IPCS INCHEM Home




    IPCS/CEC EVALUATION OF ANTIDOTES SERIES



    VOLUME 2


    ANTIDOTES FOR POISONING BY CYANIDE







    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

    T.J. MEREDITH
    Department of Health, London, United Kingdom

    D. JACOBSEN
    Ulleval University Hospital, Oslo, Norway

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

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

    Guest Editor

    A.N.P. van HEIJST
    Formerly of the Dutch National Poison Control Centre,
    Utrecht, The Netherlands

    EUR 14280 EN

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

    CAMBRIDGE UNIVERSITY PRESS


    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 14280 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 45458 1 hardback


    CONTENTS

         PREFACE
         ABBREVIATIONS
    1. OVERVIEW
         1.1. Historical review
         1.2. Potential sources of cyanide
               1.2.1. Industrial sources
               1.2.2. Non-industrial sources
               1.2.3. Natural sources
               1.2.4. Iatrogenic sources
         1.3. Toxicity of cyanide in man
               1.3.1. Acute poisoning
               1.3.2. Chronic poisoning
         1.4. Mechanism of toxicity
         1.5. Clinical features
         1.6. Laboratory findings
               1.6.1. Lactic acidosis
               1.6.2. Hyperglycaemia
               1.6.3. Cyanide concentration in blood and plasma
         1.7. Biological detoxification of cyanide
               1.7.1. Thiocyanate toxicity
         1.8. Protective measures for occupational exposure
         1.9. Treatment
               1.9.1. Supportive treatment
               1.9.2. Antidotal treatment
                       1.9.2.1    Oxygen
                       1.9.2.2    Sodium thiosulfate
                       1.9.2.3    Amyl nitrite
                       1.9.2.4    Sodium nitrite
                       1.9.2.5    4-Dimethylaminophenol
                       1.9.2.6    Hydroxocobalamin
                       1.9.2.7    Dicobalt edetate
                       1.9.2.8    Antidotes to methaemoglobin-forming
                                  agents
         1.10. Summary of treatment recommendations
               1.10.1. First aid and treatment measures at the site of
                       the incident
               1.10.2. Hospital treatment
                       1.10.2.1   Severe poisoning
                       1.10.2.2   Moderately severe poisoning
                       1.10.2.3   Mild Poisoning
         1.11. Summary of analytical aspects
         1.12. Proposed areas for research
         1.13. New developments in cyanide antidotes
               1.13.1. Nonspecific agents
               1.13.2. Sodium pyruvate
               1.13.3. Ifenprodil
               1.13.4. Rhodanese
               1.13.5. Alpha-ketoglutaric acid

               1.13.6. Stroma-free methaemoglobin solution
         1.14. References

    2. OXYGEN
         2.1. Introduction
         2.2. Name and chemical formula of antidote
         2.3. Physico-chemical properties of molecular oxygen
         2.4. Synthesis
         2.5. Analytical methods
               2.5.1. Quality control procedures
                       2.5.1.1    Tests
                       2.5.1.2    Assay for oxygen
               2.5.2. Methods for identification
               2.5.3. Methods for analysis of the antidote in
                       biological samples
                       2.5.3.1    In the gas phase
                       2.5.3.2    In solution
               2.5.4. The saturation of haemoglobin by oxygen
         2.6. Storage conditions
         2.7. General properties
         2.8. Animal studies
               2.8.1. Pharmacokinetics
               2.8.2. Pharmacodynamics
               2.8.3. Toxicology
                       2.8.3.1    Mechanism of injury
         2.9. Volunteer studies of pulmonary oxygen toxicity
         2.10. Clinical studies of oxygen toxicity
               2.10.1. Eyes
               2.10.2. Central nervous system
         2.11. Clinical studies - case reports
               2.11.1. Patients treated alone with supportive therapy
                       and who survived
               2.11.2. Hyperbaric oxygen therapy in cyanide poisoning
               2.11.3. Cyanide poisoning due to smoke inhalation
         2.12. Summary of evaluation
         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. SODIUM THIOSULFATE
         3.1. Introduction
               3.1.1. Indications
               3.1.2. Rationale for the choice of the antidote
               3.1.3. Risk groups
         3.2. Name and chemical formula of antidote
         3.3. Physico-chemical properties

               3.3.1. Melting point, boiling point
               3.3.2. Solubility in vehicle for administration
               3.3.3. Optical properties
               3.3.4. Acidity
               3.3.5. pKa
               3.3.6. Stability
               3.3.7. Refractive index, specific gravity
               3.3.8. Loss of weight on drying
               3.3.9. Excipients
               3.3.10. Incompatibility
               3.3.11. Other information
         3.4. Synthesis
         3.5. Analytical methods
               3.5.1. Quality control procedures for sodium thiosulfate
               3.5.2. Methods for identifying sodium thiosulfate
               3.5.3. Assay
               3.5.4. Methods for analysis of sodium thiosulfate in
                       biological samples
         3.6. Shelf-life
         3.7. General properties
               3.7.1. Mechanism of antidotal activity
               3.7.2. Other biochemical/pharmacological profiles
         3.8. Animal studies
               3.8.1. Pharmacokinetics
               3.8.2. Pharmacodynamics
               3.8.3. Toxicology
         3.9. Volunteer studies
         3.10. Clinical studies
         3.11. Clinical studies - case reports
         3.12. Summary of evaluation
               3.12.1. Indications
               3.12.2. Route of administration
               3.12.3. Dose
               3.12.4. Other consequential or supportive therapy
         3.13. Model information sheet
               3.13.1. Uses
               3.13.2. Dosage and route of administration
               3.13.3. Precautions and contraindications
               3.13.4. Adverse effects
               3.13.5. Use in pregnancy/lactation
               3.13.6. Storage
         3.14. References

