INTOX Home Page


    VOLUME 1


    IPCS/CEC Evaluation of Antidotes Series

    IPCS   International Programme on Chemical Safety
    CEC    Commission of the European Communities

    Volume 1   Naloxone, flumazenil and dantrolene as antidotes
    Volume 2   Antidotes for poisoning by cyanide

    This important new series will provide definitive and authoritative
    guidance on the use of antidotes to treat poisoning.  The
    International Programme on Chemical Safety (IPCS) and the Commission
    of the European Communities (CEC) (ILO/UNEP/WHO) have jointly
    undertaken a major programme to evaluate antidotes used clinically
    in the treatment of poisoning.  The aim of this programme has been
    to identify and evaluate for the first time in a scientific and
    rigorous way the efficacy and use of a wide range of antidotes.
    This series will therefore summarise and assess, on an antidote by
    antidote basis, their clinical use, mode of action and efficacy. The
    aim has been to provide an authoritative consensus statement which
    will greatly assist in the selection and administration of an
    appropriate antidote. This scientific assessment is complemented by
    detailed clinical information on routes of administration,
    contraindications, precautions and so on.  The series will therefore
    collate a wealth of useful information which will be of immense
    practical use to clinical toxicologists and all those involved in the
    treatment and management of poisoining.

    Scientific Editors

    Department of Health, London, United Kingdom

    Ulleval University Hospital, Oslo, Norway

    International Programme on Chemical Safety,
    World Health Organization, Geneva, Switzerland

    Health and Safety Directorate,
    Commission of the European Communities, Luxembourg

    EUR 14797 EN

    Published by Cambridge University Press on behalf of the World Health
    Organization and of the Commission of the European Communities


    The mention of specific companies or of certain manufacturers'
    products does not imply that they are endorsed or recommended by the
    World Health Organization in preference to others of a similar
    nature that are not mentioned.

    Neither the Commission of the European Communities nor any person
    acting on behalf of the Commission is responsible for the use which
    might be made of the information contained in this report.

    (c) World Health Organization, Geneva, 1993 and
    ECSC-EEC-EAEC, Brussels-Luxembourg, 1993

    First published 1993

    Publication No. EUR 14797 EN of the Commission of the European
    Communities, Dissemination of Scientific and Technical Knowledge
    Unit, Directorate-General Information Technologies and Industries,
    and Telecommunications, Luxembourg

    ISBN 0 521 45459 X hardback






         2.1. Introduction
         2.2. Name and chemical formula
         2.3. Physico-chemical properties
         2.4. Pharmaceutical formulation and synthesis
         2.5. Analytical methods
               2.5.1. Quality control
               2.5.2. Identification
               2.5.3. Quantification of the antidote
               2.5.4. Analysis of toxic agents
         2.6. Shelf-life
         2.7. General properties
         2.8. Animal studies
               2.8.1. Pharmacodynamics
               2.8.2. Pharmacokinetics
               2.8.3. Toxicology
         2.9. Volunteer studies
               2.9.1. Pharmacokinetics
               2.9.2. Pharmacodynamics
               2.9.3. Effects of high doses of naloxone
         2.10. Clinical studies - clinical trials
               2.10.1. Effects in therapeutic use of opioids
               2.10.2. Effects in acute opioid poisoning
         2.11. Clinical studies - case reports
               2.11.1. Naloxone in clonidine poisoning
         2.12. Summary of evaluation
               2.12.1. Indications
               2.12.2. Advised routes and dose
               2.12.3. Other consequential or supportive therapy
               2.12.4. Areas where there is insufficient information
                       to make recommendations
               2.12.5. Proposals for further studies
               2.12.6. Adverse effects
               2.12.7. Restrictions of use
         2.13. Model information sheet
               2.13.1. Uses
               2.13.2. Dosage and route
               2.13.3. Precautions/contraindications
               2.13.4. Adverse effects
               2.13.5. Use in pregnancy and lactation
               2.13.6. Storage
         2.14. References


         3.1. Introduction
         3.2. Name and chemical formula of antidote
         3.3. Physico-chemical properties
         3.4. Pharmaceutical formulation and synthesis
         3.5. Analytical methods
               3.5.1. Identification of the antidote
                Infrared spectroscopy
                Ultraviolet absorption
                Thin-layer chromatography
               3.5.2. Quantification of the antidote in biological
               3.5.3. Analysis of the toxic agent in biological
         3.6. Shelf-life
         3.7. General properties
         3.8. Animal studies
               3.8.1. Pharmacodynamics
               3.8.2. Pharmacokinetics
               3.8.3. Toxicology
                Acute toxicity
                Subacute toxicity
                Chronic toxicity
         3.9. Volunteer studies
               3.9.1. Pharmacodynamics
                BZD antagonist effect
                Intrinsic effects
               3.9.2. Pharmacokinetics
               3.9.3. Tolerance of flumazenil
               3.9.4. Other studies
         3.10. Clinical studies - clinical trials
               3.10.1. Anaesthesiology
               General anaesthesia
               Conscious sedation
               3.10.2. Benzodiazepine overdose or intoxication
         3.11. Clinical studies - case reports
         3.12. Summary of evaluation
               3.12.1. Indications
               3.12.2. Dosage and route
               3.12.3. Other consequential or supportive therapy
               3.12.4. Areas where there is insufficient information to
                       make recommendations

               3.12.5. Proposals for further study
               3.12.6. Adverse effects
               3.12.7. Restrictions of use
         3.13. Model information sheet
               3.13.1. Uses
               3.13.2. Dosage and route
               3.13.3. Precautions/contraindications
               Pharmaceutical precautions
               Other precautions
               3.13.4. Adverse effects
               3.13.5. Use in pregnancy and lactation
               3.13.6. Storage
               3.13.7. Special risk groups
         3.14. References


         4.1. Introduction
         4.2. Name and chemical formula of antidote
         4.3. Physico-chemical properties
         4.4. Pharmaceutical formulation and synthesis
         4.5. Analytical methods
               4.5.1. Identification and quantification of dantrolene
                       sodium and its formulation
               4.5.2. Quantification of dantrolene in body fluids
                High-performance liquid chromatography
         4.6. Shelf life
         4.7. General properties
         4.8. Animal studies
               4.8.1. Pharmacodynamics
                Effect on skeletal muscle
                Effects on other tissues
                Studies in malignant hyperthermia-
                                 susceptible pigs
               4.8.2. Pharmacokinetics
               4.8.3. Toxicology
                Acute toxicity
                Subacute toxicity
                Chronic toxicity
         4.9. Volunteer studies
               4.9.1. Administration and plasma concentrations
               4.9.2. Distribution
                Distribution to the fetus and
                                 newborn baby
               4.9.3. Elimination
               4.9.4. Human  in vitro pharmacodynamics
         4.10. Clinical studies - clinical trials
         4.11. Clinical studies - case reports

               4.11.1. Use in malignant hyperthermia
               Prophylaxis of malignant hyperthermia
               Prophylaxis of malignant hyperthermia
                                 during pregnancy
               4.11.2. Use in neuroleptic malignant syndrome
               4.11.3. Use in other drug-induced hyperthermia
         4.12. Summary of evaluation
               4.12.1. Indications
               Treatment of malignant hyperthermia
               Treatment of neuroleptic malignant
               Treatment of hyperthermia induced by
                                 muscle rigidity in poisoning
               4.12.2. Advised routes and doses
               Treatment of severe drug-induced
                                 hyperthermia, including malignant
               Prophylaxis of malignant hyperthermia
                                 prior to anaesthesia in susceptible
               4.12.3. Other consequential or supportive therapy
               4.12.4. Controversial issues and areas of insufficient
               4.12.5. Proposals for further studies
               4.12.6. Adverse effects
               Interaction with calcium antagonists
               4.12.7. Restrictions for use
         4.13. Model information sheet
               4.13.1. Uses as an antidote
               4.13.2. Dosage and route
               4.13.3. Precautions and contraindications
               4.13.4. Pharmaceutical incompatibilities and drug
               4.13.5. Adverse effects
               4.13.6. Use in pregnancy and lactation
               4.13.7. Storage
         4.14. References

         APPENDIX I    List of antidotes

         APPENDIX II   Principles for the evaluation of antidotes

         APPENDIX III  Proforma for monographs on antidotes for
                       specific toxic agents



    Dr D.N. Bateman, Department of Clinical Pharmacology, University of
    Newcastle, Newcastle-upon Tyne, United Kingdom

    Professor C. Bismuth, Hpital Fernand Widal, Paris, France

    Dr R.E. Ferner, West Midlands Poisons Unit, Dudley Road Hospital,
    Birmingham, United Kingdom  (Joint Rapporteur)

    Dr T.J. Meredith, Department of Health, London, United Kingdom

    Dr H. Persson, Poison Information Centre, Karolinska Sjukhuset,
    Stockholm, Sweden  (Joint Chairman)

    Professor L. Prescott, Scottish Poison Information Service, The Royal
    Infirmary, Edinburgh, Scotland  (Joint Chairman)

    Dr M.-L. Ruggerone, Ospedale Niguarda, Centro Antiveleni, Milan, Italy

    Dr H. Smet, Centre Belge Anti-Poisons, Brussels, Belgium

    Dr U. Taitelman, National Poisons Information Centre, Rambam Medical
    Centre, Haifa, Israel

    Dr W. Temple, National Toxicology Group, Otago University Medical
    School, Dunedin, New Zealand  (Joint Rapporteur)

    Professor A.N.P. van Heijst, Bosch en Duin, The Netherlands

    Dr G. Volans, Poisons Unit, New Cross Hospital, London, United Kingdom

    Dr E. Wickstrom, National Poison Centre, Oslo, Norway


    Dr G. Olibet, Centro Antiveleni, Milan, Italy


    Dr J.-C. Berger, Health and Safety Directorate, Commission of the
    European Communities, Luxembourg

    Dr J.A. Haines, International Programme on Chemical Safety, World
    Health Organization, Geneva, Switzerland

    Dr M. ten Ham, Pharmaceuticals Programme, World Health Organization,
    Geneva, Switzerland


         At a joint meeting of the World Federation of Associations of
    Clinical Toxicology and Poison Control Centres, the International
    Programme on Chemical Safety (IPCS), and the Commission of the
    European Communities (CEC), held at the headquarters of the World
    Health Organization in October 1985, the evaluation of antidotes used
    in the treatment of poisonings was identified as a priority area for
    international collaboration.  During 1986, the IPCS and CEC undertook
    the preparatory phase of a joint project on this subject.  For the
    purpose of the project an antidote was defined as a therapeutic
    substance used to counteract the toxic action(s) of a specified
    xenobiotic.  Antidotes, as well as other agents used to prevent the
    absorption of poisons, to enhance their elimination and to treat their
    effects on body functions, were listed and preliminarily classified
    according to the urgency of treatment and efficacy in practice.  With
    respect to efficacy in practice, they were classified as: (1) those
    generally accepted as useful; (2) those widely used and considered
    promising but not yet universally accepted as useful and requiring
    further research concerning their efficacy and/or their indications
    for use; and (3) those of questionable usefulness.  Additionally,
    certain antidotes or agents used for specific purposes were considered
    to correspond to the WHO criteria for essential drugs (see Criteria
    for the Selection of Essential Drugs, WHO Technical Report Series 722,
    Geneva, 1985).

         A methodology for the principles of evaluating antidotes and
    agents used in the treatment of poisonings and a proforma for
    preparing monographs on antidotes for specific toxins were drafted
    (Appendices II and III respectively).

         Monographs are being prepared, using the proforma, for those
    antidotes and agents provisionally classified in category 1 as regards
    efficacy in practice.  For those classified in categories 2 and 3,
    where there are insufficient data or controversy regarding efficacy in
    practice, it was agreed that further study was necessary. 
    Accordingly, several were selected for initial review and evaluation,
    among which were naloxone as an antagonist for opioids, flumazenil as
    a benzodiazepine antagonist and dantrolene for malignant hyperthermia.

