung und Propyhlaxe

psychischer Störungen auf der Intensivstation. Intensivmed 1992;29(Suppl):67-71.

29

Midazolam

26 Treggiari-Venzi M, Borgeat A, Fuchs-Buder T, Gachoud JP, Suter PM. Overnight

sedation with midazolam or propofol in the ICU: effects on sleep quality, anxiety

and depression. Intensive Care Med 1996;22:1186-90.

27 Conti G, Merdurio G, Iacobone E, Auricco D, Liberati Q. Sedation in the intensive

care unit. Minerva Anestesiologica 2002;68:240-4.

28 Soliman HM, Mélot C, Vincent J-L. Sedative and analgesic practices in the inten-

sive care unit. Br J Anaesth 2001;87:186-92.

29 Bolon M, Boulieu R, Flamens C, Paulus S, Bastien O. Sedation par le midazolam

en reanimation: aspects pharmacologiques et pharmacocinetiques. Ann Fr Anesth

Reanim 2002;21:478-92.

30 Ulvi H, Yoldas T, Mungen B, Yigiter R. Continuous infusion of midazolam in the

treatment of refractory generalized convulsive status epilepticus. Neurol Sci

2002;23:177-82.

31 Schou J. Midazolam zur Sedierung und Krampfbehandlung in der Notfallmedizin.

Europäischer Anästhesiekongress, Wien 1986, Abstract vol. III, Nr. 862.

32 Galvin GM, Jelinek GA. Midazolam: an effective intravenous agent for seizure

control. Arch Emerg Med 1987;4:169-72.

33 Lahat E, Goldman M, Barr J, Bistritzer T, Berkovitch M. Comparison of intranasal

midazolam with intravenous diazepam for treating febrile seizures in children:

prospective randomised study. BMJ 2000;321:83-6.

34 Scott RC, Besag FM, Neville BG. Buccal midazolam and rectal diazepam for

treatment of prolonged seizures inchildhood and adolescence: a randomised trial.

Lancet 1999;353:623-6.

35 Kutlu NO, Dogrul M, Yakinci C, Soylu H. Buccal midazolam for treatment of

prolonged seizures in children. Brain Dev 2003;25:275-8.

36 Bourgoin A, Albanese J, Wereszczynski N, Charbit M, Vialet R, Martin C. Safety of

 sedation with ketamine in severe head injury patients: comparison with sufentanil.

Crit Care Med 2003;31:711-7.

37 Schou J. Three techniques for prehospital emergency anaesthesia. JEUR

1994;3:139-45.

38 Scollo-Lavizzari G. The anticonvulsant effect of the benzodiazepine antagonist

Ro 15-1788. Eur Neurol 1984;23:1-6.

30

39 Strohle A, Wiedemann K. Flumazenil attenuates the pituitary response to CRH in

healthy males. Eur Neuropsychopharmacol 1996;6:323-5.

40 Scollo-Lavizzari G, Matthis H. Benzodiazepine antagonist (RO 15-1788) in ethanol

intoxication: a pilot study. Eur Neurol 1985;24:352-4.

41 Goulenok C, Bernard B, Cadranel JF, Thabut D, Di Martino V, Opolon P, Poynard T.

Flumazenil vs. placebo in hepatic encephalopathy in patients with cirrhosis:

a meta-analysis. Aliment Pharmacol Ther 2002;16:361-72.

42 Schou J, Deklerk J, Scherb M, Kübler J. Antagonism vs. ventilation in drug overdose.

JEUR 1995;8:136-9.

43 Gross JB, Blouin RT, Zandsberg S, Conard PF, Haussler J. Effect of flumazenil on

ventilatory drive during sedation with midazolam and alfentanil. Anesthesiology

1996;85:713-20.

44 Ebert U, Oertel R, Kirch W. Physostigmine reversal of midazolam-induced

electroencephalographic changes in healthy subjects. Clin Pharmacol Ther

2000;67:538-48.

45 Bonfiglio MF, Fisher-Katz LE, Saltis LM, Traeger SM, Martin BR, Nackes NA, Perkins

TA. A pilot pharmacokinetic-pharmacodynamic study of benzodiazepine

antagonism by flumazenil and aminophylline. Pharmacotherapy 1996;16:1166-72.

31

Midazolam

32

John Schou, M.D.

County Hospital of Loerrach

Spitalstr. 25

D-79539 Loerrach, Germany

Alix Publishing

Wallbrunnstr. 106E · D-79539 Loerrach · Germany

© 2004

ISBN 3-928811-21-5