         The review and evaluation of these antidotes was initiated at a
    joint meeting of the IPCS and the CEC, organized by the Northern
    Poisons Unit and held at the Medical School of the University of
    Newcastle-upon-Tyne, United Kingdom, 13-17 March 1989.  In preparation
    for this meeting, monographs were drafted, using the proforma, on
    naloxone by Dr D.N. Bateman, on flumazenil by Dr A. Brovard and
    Professor C. Bismuth and on dantrolene by Dr H. Smet and Professor C.
    Bismuth.  The draft document on naloxone was reviewed by a working
    group consisting of Professor L.F. Prescott (Chairman), Dr W. Temple
    (Rapporteur), Dr D. Bateman, Dr M. Ten Ham, Dr A.N.P. van Heijst, Dr
    G.N. Volans and Dr E. Wickstrom.  The draft documents on flumazenil

    and dantrolene were reviewed by a working group consisting of Dr H.
    Persson (Chairman), Dr R.E. Ferner (Rapporteur), Dr J.-C. Berger,
    Professor C. Bismuth, Dr G. Olibet, Dr M.-L. Ruggerone, Dr H. Smet and
    Dr U. Taitelman.

         Following the meeting further drafting work was undertaken by the
    authors, with the assistance of Drs R.E. Ferner, B. Britt (Department
    of Anaesthesia, Faculty of Medicine, University of Toronto, Canada),
    and T. Fagerlund (Institute of Medical Genetics, University of Oslo,
    Norway) in the redrafting of the dantrolene monograph.  Draft texts
    were further revised by the series editors (Dr T.J. Meredith, Dr D.
    Jacobsen, Dr J.A. Haines, and Dr J.-C. Berger), who also prepared an
    introduction to the series.  This introduction summarizes the results
    of the preparatory phase and indicates the volumes currently planned
    for this series.  The efforts of all who helped in the preparation and
    finalization of this volume are gratefully acknowledged.


    BZD     benzodiazepine
    CAT     computer-assisted tomography
    CNS     central nervous system
    GABA    gamma-aminobutyric acid
    GLC     gas-liquid chromatography
    HIV     human immunodeficiency virus
    HPLC    high-performance liquid chromatography
    LSD     lysergic acid diethylamide
    RIA     radio-immunoassay
    TLC     thin-layer chromatography
    UV      ultraviolet


         Antidotes play a vital role in the treatment of poisoned
    patients.  Good supportive care, directed particularly at the cardiac
    and respiratory systems, and the use of elimination techniques when
    indicated, enable the majority of poisoned patients to make a full
    recovery.  However, in certain circumstances the use of antidotes can
    be life-saving, and in other circumstances the use of antidotes may
    reduce morbidity as well as medical and other resources required in
    the care of a patient.  In areas remote from hospital care, and
    particularly in developing countries where facilities for supportive
    care outside hospital are often limited, the availability of certain
    antidotes is even more essential for the successful treatment of a
    poisoned patient.

         However, there remains controversy about the clinical efficacy
    and indications for use of many of the antidotes conventionally
    employed in the treatment of poisoning.  There is also sometimes
    difficulty in obtaining antidotes in an emergency situation,
    particularly if the substance in question is not available as a
    pharmaceutical preparation.

         The need for an international evaluation of the clinical efficacy
    of antidotes and other substances used in the treatment of poisoning
    was first recognized at a joint meeting of the World Federation of
    Associations of Clinical Toxicology Centres and Poisons Control
    Centres, the International Programme on Chemical Safety (IPCS) and the
    Commission of the European Communities (CEC), held at WHO
    headquarters, Geneva, 6-9 October 1985.  At the same time, the need to
    encourage the more widespread availability of those antidotes that are
    effective was also recognized.  As a result, a joint IPCS/CEC project
    was subsequently initiated to address these problems.

         In a preparatory phase of the project, an antidote was defined
    for working purposes as a therapeutic substance used to counteract the
    toxic action(s) of a specified xenobiotic.  A preliminary list of
    antidotes for review, as well as of other agents used to prevent the
    absorption of poisons, to enhance their elimination and to treat their
    effects on body functions, was established.  For the purposes of the
    review process, antidotes and other substances were classified
    according to the urgency with which treatment with the antidote was
    thought on current evidence to be required and the (currently judged)
    clinical efficacy of the antidote in practice.  Those corresponding to
    the WHO concept of an essential drug were designated as such.  Some
    have already been incorporated into the WHO list of essential
    drugsa.  Antidotes and similar substances for veterinary use were

    a  WHO (1988) Use of Essential Drugs. Model list of essential drugs
       (fifth list). Third Report of the WHO Expert Committee. WHO
       Technical Report Series 770, Geneva World Health Organization.

    also listed.  A methodology on the principles for evaluation of
    antidotes and other agents used in the treatment of poisonings was
    developed and this has subsequently been used as a framework for
    drafting monographs on specific antidotes.  The list of antidotes and
    other agents established as a result of the preparatory phase and the
    preliminary classification is given in Appendix I.  The principles for
    evaluation are detailed in Appendix II.

         Early during the course of the preparatory phase, it became
    apparent that the availability of antidotes differed from one country
    to another.  Problems of availability fell into three interrelated
    categories, namely:

    *    scientific, technical and economic aspects;

    *    regulatory and administrative requirements;

    *    geospatial and time considerations.

         Problems of availability of antidotes used in the treatment of
    poisonings were therefore examined by an IPCS/CEC Working Group,
    hosted by the Norwegian National Poisons Information Centre and held
    in Oslo, 20-22 June 1988.  The record of this meeting is given in
    ICS/88.44.  In preparation for this meeting, a preliminary survey was
    undertaken of selected poisons control centres in order to identify
    more precisely the practical difficulties encountered in obtaining
    antidotes.  The survey showed that, in general, poisons centres in
    industrialized countries had few problems in obtaining most antidotes,
    although lack of suitable preparations/importers/manufacturers
    together with administrative difficulties did hinder access to certain
    antidotes.  In contrast, centres in developing countries reported many
    problems in obtaining even those antidotes that are readily available

         A report was prepared by the IPCS/CEC Working Group setting out
    the problems associated with the availability of antidotes and
    suggesting ways in which the availability of antidotes might be
    ensured for the treatment of poisoned individuals.  In due course, it
    is intended that this report will be brought to the attention of all
    relevant national drug regulatory and importation authorities,
    pharmaceutical manufacturers, distributors of pharmaceutical
    materials, and all poisons control centres.  The IPCS Guidelines for
    Poisons Control summarize the problems and issues of availability
    identified by the Working Groupb.

    b  WHO (in press) - Guidelines for Poisons Control, Part II, section
       6 Geneva, World Health Organization

    Aspects of the evaluation of antidotes

         The development and evaluation of substances to counteract the
    toxic action(s) of a xenobiotic is principally a task for the
    scientific community, particularly those working in experimental
    pharmacology, toxicology and clinical medicine.  The efficacy of a
    substance intended for use as an antidote must first be demonstrated
    in an appropriate animal model.  The next step, demonstration of
    efficacy in humans, it is often more difficult because there is rarely
    an opportunity for controlled clinical trials.  Even if a substance
    is shown to be effective as an antidote, the potential intrinsic
    toxicity of the substance also needs to be considered prior to its
    more widespread use, and, as with all drugs, the possibility of an
    adverse drug reaction should be considered.  A clinician is more
    likely to be prepared to use a relatively "non-toxic" antidote (even
    one whose efficacy has still to be established with certainty) than
    one with intrinsic toxicity.  An antidote which is potentially toxic
    should only be used if it is therapeutically effective and the
    indication for use is clear.  Although possible long-term adverse
    effects and chronic toxicity need to be considered, they are usually
    of less consequence than for an ordinary pharmaceutical agent because
    treatment with an antidote is rarely required more than once in any
    particular individual.  A final consideration in the use of an
    antidote is that increased toxicity should not result from
    mobilization of the toxin from tissue stores or from changes in tissue

    The concept of relative "efficacy" of antidotes

         It is important that clinicians employing antidotes in the
    treatment of poisoned patients recognize that the clinical "efficacy"
    of antidotes varies considerably.  On the one hand there are antidotes
    whose clinical effect is both rapid and dramatic.  Examples would be
    naloxone or flumazenil, which act as very specific competitive
    antagonists at opioid and benzodiazepines receptors, respectively.

         On the other hand, there are antidotes that are able to counter
    only some of the toxic effects of a particular compound; if the dose
    of the compound in question is sufficiently high then the patient is
    likely to die despite the use of an antidote.  Chelating agents
    provide good examples of antidotes that fall into this category of
    efficacy.  Nevertheless, chelating agents have a valuable role to play
    in the treatment of heavy metal poisoning, and many are recommended
    for this purpose in volume V of this series.

         Some agents are loosely termed antidotes even though they may
    have little or no true antidotal effect; they may nonetheless form
    valuable adjuncts to treatment.  Diazepam, used in the treatment of
    organophosphate poisoning (volume IV), is one such example.

    Provisional list of volumes in the IPCS/CEC antidotes series

         It is intended that the IPCS/CEC series of monographs on
    antidotes will cover all antidotes that are commonly employed - or
    which have been proposed for use - in the treatment of human
    poisoning.  Once this aim has been achieved, it is intended that the
    volumes will be periodically updated in order to meet the needs of
    health care professionals.  At present, the proposed volumes for this
    series include:

    Volume 2

    Evaluation of antidotes for cyanide poisoning:

    *    oxygen
    *    sodium thiosulfate
    *    hydroxocobalamin
    *    dicobalt edetate
    *    amyl nitrite
    *    sodium nitrite
    *    4-dimethylaminophenol
    *    antidotes to methaemoglobin-forming agents (methylene blue,
         toluidine blue)
    *    analytical methods for cyanide alone and in combination with
         cyanide antidotes

    Volume 3

    Evaluation of antidotes for paracetamol poisoning

    *    overview
    *    N-acetylcysteine
    *    methionine

    Volume 4

    Evaluation of antidotes for organophosphate poisoning

    *    overview
    *    atropine
    *    diazepam
    *    obidoxime
    *    pralidoxime

    Volume 5

    Evaluation of chelating agents for heavy metal poisoning

    *    overview
    *    deferoxamine
    *    prussian blue
    *    trientine
    *    calcium disodium edetate
    *    DTPA
    *    DMPS
    *    DMSA
    *    dimercaprol
    *    penicillamine and N-acetyl penicillamine

    Volume 6

    Antidotes for methanol and ethylene glycol poisoning.

    Volume 7

    Antidotes for amatoxin, gyrometrine and isoniazid poisoning

    Volume 8

    Evaluation of the various pharmaceutical substances used for enhanced
    elimination and prevention of absorption.

    Further volumes are planned for:

    *    General antidotes and sorbents
    *    Antidotes based on immunotoxicology

    International evaluation process

         Experts are requested by the IPCS to prepare draft monographs on
    specific antidotes or agents, or on specific aspects associated with
    their therapeutic use.  Original literature references must be used
    according to the criteria established for Environmental Health
    Criteria documents.  In order to ensure that monographs are written
    according to agreed standards, a common format has been established
    following the methodology on principles for evaluation of antidotes
    (Appendix II) and the guidelines to authors (Appendix III).  The
    series editors examine the drafts to ensure that they conform to the
    standard format and are of acceptable quality for peer review.  For
    certain volumes a guest editor is also appointed.  The IPCS sends the
    drafts to selected experts for comment and for possible additional

    information.  A working group of authors and experts in the field is
    then convened by the IPCS and CEC.  The task of this group is to:

    (i)     examine the literature referred to in the monographs for its
            relevance, including case data experience;

    (ii)    identify any gaps in knowledge or scientific unknowns;

    (iii)   make an evaluation of the clinical efficacy of the antidote
            for a particular poisoning or pathological condition resulting
            from the poisoning;

    (iv)    provide guidance on the treatment regimens, under various
            conditions of use of the antidote, including, where
            appropriate, field and primary health care use, advise on the
            accompanying supportive care, and give particular attention to
            paediatric doses, contraindications and special

         Following the working group meeting further drafting may need to
    be undertaken by the original author in consultation with the series
    and guest editors.  An overview chapter summarizing the issues and
    giving the evaluation of a series of antidotes for specific types of
    poisoning cases is drafted by the editors or invited experts.  The
    IPCS and CEC may convene a further editorial meeting to finalize the
    monographs for a particular volume and to approve the overview
    chapter.  The volume is then processed by the WHO editor for
    publication by Cambridge University Press.

    2.  NALOXONE

    2.1  Introduction

         Naloxone is an opioid antagonist acting at all three types of
    opioid receptors.  It appears devoid of agonist activity (Martin,
    1976).  Naloxone is indicated in the treatment of opiate poisoning.

         Although naloxone has also been reported to be of benefit as an
    antidote in benzodiazepine (BZD) poisoning (Bell, 1975), other workers
    failed to demonstrate an effect in a double-blind study of
    diazepam-induced sedation (Christensen & Huttel, 1979).  However,
    Jordan (1980) demonstrated some reversal of diazepam-induced
    respiratory depression by naloxone.  Thus there is a need for further
    controlled studies, particularly in cases of poisoning.

         Naloxone has also been claimed to have an effect on
    ethanol-induced central nervous system (CNS) depression, and in one
    study appeared to cause an improvement in 20% of treated cases
    (Jefferys et al., 1980).  However, this finding has not been confirmed
    by other workers (Handal et al., 1983; Nuotto et al., 1984).

         The possible beneficial effects of naloxone in non-opiate
    poisoning probably reflect the involvement of endogenous opioids in
    the depressant action of some non-opioid drugs (McNicholas & Martin,

    2.2  Name and Chemical Formula

    Empirical formula: C19 H21 NO4
    Relative molecular mass: 327
    CAS number:  465-65-6
    Trade names: Narcan, Nalone, Narcanti (Du Pont Pharmaceuticals)

         Naloxone is available for clinical use as the hydrochloride salt,
    which may be anhydrous (CAS-357-08-4) or contain 2 molecules of water
    of hydration (CAS 51481-60-8).  The relative molecular mass of the
    free base is 327.37 and of the anhydrous salt 363.84.

    Conversion table:  1 g = 3.1 mmol
                       1 mmol = 327.4 mg
                       1 mg/ml = 3.1 mmol/l
                       1 mmol/l = 0.33 mg/ml

    The molecular structure of naloxone hydrochloride is shown below.


    2.3  Physico-chemical Properties

         Naloxone hydrochloride has a melting range of 200-205 C.  It is
    soluble in water, dilute acids and strong alkalis, and is slightly
    soluble in alcohol but practically insoluble in ether.  Aqueous
    solutions are acidic (pH 3 to 4.5) (United States Pharmacopeia, 1980)
    and an 8.08% solution in water is isotonic with serum (Hassan et al.,
    1985).  A 25% solution of naloxone hydrochloride rotates light between
    -170 and -181.  Naloxone crystals from ethyl acetate have a specific
    optical rotation at 20 C ([alpha]D20; 9.3 g/l chloroform) of
    -194.5  (Windholz, 1983).

         Naloxone has a pKa (20 C) values for the nitrogen and phenolic
    H groupings of 7.94 and 9.44, respectively (Kaufman et al., 1975).

         On drying at 105 C, the anhydrous form loses not more than 0.5%
    and the hydrated form not more than 11% of its weight.

         The solution for injection is made up in water and should be
    protected from light.  Naloxone can be diluted in 0.9% saline or 5%
    dextrose and should then be used within 24 h.  It should not be mixed
    in solutions containing metasulfite, metabisulfite, or long-chain or
    high relative molecular mass anions, or in those with an alkaline pH.

    2.4  Pharmaceutical Formulation and Synthesis

         Three synthetic routes for the production of naloxone have been
    reported (Hassan et al., 1985).  Oxymorphone is a starting point for
    two of the synthetic processes and 14-hydroxycodeinone for the third.

    Noroxymorphone hydrochloride is a potential impurity from the
    manufacturing process.

    2.5  Analytical Methods

    2.5.1  Quality control

         Naloxone hydrochloride can be assayed by gas chromatography with
    flame ionization detection (United States Pharmacopeia, 1980).

    2.5.2  Identification

         About 150 mg of the unknown substance is dissolved in 25 ml of
    water and a few drops of 6N ammonium hydroxide are added.  Three 5-ml
    portions of chloroform are used for extraction and the extract is
    filtered.  The filtrate is collected, evaporated to dryness using a
    steam bath, and dried at 105 C for one hour.  The infrared absorption
    spectrum of a 1-in-50 solution of the residue obtained in chloroform
    will have maxima at the same wavelengths as those of a similar
    solution of naloxone reference standard.

         The addition of one drop of ferric chloride solution to 1 ml of
    a 1-in-100 solution of naloxone hydrochoride results in a clear
    purplish-blue colour.

    2.5.3  Quantification of the antidote

         Assay methods for naloxone in biological fluids employing
    gas-liquid chromatography (GLC) (Meffin & Smith, 1980),
    radio-immunoassay (RIA) (Berkowitz et al., 1975; Hahn et al., 1983)
    and high-performance liquid chromatography (HPLC) (Asali, 1983; Terry
    et al., 1984) have all been reported.  The GLC method involves
    derivatization and the specific antibody for the RIA is not widely
    available.  The HPLC methods reported appear sensitive and
    reproducible, and are therefore probably the methods of choice.

    2.5.4  Analysis of toxic agents

         In the majority of cases in which naloxone is used as an
    antidote, there is no way of measuring the level of the opioid poison. 
    Present assay techniques for many opiates are difficult, and RIA
    suffers from lack of specificity in many cases.  Some opiates, e.g.,
    morphine, also appear to have active metabolites (Bodd et al., 1990).
    The most widely used method for opioid detection is RIA of urine.

    2.6  Shelf-life

         The shelf-life of naloxone for intravenous injection in temperate
    countries is 3 years and has a similar length in tropical countries.

    2.7  General Properties

         Naloxone is a specific opioid antagonist (Martin, 1976) and it is
    for this reason that it is used in the treatment of poisoning.  There
    are reports that it may reverse the central effects of ethanol and BZD
    poisoning in man.  However, these are experimental uses that remain
    unproven, and any observed effects probably reflect the involvement of
    endogenous opioids in the nonspecific depressant action of those
    agents (McNicholas & Martin, 1984).

    2.8  Animal Studies

    2.8.1  Pharmacodynamics

         Naloxone is a competitive antagonist at opiate receptors, and
    appears to be effective at all three types of receptor (mu, kappa and
    sigma) (Martin, 1976).  It does not produce habituation in animal or
    human models of opiate tolerance and appears to be free of agonist
    activity in most laboratory test models (Jasinski, 1967; McNicholas &
    Martin, 1984).  It produces a parallel shift in the  in vitro dose-
    response effects of pure agonist opioids, such as morphine, and
    partial agonists, such as pentazocine (Smits & Takemori, 1970),
    buprenorphine and dextropropoxyphene.

         Since the range and relative quantities of opioid receptors vary
    in different animal tissues, a range of concentrations of naloxone is
    required to antagonize opioid effects in different test systems. 
    Confusion has arisen as to whether naloxone is a pure antagonist. 
    This is because some opioid receptors act as modulators and enhance
    nociceptive stimuli.  Thus, in some animal models naloxone appears to
    possess agonist effects, but this is in fact incorrect (Sawynok et
    al., 1979).  Naloxone has also been observed in some experiments to
    antagonize the antinociceptive effects of some non-opiate drugs. 
    Again it seems likely that this reflects an involvement of opioid
    receptors in the mechanism of action of these drugs (Sawynok et al.,
    1979). However, in a recent study in rats, Kotlinska & Langwinski
    (1990) failed to find any evidence for the participation of the opioid
    system in the mediation of acute ethanol effects in rats.

         Naloxone has been reported to either decrease or have no
    influence on barbiturate-induced anaesthesia.  This paradox may be a
    result of the dose-response relationship of the effects of naloxone,
    which at high doses may have a potentiating effect (Sawynok et al.,
    1979).  Naloxone has some activity as a GABA antagonist and may thus
    have convulsant activity.  However, this is likely to be at much
    higher concentrations that those encountered clinically (Dingledine et
    al., 1978), since in mice a dose of 100 mg/kg was required to produce

         Naloxone has also been shown to have a number of biochemical
    effects in the rat, including inhibition of lipolysis and a subsequent
    increase in circulating free tryptophan (Badawy et al., 1983).

    2.8.2  Pharmacokinetics

         Naloxone appears to be readily absorbed after oral administration
    but undergoes extensive first-pass hepatic metabolism, which results
    in a very low bioavailability (Misra, 1978).  Studies of the
    pharmacokinetics of intravenous naloxone have been performed in a
    variety of animal species including the rat, rabbit and dog.  Many of
    these studies are based on radio-immunoassay of naloxone.

         The serum concentration of naloxone found 5 min after injection
    was similar (5 mg/kg) in the rat and the dog (Ngai et al., 1976; Pace
    et al., 1979).  The half-life of the parent drug in the rat (30 min)
    was approximately half that in the dog (71 min).

         Ngai et al. (1976) also examined the brain:serum ratio of
    naloxone and found this to vary in the rat between 2.7:1 and 4.6:1.
    Intravenously administered naloxone acts rapidly on the brain.  The
    brain:serum ratio was higher, however, when the naloxone was
    administered subcutaneously.  These workers also studied, in a
    parallel group of animals, the distribution of morphine and noted that
    the brain:serum ration was 1:10.

         The initial distribution of naloxone may account for the rapid
    onset of its reversal of opiate effects when it is given
    intravenously.  The major metabolite of naloxone is the glucuronide.
    Naloxone-3-glucuronide has been found, for example, in the rabbit
    (Fujimoto, 1969).  A conjugated 6-hydroxy product of naloxone,
     N-allyl-14-hydroxy-7,8-dihydronormorphine-3-glucuronide was
    identified in the chicken by Fujimoto (1969); this conjugate was also
    identified in the rabbit by Weinstein et al. (1974) but only in small

         The relatively short action of naloxone appears to result from
    the ease with which it enters the brain after intravenous dosing and
    the subsequent rapid redistribution, elimination and consequent fall
    in brain naloxone levels (Berkowitz, 1976).

         Hydroxylated metabolites of naloxone appear to possess narcotic
    antagonist activities, but their potencies are much weaker than the
    parent compound.  Thus they are unlikely to be of significance in view
    of the small amounts produced (Fujimoto et al., 1975).

         The distribution of naloxone has not been found to be altered by
    a 25-fold range of morphine concentration in the rat (Fishman et al.,

    2.8.3  Toxicology

         Acute toxicity studies with naloxone have been performed in mice,
    rats and dogs.  The LD50 for intravenous administration was 150
    mg/kg in mice, 109 mg/kg in rats and 80 mg/kg in dogs (Social Welfare
    Board, 1976).  For 24-h-old rats the LD50 was 260 mg/kg when given
    subcutaneously (Blumberg et al., 1966).  The maximum nontoxic
    subcutaneous dose in rats was found to be of the order of 50 mg/kg
    (Blumberg et al., 1966).  This dose was tolerated for 24 days, whereas
    200 mg/kg resulted in tremor, convulsions and salivation.

         Daily doses of 0.2 mg/kg given intravenously to dogs for 16 days
    and 5 mg/kg given subcutaneously to monkeys for 30 days caused no
    toxicity.  However, a subcutaneous dose of 20 mg/kg resulted in
    lethargy and tremor in monkeys.

         No teratogenic effects were observed in mice, rats or rabbits
    when naloxone was given parenterally over the period of organogenesis
    (Social Welfare Board, 1976).  No studies on mutagenicity have been

    2.9  Volunteer Studies

         Studies of the pharmacokinetics and pharmacodynamics of naloxone
    have been performed in volunteers.

    2.9.1  Pharmacokinetics

         Using an RIA assay, the pharmacokinetics of naloxone were found
    to fit a two-compartment model, with a rapid distribution phase and a
    slower elimination phase, having a half-life of 64 min (Ngai et al.,
    1976).  More recent studies using HPLC to assay naloxone suggest that
    the apparent volume of distribution, half-life and clearance all show
    differences within groups of normal volunteers.  Thus Aitkenhead et
    al. (1984) reported a mean apparent volume of distribution at steady
    state of 3.65 l/kg (range 1.43-7.05 l/kg) and a mean half-life of
    151.2 min (range 47.1-313.2 min).  Using an HPLC assay, Goldfrank et
    al. (1986) found less variability in patients (half-life 28-55 min).

         The kinetics of naloxone in infants appear similar to those in
    adults (Stile et al., 1984).

         Orally administered radiolabelled naloxone undergoes extensive
    first-pass metabolism in normal subjects (Fishman et al., 1973). 
    After intravenous administration, most (70%) of the radioactivity was
    recovered in urine, the major part of which was conjugated as the
    glucuronide.  In addition other metabolites were found in small
    quantities, i.e. the glucuronide conjugates of 7,8-dihydro-14-hydroxy-
    normorphine, and  N-allyl-7,8-dihydro-14-hydroxy-normorphine
    (Weinstein et al., 1971).

         As a consequence of the high hepatic clearance of naloxone and
    relatively weak agonist activity of its metabolites, it is unlikely
    that dose adjustments would be necessary in cases of renal failure. 
    Naloxone is only 54% protein-bound in adult plasma (61.5% in fetal
    plasma), and this binding is not concentration-dependent over the
    range 9 ng/ml to 2.5 g/ml (Asali & Brown, 1984).  Thus protein-
    binding interactions seem unlikely.

         The elimination of naloxone might be altered in patients with
    liver disease, but no studies appear to have been performed.

    2.9.2  Pharmacodynamics

         Studies have been conducted on the duration of action and potency
    of naloxone in reversing respiratory depression induced by morphine
    (intravenous doses of 5 mg plus 10 mg) in volunteers (Kaufman et al.,
    1981).  The effect of naloxone against this therapeutic dose of
    morphine reached a peak at around 30 min, which was equatable with the
    probable peak in brain concentration.  It should be noted that the
    times of onset and peak effect of naloxone differed.  The duration of
    action of naloxone appeared to be about 1.5 h in this experimental

         Johnstone (1974) examined the effects of an infusion of naloxone
    in volunteers who had received 2 mg/kg morphine intravenously and been
    anaesthetized for 5 h.  Intravenous naloxone given to these volunteers
    at a rate of 40 g/kg over a 10-h period reversed the central
    depressant effects of morphine on respiratory function (measured by
    CO2 responsiveness) and higher functions (assessed by a vigilance
    test).  No tachyphalaxis to the effects of naloxone was observed over
    this period (Johnstone et al., 1974).

         It has been suggested that ethanol may exert some of its effects
    via the endogenous opiate system, as illustrated by the study by
    Jeffferys et al. (1980) and Jeffcoate et al. (1979) where naloxone was
    found to antagonize some of the ethanol effects. However, these
    findings could not be confirmed by Handal et al. (1983) or Nuotto et
    al. (1984).  In the latter study, the effect of naloxone on ethanol-
    induced impairment of psychomotor performance was first studied in two
    placebo-controlled, double-blind, cross-over trials in 17 healthy male
    volunteers.  The main conclusion was that naloxone (intravenous doses
    of 0.4 plus 2 mg) had no significant antagonizing effects on the
    impairment induced by ethanol (1.5 g/kg).  However, a slight but
    significant effect on ethanol-induced nystagmus was noted.  A placebo-
    controlled, double-blind study was subsequently conducted on male
    alcoholics admitted for acute ethanol intoxication (the mean blood
    ethanol level was 2.9 g/l (64 mmol/l)). In this case, neither naloxone
    (intravenous doses of 0.4 plus 2 mg; n=11) nor saline (n=7) had any
    effect, as judged from a clinical inebriation test (Nuotto et al.,

    2.9.3  Effects of high doses of naloxone

         Naloxone has been administered to healthy volunteers at dose
    levels of 0.3-4 mg/kg.  These high dose levels produced dose-dependent
    dysphasia and memory impairment.  In addition, increases in blood
    pressure and respiratory rate were noted, together with increases in
    cortisol and growth hormone levels (Cohen et al., 1983).  These
    findings have been used to support the hypothesis that endogenous
    opioids play a normal regulatory physiological role, but obviously
    have potential therapeutic implications if large doses of naloxone are
    used to treat poisoned patients.

    2.10  Clinical Studies - Clinical Trials

         Naloxone has been investigated in clinical studies on both
    patients who have received a therapeutic dose of an opiate (see
    section 2.9) and those who have been poisoned with opiates.  Since
    naloxone is a competitive antagonist, the dose required to reverse the
    clinical effects of a specific opiate will depend on the dose of the
    opiate, its duration of action, and its pharmacological properties,
    particularly whether it has partial agonist activity or shows
    selectivity at one type of opioid receptor subgroup (Martin, 1976).

    2.10.1  Effects in therapeutic use of opioids

         An alternative method of studying the response to naloxone was
    reported by Drummond et al. (1977). They studied patients who had been
    anaesthetized and had received the synthetic opiate fentanyl. 
    Naloxone produced a dose-dependent increase in respiratory function
    (measured as minute volume or respiratory rate) with intravenous doses
    of 0.1, 0.2 and 0.4 mg.

         Hatano et al. (1975) reported an open study on 80 patients
    undergoing a variety of surgical procedures including cardiopulmonary
    bypass.  Premedication included pethidine (meperidine) and induction
    was achieved with pentazocine and diazepam.  The doses of pentazocine
    in males were 2 mg/kg and females 1.5 mg/kg, and those of diazepam
    were 0.4 and 0.3 mg/kg, respectively.  The authors used a stepwise
    increment of naloxone (0.2-mg intravenous boluses) to achieve reversal
    of the opiate effect of pentazocine at the end of the operative
    procedure and noted a stepwise reversal of the opiate effects in their
    patients as the opiate dose was increased (the average total dose
    given was 2.5 mg/kg body weight).

         The duration of action of naloxone in reversing the effects of
    morphine (5 or 10 mg, intramuscular) in patients recovering from
    surgery is relatively short (Longnecker et al., 1973).  The authors
    suggested that the use of a combination of intravenous and
    intramuscular naloxone might be an appropriate regimen in the post-
    operative situation;  this has also been suggested for the treatment
    of acute overdoses in heroin addicts (see sections 2.12.2 & 2.13.2).

    2.10.2  Effects in acute opioid poisoning

         Two important studies have demonstrated the efficacy of naloxone
    in reversing opiate poisoning.  Evans et al. (1973) reported a study
    in which naloxone (0.4-1.2 mg, intravenous) resulted in recovery of
    consciousness within 1-2 min in nine patients with a history of opiate
    ingestion.  This was associated with improvement in respiratory
    function in the six patients in whom this could be measured with
    minute volume and respiratory rate.  The opiates taken by these
    patients were reported as dipipanone (3), pethidine (2),
    dihydrocodeine (2), pentazocine (1) and heroin (1).  In contrast, none
    of 13 patients overdosed with a variety of other central nervous
    system depressants showed improvement after having been given a total
    intravenous dose of 1.2 mg naloxone.  This rapid and clear benefit of
    therapy was also reported by Buchner et al. (1972), who studied the
    effects of naloxone (0.005 to 0.01 mg/kg) in 10 children with
    methadone poisoning.  Although they did not study a control group,
    they did confirm the presence of methadone in biological fluids in
    some of their patients.  These authors stress the importance of an
    adequate period of observation for patients poisoned with long-acting
    opiates and the necessity of repeated doses of naloxone.

         Since the onset of the effects of naloxone is so rapid, it has
    proved relatively easy to confirm its effectiveness in opiate
    poisoning at restoring consciousness and improving respiration. 
    Further extensive clinical trials in opiate poisoning have, therefore,
    not been performed.

         Henry & Volans (1984) have stressed the importance of classifying
    drugs correctly as opioids.  A list of opioids is a useful reminder
    (Table 1) that agents such as loperamide and diphenoxylate may produce
    significant systemic toxicity in overdose.

         One particular aspect of naloxone use that requires consideration
    is that of the most appropriate dosage regimen.  Early human studies
    confirmed that the duration of action of naloxone was shorter than
    might have been expected from its plasma half-life (Berkowitz et al.,
    1975).  The long duration of action of some opiates is also a factor
    in the need to repeat the initial dose of naloxone in poisoned
    patients (Gober et al., 1979).  As an alternative to repetitive
    dosing, several research workers have suggested that intravenous
    loading doses followed by a steady-state infusion of the drug would be
    appropriate both in children (Gourlay & Coulthard, 1983; Tenenbein,
    1984) and in adults (Bradberry & Raebel, 1981; Goldfrank et al., 1986)
    suffering opiate poisoning.  These regimens have appeared safe and
    effective in clinical use, but do not obviate the need for close
    monitoring during treatment of respiratory function, conscious level
    and cardiovascular function.  It is important to remember that some
    synthetic opioids, e.g., dextropropoxyphene, have been reported to
    produce toxic effects at high doses, which are not reversible by

    naloxone (Barraclough & Lowe, 1982).  These effects may be due to a
    direct action of dextropropoxyphene on cardiac cell membranes.

    Table 1.  Alphabetical list of opioid drugsa


    Alletorphine                    Levorphanol
    Alphaprodine                    Loperamide
    Anileridine                     Meptazinol
    Azidomorphine                   Methadone
    Bezitramide                     Metofoline
    Buprenorphine                   Morphine
    Butorphanol                     Nalbuphine
    Codeine                         Norpipanone
    Dextromoramide                  Opium
    Dextropropoxyphene              Oxycodone
    Diamorphine (Heroin)            Oxymorphone
    Difenoxin                       Papaveretum
    Dihydrocodeine                  Pentazocine
    Diphenoxylate                   Pethidine (Meperidine)
    Dipipanone                      Phenadoxone
    Ethoheptazine                   Phenazocine
    Ethylmorphine                   Phenoperidine
    Etorphine                       Piminodine
    Fentanyl                        Piritramide
    Hydrocodone                     Thebacon
    Hydromorphone                   Tilidate
    Ketobemidone                    Tramadol
    Levomethadyl                    Trimeperidine

    a  From Martindale (1982).  Some of these drugs may be marketed as
       part of a combination preparation.

    2.11  Clinical Studies - Case Reports

         Individual published case reports have confirmed efficacy for the
    majority of opiates (Handal et al., 1983).  In patients who are
    narcotic addicts, naloxone may precipitate features of acute opiate
    withdrawal.  Doses of up to 20 mg naloxone have been used in children
    without associated adverse effects (Handal et al., 1983).

         If patients with acute renal failure are given morphine over
    several days for various reasons (e.g., for sedation while on a
    respirator), opioid toxicity may occur due to accumulation of the
    active metabolite morphine-6-glucuronide, which is renally excreted
    (Bodd et al., 1990).  In such cases, the opioid toxicity may last for
    up to two weeks after the cessation of morphine therapy, and the
    patient will need naloxone infusion in order to avoid respiratory

    2.11.1  Naloxone in clonidine poisoning

         Clonidine hydrochloride is a central and peripheral
    alpha-adrenergic antagonist that is still used in the treatment of
    hypertension. It has also been suggested for the treatment of opiate
    withdrawal (Gold et al., 1980). The mechanism for this effect and for
    the claimed effect of naloxone in some cases of clonidine poisoning
    (North et al., 1981; Kulig et al., 1982) is not clear, but the
    involvement of endogenous opioids has been suggested.  However, the
    effect of naloxone in clonidine poisoning could not be confirmed by
    Banner et al. (1983).  In a retrospective study of 47 consecutive
    children admitted for clonidine poisoning (Wiley et al., 1990), only
    3 out of the 19 given naloxone showed a temporary response. One child
    had an episode of severe hypertension associated with naloxone
    administration (0.1 mg/kg). Thus, there is no clear documentation for
    the beneficial effect of naloxone in clonidine poisoning.

    2.12  Summary of Evaluation

    2.12.1  Indications

         Naloxone has been reported to significantly antagonize acute
    opioid toxicity and opioid effects within anaesthesia. Its high
    therapeutic index and possible beneficial effect in other poisonings
    allow for diagnostic use in critically ill patients when opioid
    poisoning may be a differential diagnosis.

    2.12.2  Advised routes and dose

         In patients with  definite opiate poisoning, naloxone should be
    given by the intravenous route until an improvement in conscious level
    and respiration is observed.  This may involve the administration of
    several milligrams of naloxone if partial opioid agonists are given,
    but 0.8-1.2 mg is usually sufficient in morphine or heroin poisonings. 
    It is important to stress that a pharmacologically active dose of
    naloxone in opiate poisoning may be more than that normally
    recommended in anaesthetic practice.

         In patients with  suspected opiate poisoning, an intravenous
    injection of up to 2 mg naloxone should be administered and the
    patient's response closely monitored.  If there is improvement in
    conscious level, respiratory rate or cardiovascular parameters,
    further doses of naloxone should be administered.  The effect of
    naloxone should be visible within 1 to 2 min after administration.

         Once a patient has regained consciousness, it is necessary to
    continue to monitor respiration and cardiovascular status at regular
    intervals.  In the patient who has taken a large opiate overdose or an
    overdose of a long-acting opiate, it may be necessary to repeat dosing
    with naloxone.  This may be conveniently done by establishing an
    intravenous infusion of naloxone.  A guide to the required dosage has

    been suggested by Goldfrank et al. (1986).  From studies of the
    pharmacokinetics of naloxone in patients suffering opiate poisoning,
    they calculated that an hourly infusion of two-thirds of the dose
    required initially to reverse the effects of the opiate would maintain
    naloxone levels at approximately those present 30 min after the
    initial bolus administration.

         Another approach to opioid poisoning that may sometimes be
    usefully employed in addicts is to give 0.8-1.2 mg naloxone
     intramuscularly before awakening the patient with an intravenous
    naloxone dose of 0.4-0.8 mg (higher doses are rarely needed) (personal
    communication by D. Jacobsen, 1991).  This has been shown to be a
    useful practical approach, since many addicts leave the hospital
    immediately following the effect of the intravenous dose.  Since
    naloxone has a shorter duration of action than the opiate, patients
    are commonly readmitted within one hour with miosis, coma and impaired
    respiration.  This approach to treatment, however, requires adequate
    ventilatary support for the patient because of the short delay before
    the intravenous dose is given.

         Naloxone may also be given as a continuous intravenous infusion
    (about 0.5 mg/h in isotonic saline) to counteract effects of morphine
    metabolites in patients with acute renal failure (Bodd et al., 1990).

    2.12.3  Other consequential or supportive therapy

         Since many of these patients suffer from impaired respiration or
    respiratory arrest, it is extremely important to give oxygen and to
    support ventilation immediately while waiting for naloxone to be
    available for injection. If ventilation is under control and cyanosis
    is regressing, one should consider giving an intramuscular dose of
    naloxone  before the intravenous dose (see section 2.12.2).

         Pulmonary congestion or oedema is occasionally seen in opioid
    (heroin) poisoning. It is usually transient and responds to supportive
    therapy (oxygen and ventilation support) and naloxone.

    2.12.4  Areas where there is insufficient information to make

         There are anecdotal reports of beneficial effect of naloxone in
    other types of acute poisoning, e.g., with ethanol or clonidine. In
    the case of ethanol, these results have not been confirmed in well-
    controlled studies on volunteers or in intoxicated patients (Nuotto et
    al., 1984). The claimed effect in clonidine poisoning has also been
    challenged (Wiley et al., 1990).  There are insufficient data to
    recommend the use of naloxone in poisonings other than those involving

    2.12.5  Proposals for further studies

         Studies of the effect of naloxone in other acute poisonings
    should be encouraged. It could, however, be argued that enough studies
    have been performed on the use of naloxone in ethanol intoxication to
    rule out a possible beneficial effect. On the other hand, there is
    certainly a lack of controlled studies on the possible effect of
    naloxone in clonidine poisoning.

         If effects of naloxone are observed in patients assumed to have
    been poisoned by non-opioids, urine specimens should be collected and
    analysed by RIA for presence of opioids.  Otherwise such "case
    reports" are of little value.

    2.12.6  Adverse effects

         Naloxone possesses a high therapeutic index, but it may provoke
    withdrawal signs and symptoms, e.g., seizures, in (heroin) addicts.
    Other adverse reactions, as described below, are very rarely seen.

         Cardiac arrhythmias and, in particular, ventricular fibrillation
    have resulted from rapid reversal of opiate effects with naloxone. 
    Such events may be a particular problem in patients who have recently
    undergone surgery or those habituated to opiates (Cuss et al., 1984). 
    These reactions may result from a release of sympathetic transmitters,
    since a rise in blood pressure and tachycardia have also been

         Some cases of pulmonary oedema following naloxone use in
    anaesthetic practice have been reported, but it is unclear in this
    situation which is the responsible agent: the anaesthetic, the opiate
    or the antagonist (Partridge & Ward, 1986).

    2.12.7  Restrictions of use

         The fear of provoking withdrawal signs and symptoms should not
    hinder use of naloxone in those who need it clinically.

    2.13 Model Information Sheet

    2.13.1 Uses

         Naloxone is indicated in the management of opiate poisoning, both
    definite and suspected. Opiate poisoning should be considered in
    comatose patients with impaired respiration. Miosis is an unreliable
    sign and is not required for a diagnosis of opioid poisoning.  The
    high wide therapeutic index of naloxone allows its use when a
    diagnosis of opioid poisoning is uncertain.

    2.13.2  Dosage and route

         Since naloxone is a competitive antagonist of opiate poisoning,
    there can be no absolute guidelines on dosage.  Naloxone should be
    given intravenously, in successive doses of 0.4 to 2.0 mg, until the
    desired response has been obtained.  It should be noted that to
    reverse the effects of partial agonists/antagonists, e.g.,
    pentazocine, buprenorphine and dextropropoxyphene, much larger doses
    may be required, and it may prove impossible to reverse the effects of

         Failure to respond to a total dose of 10 mg usually indicates: a)
    that poisoning is not due to opiates; b) that poisoning is due to a
    partial agonist/antagonist; or c) that hypoxic brain damage has
    occurred. It should be noted that dextropropoxyphene has been reported
    to produce cardiac toxicity that is  not reversible by naloxone

         The duration of action of naloxone is short; careful monitoring
    is required and repeated doses may be necessary.  The alternative is
    an intravenous infusion of naloxone.  The use of an hourly infusion of
    two-thirds of the dose of naloxone required to resuscitate the patient
    has been reported to be effective, but dosage should be always
    titrated to the individual patient.

         Another alternative, which may be appropriate for opiate addicts,
    is to give naloxone (0.8-1.2 mg)  intramuscularly before waking the
    patient with an intravenous dose of 0.4-0.8 mg.  However, adequate
    ventilatory support must be given.  The patient then has a "depot" of
    antidote in case he/she departs soon after the initial treatment (as
    many addicts do).

         The dose given to  children should be reduced according to body
    weight (0.01 mg/kg initially).

    2.13.3  Precautions/contraindications

         Naloxone may induce symptoms and signs of acute opiate withdrawal
    in addicts. If seizures occur they are best controlled with diazepam
    (10-30 mg, intravenously).  No dosage alterations seem necessary in

    the case of changes in renal function.  The dose in children should be
    adjusted on a body-weight basis to that used in adults.

         Appropriate protective precautions need to be taken by hospital
    staff in the case of opiate addicts, bearing in mind the risk of
    infection from blood-borne diseases such as hepatitis B and human
    immunodeficiency virus (HIV).

    2.13.4  Adverse effects

         Naloxone has a very high therapeutic index and adverse effects
    are rarely seen. Ventricular arrhythmias including ventricular
    fibrillation have been reported following rapid reversal of severe
    opiate intoxication. This may be avoided if oxygen and adequate
    ventilatory support are also given.  The management of withdrawal
    symptoms in addicts is discussed in section 2.13.3.

    2.13.5 Use in pregnancy and lactation

         Naloxone is not teratogenic in animals, but no relevant human
    data exist.  Naloxone treatment does not appear to be a
    contraindication to breast feeding, although the opiate poisoning
    being treated may itself be a contraindication.

    2.13.6  Storage

         Naloxone for injection should be stored protected from light. 
    Its shelf-life is 3 years.

    2.14  References

    Aitkenhead AR, Derbyshire DR, Pinnock CA, Achola K, & Smith G  (1984) 
    Pharmacokinetics of intravenous naloxone in healthy volunteers. 
    Anaesthesiology, 61: A381.

    Asali LA  (1983)  Determination of naloxone in blood by high
    performance liquid chromatography. J Chromatogr, 278: 329-335.

    Asali LA & Brown KF (1984)  Naloxone protein binding in adult and
    fetal plasma.  Eur J Clin Pharmacol, 27: 459-464.

    Badawy AA-R, Evans M, Punjani NF, & Morgan CJ (1983)  Does naloxone
    always act as an opiate antagonist?  Life Sci, 33(Suppl 1): 739-742.

    Banner W, Lund ME, & Clawson L (1983)  Failure of naloxone to reverse
    clonidine toxic effect. Am J Dis Child, 137: 1170-1171.

    Barraclough CJ & Lowe RA (1982)  Failure of naloxone to reverse the
    cardiotoxicity of Distalgesic overdose. Postgrad Med J, 58: 667-668.

    Bell EF (1975)  The use of naloxone in the treatment of diazepam
    poisoning.  J Paediatr, 87: 803-804.

    Berkowitz BA, Ngai SH, Hempstead J, & Spector S (1975) Disposition of
    naloxone: Use of a new radio-immunoassay.  J Pharm Exp Ther, 195(2):

    Berkowitz BA (1976)  The relationship of pharmacokinetics to
    pharmacological activity: Morphine, methadone and naloxone.  Clin
    Pharmacokinet, 1: 219-230.

    Blumberg H, Wernick T, Dayton HB, Hansen RE, & Rapaport DN  (1966) 
    Toxicological studies on the narcotic antagonist naloxone.  Toxicol
    Appl Pharmacol, 8: 335.

    Bodd E, Jacobsen D, Lund E, Ripel A, Morland J, & Wiik-Larsen E 
    (1990) Morphine-6-glucuronide might mediate the prolonged opioid
    effect of morphine in acute renal failure. Hum Exp Toxicol, 9:

    Bradberry JC & Raebel MA (1981)  Continuous infusion of naloxone in
    the treatment of narcotic overdose.  Drug Intell Clin Pharm, 15:

    Buchner LH, Cimino JA, Raybin HW, & Stewart B (1972)  Naloxone
    reversal of methadone poisoning.  NY State J Med, 72: 2305-2309.

    Christensen KN & Huttel M (1979)  Naloxone does not antagonise
    diazepam-induced sedation.  Anaesthesiology, 51: 187.

    Cohen MR, Cohen RM, Pickar D, Weingartner H, & Murphy DL  (1983)  High
    dose naloxone infusions in normals.  Dose-dependent behavioural,
    hormonal and physiological responses.  Arch Gen Psychiatry, 40:

    Cuss FM, Colao CB, & Baron JH (1984)  Cardiac arrest after reversal
    of effects of opiates with naloxone. Br Med J, 288: 363-364.

    Dingledine R, Inversen LL, & Breuker E (1978)  Naloxone as a GABA
    antagonist: evidence from iontophoretic, receptor binding and
    convulsant studies.  Eur J Pharmacol, 47: 19-27.

    Drummond GB, Davie IT, & Scott DB (1977)  Naloxone: dose-dependent
    antagonism of respiratory depression by fentanyl in anaesthetised
    patients.  Br J Anaesth, 49: 151-154.

    Evans LEJ, Roscoe P, Swainson CP, & Prescott LF (1973)  Treatment of
    drug overdose with naloxone, a specific narcotic antagonist.  Lancet,
    1: 452.

    Fishman J, Roffwarg H, & Hellman, L. (1973)  Disposition of naloxone
    - 7,8,3H in normal and narcotic-dependent men.  J Pharmacol Exp Ther,
    187: 575-580.

    Fishman J, Hahn EF, & Norton BI (1975)  Comparative in vivo
    distribution of opiate agonists and antagonists by means of double
    isotope techniques. Life Sci, 17: 1119-1126.

    Fujimoto JM (1969)  Isolation of two different glucuronide metabolites
    of naloxone from the urine of rabbit and chicken.  J Pharmacol Exp
    Ther, 168: 180-186.

    Fujimoto JM, Roerig S, Wang RIH, Chatterjie N, & Intrirrisi CE (1975) 
    Narcotic antagonist activity of several metabolites of naloxone and
    naltrexone tested in morphine dependent mice (38558).  Proc Soc Exp
    Biol Med, 148: 443-448.

    Gober AE, Kearns GL, Yorkel RA, & Danziger L (1979)  Repeated naloxone
    administration for morphine overdose in a 1 month old infant.
    Paediatrics, 63: 606-608.

    Gold MS, Pottash AC, Sweeney DR, & Kleber HD (1980) Opiate withdrawal
    using clonidine. A safe, effective, and rapid nonopiate treatment. J
    Am Med Assoc, 243: 343-346.

    Goldfrank L, Weisman RS, Errick JK, & Lo MW (1986)  A dosing nomogram
    for continuous infusion intravenous naloxone.  Ann Emergency Med, 15:

    Gourlay GK & Coulthard K (1983)  The role of naloxone infusions in the
    treatment of overdoses of long half-life narcotic agonists.
    Application to nor-methadone.  Br. J Clin Pharmacol., 15: 269-271.

    Handal KA, Schauben JL, & Salamone FR (1983)  Naloxone.  Ann Intern
    Med, 12: 438-445.

    Hahn EF, Lahita R, Kreek J, Duma C, & Intrurrisi CE (1983)  Naloxone
    radio-immunoassay: an improved antiserum.  J Pharm Pharmacol, 35:

    Hassan MMA, Mohammed ME, & Mian MS (1985)  Naloxone hydrochloride. 
    Anal Profiles Drug Subst, 14: 453-489.

    Hatano S, Keane DM, Wade MA, & Sadove MS (1975)  Naloxone reversal for
    anaesthetic dosages of pentazocine. Anaesth Rev, 2: 11-15.

    Henry J & Volans G (1984)  ABC of Poisoning: Analgesics: Opioids.  Br
    Med J, 289: 990-993.

    Jasinski DR, Martin WR, & Haertzen CA (1967)  The human pharmacology
    and abuse potential of N-allyl noroxymorphone (naloxone).  J Pharm Exp
    Ther, 157: 420-426.

    Jeffcoate WJ, Herbert M, Cullen MH, Hastings AG, & Walder CP (1979) 
    Prevention of effects of alcohol intoxication by naloxone. Lancet, 2:

    Jefferys DB, Flanagan RJ, & Volans GN (1980)  Reversal of ethanol-
    induced coma with naloxone.  Lancet, 1: 308-309.

    Johnstone RE, Jobes DR, Kennell EM, Behar MG, & Smith TC (1974) 
    Reversal of morphine anaesthesia with naloxone.  Anaesthesiology, 41:

    Jordan C (1980)  Respiratory depression following diazepam: reversal
    with high-dose naloxone.  Anaesthesiology, 53: 293-298.

    Kaufman RD, Gabathuler ML, & Bellville JW (1981)  Potency, duration of
    action and pA2 in man of intravenous naloxone  measured by reversal of
    morphine-depressed respiration.  J Pharmacol Exp Ther, 219: 156-162.

    Kaufman JJ, Semo NM, & Koski WS (1975)  Microelectrometric titration
    measurements of the pKa's and partition and drug distribution
    coefficients of narcotics and narcotic antagonists and their pH and
    temperature dependence.  J Med Chem, 18: 647-655.

    Kotlinska J & Langwinski R (1990) The lack of effect of opioid
    agonists and antagonists on some acute effects of ethanol. Pol J
    Pharmacol Pharm, 42: 129-135.

    Kulig K, Duffy J, & Rumack BH (1982) Naloxone for treatment of
    clonidine overdose. J Am Med Assoc, 247: 1697.

    Longnecker DD, Grazis PA, & Eggors WWN (1973)  Naloxone for antagonism
    of morphine-induced respiratory depression.  Anaesth Analg, 53:

    McNicholas LF & Martin WR (1984)  New and experimental therapeutic
    roles for naloxone and related opioid antagonists.  Drugs, 27: 98-93.

    Martin WR (1976)  Naloxone. Ann Intern Med, 85: 765-768.

    Martindale (1982) In: Reznolds JEF ed.  The extra pharmacopoeia, 28th
    ed.  London, Pharmaceutical Press.

    Meffin PF & Smith KJ (1980)  Gas chromatographic analysis of naloxone
    in biological fluids. J Chromatogr, 183: 352-356.

    Misra AL (1978)  Metabolism of opiates.  In: Factors affecting the
    action of narcotics.  New York, Raven Press, pp 297-343.

    Ngai SH, Berkowitz BA, Yang JC, Hampstead J, & Spector S (1976) 
    Pharmacokinetics of naloxone in rats and in man: Basis for its potency
    and short duration of action.  Anaesthesiology, 44: 398-401.

    North DS, Wieland MJ, & Peterson CD (1981) Naloxone administration in
    clonidine overdosage. Ann Emergency Med, 10: 397.

    Nuotto E, Palva ES, & Seppala T (1984) Naloxone-ethanol interaction in
    experimental and clinical situations. Acta Pharmacol Toxicol, 54:

    Pace NL, Parrish RG, Lieberman MM, Wong KC, & Blatnick RA (1979) 
    Pharmacokinetics of naloxone and naltrexone in the dog.  J Pharmacol
    Exp Ther, 208: 254-256.

    Partridge BL & Ward CF (1986)  Pulmonary edema following low-dose
    naloxone administration. Anesthesiology, 65: 709-710.

    Sawynok J, Pinsky C, & Labella FS (1979)  Mini review on the
    specificity of naloxone as an opiate antagonist.  Life Sci, 25:

    Smits SE & Takemori AE (1970)  Quantitative studies on the antagonism
    by naloxone of some narcotic and narcotic-antagonist analgesics.  Br
    J Pharmacol, 39: 627-638.

    Social Welfare Board (1976)  Nalone (naloxon). Uppsala, Sweden, Social
    Welfare Board, Pharmaceuticals Department, pp 7-9.

    Stile IL, Fort M, Marotta F, Wurzburger R, Hiatt IM, & Hegyi T (1984) 
    Pharmacokinetics of naloxone in premature infants.  Paediatr Res,
    18(392): 161A.

    Tenenbein M (1984)  Continuous naloxone infusion for opiate poisoning
    in infancy.  J Pediatr, 105: 645-647.

    Terry MD, Hisayasu GH, Kern JW, & Cohen JL (1984)  High performance
    liquid chromatographic analysis of naloxone in human serum.  J
    Chromatogr, 311: 213-217.

    United States Pharmacopeia (1980) 20th ed.  Rockville, Maryland,
    United States Pharmacopeial Convention, Inc.

    Weinstein SH, Pfeffer M, & Schor JM (1974)  Metabolism and
    pharmacokinetics of naloxone.  Adv Biochem Psychopharmacol, 8:

    Weinstein SH, Pfeffer M, Schor JM, Indindoli L, & Mintz M (1971) 
    Metabolites of naloxone in human urine.  J Pharm Sci, 60: 1567-1568.

    Wiley JF, Wiley CC, Torrey SB, & Henretig FM (1990) Clonidine
    poisoning in young children. J Pediatr, 116: 654-658.

    Windholz M ed.  (1983)  The Merck index: An encyclopedia of chemicals,
    drugs and biologicals, 10th ed.  Rahway, New Jersey, Merck and Co,


    3.1 Introduction

         Acute poisoning is currently one of the main causes of hospital
    admission in developed countries.  Benzodiazepines (BZDs) are the most
    commonly used drugs throughout the world and their abuse may be
    responsible for the impairment of memory and for dependence.  An acute
    overdose can result in long-lasting coma, which is generally treated
    with supportive measures. Flumazenil, an imidazobenzo-diazepine
    (AnexateTM), has been shown to reverse the sedative, anti-
    convulsant, and muscle-relaxant effects of BDZs.  It has no convulsive
    action in itself and its use has therefore been proposed to counteract
    benzodiazepine action in anaesthetics, clinical toxicology and
    intensive care.

    3.2  Name and Chemical Formula of Antidote

    *    Flumazenil AnexateR (Roche Laboratories)
    *    Ethyl-8-fluoro-5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a][1,4]
    *    Empirical formula: C15H14O3N3F
    *    Relative molecular mass: 303.3
    *    Therapeutic class : Imidazobenzodiazepine
    *    CAS number: 78 755-81-4
    *    Conversions:1 mmol   =    303.3 mg
                     1 g      =    3.3 mmol
                     mol/l   =    3.3 x g/ml
                     g/ml    =    0.3 x mol/l

    3.3  Physico-chemical Properties

         Physico-chemical properties of flumazenil are given in Table 1.

         Flumazenil remains stable when exposed to light and when stored
    for 2 years at 35 C.  The loss of weight on drying is up to 1%.

    3.4  Pharmaceutical Formulation and Synthesis

         No information is available on the routes of synthesis and

         Flumazenil is supplied for parenteral administration in vials
    containing 5 or 10 ml aqueous solution (0.1 mg/ml).  It is available
    for oral administration as tablets of 10, 20 or 30 mg.

    Table 1.  Physico-chemical properties of flumazenila


    Melting point                                      198-202 C

    Solubility in water                                   < 1 g/l

    Solubility in organic solvents (g/l)
        chloroform                                          < 250
        methanol                                            <  17
        ethyl acetate                                       <   3
        diethyl ether                                       <   1

    Solubility at various pH values (g/l)
        (in aqueous buffered solution at 37 C)
        pH 1.2                                                  3
        pH 5.3                                                0.7
        pH 7.5                                                0.6

    Acidity (10% aqueous solution)                        4.5-7.5

    pKa (in weak base)                                        1.7

    a  Personal communication from Roche Laboratories (1988)

    3.5  Analytical Methods

    3.5.1  Identification of the antidote

         Information on the identification of flumazenil was provided by
    Roche Laboratories (personal communication, 1988).  Infrared spectroscopy

         The infrared spectrum (625-4000 cm-1) of a sample (a 1:300
    solid dispersion in potassium bromide) is compared qualitatively with
    that of a reference substance.  Ultraviolet absorption

         A portion (85-95 mg) of the sample is dissolved in approximately
    100 ml of ethanol and diluted to 150 ml with ethanol (solution 1). 
    This solution is then diluted ten-fold with ethanol to give solution
    2, which is further diluted ten-fold with ethanol to give solution 3.
    The position and absorbance of solution 3 is measured
    spectrophotometrically at the maximum (245 nm) and minimum (228 nm)
    wavelengths, against ethanol, in quartz cells.  Thin-layer chromatography

         The TLC details are as follows:

         *     layer:  Silica gel 60 F254

         *     mobile phase:  Chloroform/ethanol (90/10 v/v)

         *     sample solution: 10 ml of the ampoule solution is extracted
               with 1 ml of chloroform

         *     standard solution: 5 mg of flumazenil is dissolved in 5 ml
               of chloroform saturated with water

         *     front distance: 12 cm

         *     migration time approx: 30 min

         *     detection: the plate is dried in a current of warm air for
               5 min, and examined under shortwave light.  Decreasing
               fluorescence due to flumazenil occurs at 254 nm
               (ultraviolet region). When the plate is sprayed with
               Dragendorff's reagent, flumazenil appears as an orange
               spot.  The Rf value is approximately 0.5.

    3.5.2  Quantification of the antidote in biological samples

         The determination of flumazenil in plasma by gas-liquid
    chromatography (GLC) with nitrogen phosphorus detection is a sensitive
    and specific method, the detection limit being 3 ng/ml (Abernethy et
    al., 1983).  An ethyl acetate extraction (neutral pH) of 0.1-3 ml
    plasma is used for sample preparation.  When methylclonazepam is used
    as an internal standard, the graph is linear for plasma concentrations
    up to 200 ng/ml.  The retention time for flumazenil is 3.96 min.

         High-performance liquid chromatography (HPLC) with UV detection
    at 254 nm is a sensitive method for determination in urine or plasma,
    the detection limit being about 10 ng/ml (Timm & Zell, 1983; Bun et
    al., 1989).  When the  n-propyl ester analogue is used as an internal
    standard, the graph is linear for plasma concentrations up to 320

    3.5.3  Analysis of the toxic agent in biological samples

         Three major methods for the quantitative analysis of BZDs in
    plasma or serum are used:

    *    HPLC with UV detection at 246 nm (detection limits are 5-50 g/ml
         of serum) (Rocher, 1984);

    *    immunoenzymology by the EMIT method for a semiquantitative
         determination (metabolites also measured) of diazepam levels,
         completed by a chromatographic method (sensitivity from 0.3 to 2
         g/ml) (Rocher, 1984);

    *    gas-liquid chromatography (Pellerin, 1986).

    3.6  Shelf-life

         Vials ready for use are stable at room temperature (15-25 C) for
    three years.

    3.7  General Properties

         Flumazenil has been shown to block all the typical BZD effects
    (anticonvulsive, sedative, anxiolytic, muscle relaxant, and amnesic). 
    It acts as a potent BZD-specific antagonist by competing at the
    central synaptic gamma-aminobutyric acid (GABA) receptor sites in a
    dose-dependent manner, but does not seem to antagonise BZD effects at
    peripheral GABA-ergic (renal, cardiac, etc.) receptor sites (Mohler et
    al., 1981).  It possesses agonist properties and has a specific, but
    discreet, anticonvulsive effect without inducing drowsiness or muscle
    relaxation (Abernethy et al., 1983; Timm & Zell, 1983; Haaefely, 1983;
    Rocher, 1984; Scollo-Lavizzari, 1984; personal communication by Roche
    Laboratories, 1988).  In addition, it antagonizes the sedative effects
    of other compounds that act through GABA receptors, such as zopiclone
    (Mohler et al., 1981).

    3.8  Animal Studiesc

    3.8.1  Pharmacodynamics

         Flumazenil has been tested for its ability to induce withdrawal
    signs in animals pretreated with benzodiazepine; the signs included
    emesis, tremors, rigidity and clonic convulsions.


    c  Personal communication by Roche Laboratories to the IPCS, 1988

         Rats that had been pretreated with an oral dose (10 or 100 mg/kg)
    of diazepam for 12 days were administered flumazenil (10 mg/kg)
    intravenously.  Signs were very mild even at 100 mg/kg.

         Cats were pretreated intraperitoneally for 16 days with either a
    10-mg/kg dose of lorazepam twice daily or a 1-mg/kg dose of triazolam
    once daily. Flumazenil (100 mg/kg) was then administered
    intraperitoneally either immediately or 1.5, 6, 12, 48, and 60 h after
    the last dose.  Symptoms such as rigidity, vocalization and tachypnoea
    lasted 30 min, whereas others such as hypersalivation lasted 2 h.

         Flumazenil (1 to 15 mg/kg) was administered intragastrically to
    rats that had been pretreated with daily diazepam doses of 113 mg/kg
    for about 6 months. Abstinence syndromes increased with increasing
    dose of flumazenil and reached a plateau.

         The intragastric administration of flumazenil (15 mg/kg per day)
    to cats pretreated with flurazepam (5 mg/kg per day) for 35 days led
    to withdrawal symptoms (increasing muscle tone, tremors, piloerection,
    mydriasis, and hypersalivation) 24 h after the last dose of
    flurazepam.  No convulsions were observed.

         Intramuscular administration of flumazenil (5 mg/kg) to squirrel
    monkeys and baboons, pretreated with oral doses of lorazepam,
    triazolam (3 mg/kg per day), oxazepam (40 or 80 mg/kg per day) or
    diazepam (8-20 mg/kg per day), produced withdrawal signs.  However, no
    withdrawal signs were precipitated by flumazenil in monkeys treated
    with oral midazolam (30 mg/kg) or in barbital-dependent rhesus monkeys
    (the length of pretreatment with BZD was not specified).

         The severity of withdrawal signs resulting from the blocking of
    BZD receptors by flumazenil depends on the species tested, the dose of
    BDZ used to develop physiological dependence, and the duration of

    3.8.2  Pharmacokinetics  Absorption

         A single dose of flumazenil (125 mg/kg) in a carboxymethyl
    cellulose suspension produced a maximum plasma concentration in rats
    of 9.9 g/ml after 20 min.  The bioavailability was 0.55.  In rabbits,
    the maximum concentration 90 min after a single dose of flumazenil
    (150 mg/kg) was 15 g/ml.  The bioavailability was 0.60.  Distribution

         When total radioactivity was measured in rats 0.5, 7, 24, 96, and
    192 h after an intravenous dose of 14C-labelled flumazenil (2
    mg/kg), the highest level was found at 0.5 h in the kidney, liver and

    intestine.  None was found at 192 h.  The volume of distribution
    ranged from 0.71 to 1.87 l/kg.  Elimination

         Studies on rats given an oral dose (50 mg/kg) of 14C-labelled
    flumazenil and on dogs given an intravenous dose of 4 mg/kg showed
    three main inactive metabolites:

    *    Ro 15-3890 acid and major metabolite (72% in the rat, 30-60% in
         the dog);

    *    Ro 15-4965 hydroxyethyl derivative (3% in the rat);

    *    Ro 15-6877  N-demethyl derivative (1% in the rat, 1-13% in the

         Table 2 presents elimination data in three different species.  In
    these species, 90% of the intravenously or orally administered
    flumazenil was eliminated, mainly as metabolites, within 48 h.  One
    third was eliminated in the faeces and two-thirds in the urine.

    Table 2.  Elimination of flumazenil in the rat, rabbit and dog


    Species     Dose           Total plasma          T
                                clearance           (min)
                             (ml/min per kg)

    Rat         2 mg/kg            114                7.4

    Rabbit      0.5 mg/kg           24               34

    Dog         5 mg                21               48

    3.8.3  Toxicology  Acute toxicity

    a)   Intravenous administration to rats and mice

         An aqueous solution of 0.1 mg flumazenil/ml was used and was
    administered at a dose of 2.5 mg/kg to mice and 1 mg/kg to rats.  No
    abnormal clinical signs and no deaths occurred.  LD50 values were
    not determined; these doses (50 to 250 times higher than the clinical
    doses) were well tolerated in the two species.

         A flumazenil solution with a concentration of 50 mg/ml was
    subsequently used and the LD50 values given in Table 3 were obtained
    (95% confidence intervals).  Deaths occurred 30 min after the
    injection, preceded by rigidity and clonic convulsions.

    Table 3.  Intravenous LD50 values (mg/kg) for the mouse and rat


    Species         Male            Female

    Mouse          143-198          145-175

    Rat             85-167          112-231

    b)   Intravenous administration to the dog

         The administration of daily doses of 0.01 to 0.03 mg/kg was well
    tolerated and no deaths were observed.  LD50 values were not
    determined; the doses (15 to 30 times higher than the clinical doses)
    were again well tolerated.

    c)   LD50 values (mg/kg) for the rat, mouse and rabbit

         When flumazenil was administered orally to rats, mice and rabbits
    (Table 4), deaths were observed within three days, associated with
    decreased motor activity, catatonic state and tremors.

    Table 4.  LD50 values (mg/kg) for the rat, mouse and rabbit


    Species           Male             Female

    Mouse             2500             1300

    Rat               4200             4200

    Rabbit            2000             2000
                                                Subacute toxicity

         Systemic tolerance was good in both rats and dogs administered
    flumazenil intravenously at dosages up to 10 mg/kg per day for 4
    weeks.  Chronic toxicity

         In 13-week studies using an oral aqueous solution of flumazenil,
    very good tolerance was shown by rats at dosages of 0.5, 25 and 125
    mg/kg per day and by dogs at 0.5, 20, and 80 mg/kg per day.  No
    haematological, biochemical or gross pathological abnormalities were
    observed.  Embryotoxicity

         Studies on rats (between the 7th and 16th day of gestation) and
    rabbits (between the 7th and 19th day of gestation) revealed no signs
    of embryotoxicity at dosages of 15, 50, and 150 mg/kg per day.  Mutagenicity

         Flumazenil was not mutagenic in the Ames test or micronucleus
    test, or in tests using  Saccharomyces cerevisiae or Chinese hamster
    V79 cells.

    3.9  Volunteer Studies

    3.9.1  Pharmacodynamics  BZD antagonist effect

         Efficacy studies were performed on 125 healthy volunteers with
    oral doses of flumazenil up to 20 mg, the aim being to antagonize the
    effects of diazepam, flunitrazepam and midazolam on the CNS (Darragh,
    1981; Lupolover, 1983).  These studies demonstrated the antagonist
    effect of flumazenil, which rapidly abolished the hypnotic-sedative
    BZD effects.  Other studies used meclonazepam (Darragh et al., 1981),
    diazepam (Darragh et al., 1982), flunitrazepam (Gaillard & Blois,
    1983) and midazolam (Forster et al., 1983).  In studies by Ziegler &
    Schalch (1983) and Lauven et al. (1985), flumazenil was administered
    to subjects during continuous midazolam infusion after the attainment
    of a pharmacokinetic and pharmacodynamic steady state, at which point
    subjects were deeply asleep.  The degree and duration of the effect of
    flumazenil depended on the BZD dose, the antagonist dose and the time
    that had elapsed since the BZD was given.  In the study by Ziegler &
    Schalch (1983), baseline levels of vigilance and orientation were
    reached within 1 min.  Lauven et al. (1985) used higher midazolam and
    flumazenil dosages and his patients awoke within 28 to 48 seconds.  No
    signs of BZD withdrawal effects were seen in short-term studies (one
    single dose) on healthy volunteers given flumazenil to antagonize BZDs
    (Amrein, 1987).

         The efficacy of flumazenil in antagonizing the effects of
    midazolam was also clearly demonstrated in the double-blind placebo-
    controlled study by Rouiller et al. (1987).  Intrinsic effects

         Most studies on healthy human volunteers have shown little or no
    intrinsic effect of flumazenil when administered alone.  The mild
    sedation reported by Amrein (1987) occurred after the administration
    of oral doses greater than 100 mg.

         Scollo-Lavizzari (1984) observed some anticonvulsant effects in
    epileptic patients.  Decreased amplitude of auditory evoked potentials
    has also been described (Laurian et al., 1984; Schoepf et al., 1984). 
    Mild, nonspecific effects such as increased alertness may occur after
    the administration of doses very much higher than those used
    clinically (Laurian et al., 1987).

    3.9.2  Pharmacokinetics  Absorption

         Following oral administration of a 200-mg dose of flumazenil, the
    highest plasma concentration (Cmax) ranged from 147 to 349 g/l and
    was reached within 20 to 45 min.  The mean bioavailability of the
    tablets used was about 17% and the inter-individual variability was
    7-29% (Pellerin, 1986).  Distribution

         The proportion of flumazenil bound to plasma proteins is 50%
    (two-thirds of which is bound to albumin).  Values for the mean
    steady-state volume of distribution of 0.95 l/kg (personal
    communication by Roche Laboratories, 1988) and 1.23 l/kg (Roncari et
    al., 1986) have been determined.  Elimination

         Ninety-nine per cent of the flumazenil administered is
    metabolized by the liver, and 1% is excreted unchanged in the urine. 
    Mean total blood clearance, for which values of 59 l/h (Pellerin,
    1986) and 72 l/h (Roncari et al., 1986) have been determined, is
    essentially due to the hepatic clearance.  The apparent plasma half-
    life in healthy volunteers has been reported to be 53-58 min (Roncari
    et al., 1986; personal communication by Roche Laboratories, 1988).

    3.9.3  Tolerance of flumazenil

         In the study by Rouiller et al. (1987), no objective agonist
    effects or biological toxicity of flumazenil could be demonstrated in
    six healthy volunteers.

    3.9.4  Other studies

         There is evidence that central nervous system effects of ethanol
    are mediated through the GABA system.  For this reason, the effect of
    flumazenil on psychometric performance was studied in eight healthy
    volunteers with stable blood ethanol levels of 1.6 g/l (35 mmol/l)
    under a placebo-controlled double-blind design (Clausen et al., 1990). 
    Flumazenil did not improve psychomotor functions in these ethanol-
    intoxicated subjects, which is in agreement with experience in
    clinical toxicology.

    3.10  Clinical Studies - Clinical Trials

         Flumazenil was first used clinically in patients with iatrogenic
    benzodiazepine overdose due to mechanical ventilation or status
    epilepticus (Scollo-Lavizzari, 1983).

         Clinical studies can be grouped under the headings
    anaesthesiology and toxicology (Amrein, 1986).

    3.10.1 Anaesthesiology  General anaesthesia

         Three placebo-controlled studies have been conducted in patients
    who were given flunitrazepam for general anaesthesia.

         Jensen et al. (1985) reported that a 0.3-mg to 0.7-mg dose of
    flumazenil awoke all patients within 5 min, compared with only 35% of
    the patients in the placebo-treated group (P < 0.001 for sedation,
    orientation and amnesia).

         In a study of 60 patients, Tolksdorf et al. (1986) found that
    patients treated with flumazenil were less sedated than placebo-
    treated patients (P < 0.05) following flunitrazepam sedation (from 5
    min to 1 h after the administration of flumazenil), better orientated
    at 15 min, and less amnesic.  Ellmauer et al. (1986) reported similar
    results in 57 patients given a 0.1- to 1-mg dose of flunitrazepam (P
    < 0.005).

         No significant difference was observed after 2 h between the
    placebo-treated and flumazenil-treated patients in any of the three
    studies described in this section.

         Midazolam effects were reversed by flumazenil in an open study
    including 18 intracranial surgery patients (Chiolero et al., 1988). sedation

         In a 74-patient open study (Geller et al., 1986) and a 40-patient
    placebo-controlled study (Knudsen et al., 1986), in which either
    midazolam or diazepam was used, there was a significant difference
    between flumazenil- and placebo-treated patients.  In the former
    study, patients were awakened by a 0.1- to 0.6-mg dose of flumazenil
    within 1 to 2 min.  In the study by Knudsen et al. (1986), 80% of the
    flumazenil-treated patients were awake 5 min after receiving the dose
    compared with 50% in the placebo group (P < 0.05).

    3.10.2  Benzodiazepine overdose or intoxication

         Three different studies have indicated that flumazenil may be an
    effective tool for the management of intoxication (either intentional
    or iatrogenic) with BZD in the presence or absence of other agents. 
    Owing to its safety and specificity, flumazenil could be used in the
    initial treatment of poisoning and coma of unknown origin.  In a study
    by Hofer & Scollo-Lavizzari (1985) based on 13 patients, a 1.5-to 10-
    mg dose of flumazenil administered intravenously at a rate of 1.5 to
    2.5 mg/min reversed the CNS depression induced by various BZDs within
    1 to 2 min.

         Geller et al. (1985) treated 34 patients (23 cases of intentional
    drug intoxication and 11 of iatrogenic BZD overdose) by means of
    intravenous injections of 0.1 mg flumazenil every 30 seconds until the
    patient regained consciousness.  The treatment proved to be extremely
    effective, providing reversal effects lasting up to 2 h.

         Bismuth et al. (1985) treated patients for BZD overdose in a
    double-blind randomized study, injecting a single dose of either
    flumazenil or placebo.  Two of the 20 placebo patients awoke
    partially, compared with 17 of the 20 flumazenil-treated patients (one
    experienced seizures interrupting the study).  In a second open study
    (Bismuth et al., 1986) based on 37 patients, 6 showed no response to
    doses of flumazenil ranging from 5 to 9.5 mg (mixed intoxication), 11
    showed partial awakening (no possible written response) at a dose of
    2.1  1.6 mg (mixed intoxication), and 20 were completely awakened by
    a dose of 1.4  0.7 mg.  The awakening was only temporary and return
    to coma occurred after an interval of 15 min to 5 h.  Permanent
    recovery occurred in a patient suffering intoxication due to
    triazolam, a BZD with a short half-life, after a single administration
    of flumazenil.

         More recent placebo-controlled double-blind studies have
    confirmed the beneficial effect of flumazenil in cases of BZD
    poisoning (Aarseth et al., 1988; Ritz et al., 1990).

    3.11  Clinical Studies - Case Reports

         The many controlled clinical studies of the effect of flumazenil
    limit the need for information from case reports.  In the clinical
    studies reported, there have been few adverse effects associated with

    the use of flumazenil.  There have, however, been case reports of
    seizures followed by ventricular tachycardia associated with the use
    of flumazenil in combined poisonings with cyclic antidepressants and
    BZD (Bismuth et al., 1985).

         In one report, death was claimed to have been associated with
    flumazenil administration in an old, obese and anaemic woman who had
    been sedated with midazolam (4 mg, intravenous) prior to gastroscopy
    (Lim, 1989).  During the investigation she suffered cardiac arrest;
    flumazenil was given promptly and she recovered temporarily, but then
    gradually deteriorated and died 16 h later. According to Birch &
    Miller (1990), the death of this patient was probably not related to
    flumazenil administration.

         Recently, successful treatment was achieved by administering
    flumazenil as an intravenous bolus (0.02 mg/kg) and then as a
    maintenance dose of 0.05 mg/kg per h to a newborn baby with recurrent
    apnoea due to BZDs taken by his mother (Richard et al., 1991).

         The benefit from the diagnostic use of flumazenil in coma of
    unknown origin has been reported in two recent cases (Burkhart &
    Kulig, 1990).  When flumazenil is used with caution in such
    situations, time may be saved and further expensive diagnostic
    procedures, e.g., cerebral computerized tomographic (CT) scan,

    3.12  Summary of Evaluation

         Flumazenil appears to be a antagonist to BZDs and other GABA-
    ergic agents.  This antagonism, following intravenous injection, has
    been reported to be sensitive in cases of intoxication resulting
    solely from BZDs (the reversal of BZD effects being observed with
    doses of less than 2 mg), rapid in onset (within 2 min), and short-
    lived (effects last for less than 30 min).

    3.12.1  Indications

         In controlled clinical trials, flumazenil significantly
    antagonizes BZD-induced coma arising from anaesthesia or acute
    overdose.  However, the use of flumazenil has not been shown to reduce
    mortality or sequelae in such cases.  As the mortality in pure BZD
    poisoning is extremely low, studies with mortality as end-point are
    impractical since a reasonable level of statistical significance could
    probably never be obtained.  However, in cases of mixed intoxication,
    especially with ethanol and triazolam/flunitrazepam, the use of
    flumazenil may be life-saving due to the poten-tiation of BZD toxicity
    by ethanol.  Given this situation, it is obvious that the routine use
    of flumazenil in BZD poisoning is not indicated and that
    recommendations for its use in clinical toxicology must be based on
    pragmatic considerations made by clinicians experienced in treating

    these patients.  Flumazenil is a relatively expensive drug and this
    may also influence its use, especially in areas with limited

         The use of flumazenil in BZD poisoning should, therefore, only be
    advocated in situations with complications, which are rarely seen
    except in cases of mixed ingestion.  Although not of life-saving
    significance, it also seems reasonable to advocate the use of
    flumazenil if intubations (before gastric lavage) and mechanical
    ventilation can thereby be avoided (see section 3.13.1).  The proposed
    uses of flumazenil within acute medicine and anaesthesia are listed in
    section 3.13.1.  Acute poisoning is always an important differential
    diagnosis in cases of coma in children and young adults.  The
    diagnostic use of flumazenil in such cases can be justified by its
    high therapeutic index and the fact that this may limit the use of
    other diagnostic procedures such as cerebral CT scan, clinical
    chemistry analyses and even lumbar puncture.

    3.12.2  Dosage and route

         Flumazenil is available for intravenous and oral administration. 
    The need for the latter formulation may be questioned in view of the
    fact that drugs should generally be given intravenously in the
    emergency situation and the bioavailability is low and variable.  Thus
    the intravenous route is preferable.  Doses need to be adjusted
    according to individual clinical response, bearing in mind the very
    high therapeutic index of flumazenil.

    a)   In anaesthetics and in intensive care, doses of 0.2-0.5 mg should
         be used to reduce sedation and doses of 0.5-1 mg to abolish other
         BZD effects (Amrein, 1987).

    b)   In cases of BZD overdosage, single doses of 0.3-1 mg can be given
         and repeated as necessary.  If there is no clinical response to
         2 mg flumazenil given over a period of 5-10 minutes, diagnoses
         other than BZD poisoning are likely.  It is also possible to
         administer a continuous infusion (0.3-1 mg/h) of flumazenil
         (diluted in 0.9% sodium chloride solution or 5% glucose solution)
         in patients relapsing into a coma and/or respiratory depression
         following an initial effect of flumazenil injection.

    c)   In children, experience is limited and dosage regimens less well
         documented (Lheureux & Askenasi, 1988; Wood  et al., 1987).  It
         is suggested that intravenous doses of 0.1 mg should be given
         once per minute until the child is awake.  It may be necessary to
         give a subsequent continuous intravenous infusion at a rate of
         0.1 to 0.2 mg/h.

    3.12.3  Other consequential or supportive therapy

         Treatment with flumazenil requires continuous intensive
    observation. After the administration of a single dose of flumazenil,
    the patient must be observed for at least 2 h to be certain that BZD-
    induced complications will not recur.  The termination of continuous
    infusion requires intensive care monitoring.

    3.12.4  Areas where there is insufficient information to make

         There is insufficient information to make recommendations in the
    case of hepatic encephalopathy (indication is based on the hypothesis
    that hepatic encephalopathy is associated with increased GABA-mediated
    inhibitory neurotransmission).

    3.12.5  Proposals for further study

         The use and dosages of flumazenil in children require further
    study.  Indications for utilization of oral preparations need to be
    clarified.  The use in coma of unknown origin merits further studies.

    3.12.6  Adverse effects

         The most frequent adverse effects have been reviewed by Amrein
    (1987).  When flumazenil is used in anaesthesia, the main adverse
    effects that have been reported are nausea and vomiting (placebo:
    7.5%; < 1 mg flumazenil: 12.1%; 1-10 mg flumazenil: 24.5%).  Other
    adverse effects, which have been reported in less than 5% of cases,
    are tremor, involuntary movements, dizziness, agitation, discomfort,
    tears, anxiety, and a sensation of cold.

         Minor effects occur when flumazenil is used in intensive care,
    where agitation is the commonest adverse effect (10%).  When it is
    administered to patients showing BZD habituation, the following
    features occur: anxiety, tenseness, fear, agitation, confusion,
    convulsions (Marchant & Wray, 1989) and myoclonic seizures.  Their
    frequency and intensity depend on the degree of dependency and they
    are believed to be related to some sort of BZD abstinence syndrome.

         When administered rapidly, flumazenil can cause hypertension,
    tachycardia and acute anxiety.  This equivalent of an "exercise test"
    was observed with the 1 mg/ml solution, which is no longer used.

    3.12.7  Restrictions of use

         In certain circumstances, BZD antagonism by flumazenil may be

    a)   An acute withdrawal syndrome can occur in patients showing BZD
         habituation following therapy or abuse.

    b)   Convulsions can occur in cases of mixed drug overdosage where BZD
         has been taken with a drug liable to cause convulsions (such as
         a tricyclic antidepressive agent).

    c)   Convulsions can be induced in patients treated with BZD for
         seizure disorders or in patients who for years have been using
         BZD for sleep disturbances.

         There are other limitations to the use of flumazenil.

    a)   It has a short-lived effect and repeated injection or continuous
         infusion is often necessary unless a short-acting BZD (e.g.,
         triazolam) has been ingested.

    b)   In cases of mixed drug overdosage, the patient may remain
         unresponsive when other drugs are contributing to the coma.

    c)   The treatment costs are high and supportive treatment may be

    3.13  Model Information Sheet

    3.13.1 Uses

         Flumazenil is a specific antagonist of the effects of BZD at
    central GABA-ergic receptors.

         Within the domains of intensive care and anaesthesia, flumazenil
    may be valuable in the following circumstances:

    a)   to diagnose BZD-induced unconsciousness in patients presenting
         coma of unknown origin;

    b)   to terminate long-term BZD-induced sedation in the intensive care
         unit (e.g., weaning from ventilatory support);

    c)   to reduce BZD-induced sedation or to counteract paradoxical
         anxiety reactions to BZD in anaesthesia;

    d)   to antagonise BZD-induced sedation after short diagnostic
         procedures where a long-acting BZD has been used.

         Flumazenil may be justified in the following situations in cases
    of BZD poisoning:

    a)   to facilitate gastric lavage and avoid intubation in comatose

    b)   to treat complications in severe cases of mixed poisoning where
         BZD is thought to be one of the major toxic agents;

    c)   to avoid the need for mechanical ventilation in cases where there
         is respiratory depression.

         The routine use of flumazenil for the treatment of BZD overdosage
    is not recommended.

    3.13.2  Dosage and route

         The intravenous route of administration is recommended when
    flumazenil is given as a BZD antagonist.  Doses need to be adjusted
    according to individual clinical response and the following are

    a)   In anaesthetics and in intensive care (adults), doses of 0.2-0.5
         mg should be used to reduce sedation and doses of 0.5-1 mg to
         abolish BZD effects.

    b)   In cases of BZD overdosage (adults), single doses of 0.3-1 mg can
         be given and repeated as necessary.  The absence of clinical
         response to 2 mg flumazenil within 5-10 min indicates that BZD
         poisoning is not the major cause of coma and other complications. 
         It is also possible to administer a continuous infusion (0.3 to
         1 mg/h) of flumazenil (diluted in 0.9% sodium chloride solution
         or 5% glucose solution) following an initial response to

    c)   In children, it is suggested that intravenous doses of 0.1 mg
         should be given once per minute until the child is awake.  It may
         be necessary to give a subsequent continuous intravenous infusion
         at a rate of 0.1 to 0.2 mg/h.

    3.13.3  Precautions/contraindications  Pharmaceutical precautions

         Solutions of flumazenil should be stored at +4 C.  No other drug
    should be injected or infused with the flumazenil, which should be
    made up in 0.9% sodium chloride or 5% glucose (dextrose) solution.  Other precautions

         Treatment with flumazenil requires continuous intensive
    observation. After the administration of a single dose of flumazenil,
    the patient must be observed for at least 2 h to be certain that
    BZD-induced complications will not recur.  The termination of
    continuous infusion requires intensive care monitoring.

         Note that in cases of mixed drug overdosage, the patient may
    remain unresponsive where other drugs are contributing to the coma.

         BZD antagonism by flumazenil may in certain circumstances be

    a)   An acute withdrawal syndrome can occur in patients showing BZD
         habituation following therapy or abuse.

    b)   Convulsions can occur in cases of mixed drug overdosage where BZD
         has been taken with a drug liable to cause convulsions (such as
         a tricyclic antidepressive agent).

    c)   Convulsions can be induced in patients treated with BZD for
         seizure disorders.

         The three above-mentioned situations may be considered relative
    contraindications to its use; flumazenil should only be used when it
    is strongly indicated.  In these situations, it should be given more
    slowly than usual (e.g., 0.3 mg intravenously over 3 min, a 3-min
    pause, then a further 0.3 mg at the same rate, and so on).