    4. HYDROXOCOBALAMIN
         4.1. Introduction
         4.2. Name and chemical formula of antidote
         4.3. Physico-chemical properties
               4.3.1. Characteristics
               4.3.2. Melting-point
               4.3.3. Solubility in vehicles for administration
               4.3.4. Optical properties

               4.3.5. Acidity
               4.3.6. Stability in light
               4.3.7. Thermal stability
               4.3.8. Interference with other compounds
         4.4. Synthesis
         4.5. Analytical methods
               4.5.1. Identification of hydroxocobalamin
                       4.5.1.1    UV spectroscopy
                       4.5.1.2    Colorimetric method
               4.5.2. Quality controls
               4.5.3. Raw materials
               4.5.4. Finished galenic form
               4.5.5. Measurement
                       4.5.5.1    In raw materials and in finished form
                       4.5.5.2    In biological samples
         4.6. Shelf-life
         4.7. General properties
         4.8. Animal studies
               4.8.1. Pharmacokinetics
               4.8.2. Pharmacodynamics in the presence of the toxin
               4.8.3. Toxicology
                       4.8.3.1    Acute toxicity
                       4.8.3.2    Sub-acute and chronic toxicity
         4.9. Volunteer studies
         4.10. Clinical studies
         4.11. Clinical studies - case reports
         4.12. Summary of evaluation
               4.12.1. Indications
               4.12.2. Advised route and dosage
               4.12.3. Practical advice
               4.12.4. Side effects
         4.13. Model information sheet
               4.13.1. Uses
               4.13.2. Dosage and route
               4.13.3. Precautions/contraindications
               4.13.4. Adverse effects
               4.13.5. Use in pregnancy and lactation
               4.13.6. Storage
         4.14. References

    5. DICOBALT EDETATE
         5.1. Introduction
         5.2. Name and chemical formula
         5.3. Physico-chemical properties
         5.4. Synthesis
               5.4.1. Source of materials
                       5.4.1.1    Cobalt carbonate
                       5.4.1.2    Ethylenediaminetetraacetic acid
                       5.4.1.3    Glucose

         5.5. Analytical methods
               5.5.1. Free cobalt
               5.5.2. Dicobalt edetate
               5.5.3. Analysis in biological fluids
         5.6. Stability and shelf-life
         5.7. General properties
         5.8. Animal studies
               5.8.1. Pharmacokinetics
               5.8.2. Pharmacodynamics
                       5.8.2.1    Efficacy in animals
                       5.8.2.2    Comparison of dicobalt edetate with
                                  other compounds
                       5.8.2.3    Interactions with other drugs
               5.8.3. Toxicology
                       5.8.3.1     In vitro studies
                       5.8.3.2    Acute toxicity studies
                       5.8.3.3    Repeated dose toxicity
                       5.8.3.4    Circulatory effects in dogs
                       5.8.3.5    Other toxicity studies
         5.9. Volunteer studies
         5.10. Clinical trials
         5.11. Clinical studies - case reports
               5.11.1. Successful use
               5.11.2. Use in pregnant women and children
               5.11.3. Adverse effects
               5.11.4. Use in combination with other antidotes
         5.12. Summary of evaluation
               5.12.1. Indications
               5.12.2. Administration
               5.12.3. Other consequential or supportive therapy
               5.12.4. Contraindications
               5.12.5. Comparison with other antidotes
         5.13. Model information sheet
               5.13.1. Uses
               5.13.2. Dosage and route
               5.13.3. Precautions/contraindications
               5.13.4. Adverse effects
               5.13.5. Use in pregnancy and lactation
               5.13.6. Storage
         5.14. References

    6. AMYL NITRITE
         6.1. Introduction
         6.2. Name and chemical formula
         6.3. Physico-chemical properties
         6.4. Synthesis
         6.5. Analytical methods
               6.5.1. Identification
               6.5.2. Purity
                       6.5.2.1    Acidity
                       6.5.2.2    Non-volatile residue

                       6.5.2.3    Assay for total nitrites
         6.6. Shelf-life
         6.7. General properties
         6.8. Animal studies
               6.8.1. Pharmacokinetics
               6.8.2. Pharmacodynamics
               6.8.3. Toxicology
         6.9. Volunteer studies
         6.10. Clinical studies
         6.11. Clinical studies - case reports
         6.12. Summary of evaluation
               6.12.1. Indications
               6.12.2. Advised routes and dose
               6.12.3. Other consequential or supportive therapy
         6.13. Model information sheet
               6.13.1. Uses
               6.13.2. Dosage and route
               6.13.3. Precautions/contraindications
               6.13.4. Storage
         6.14. References

    7. SODIUM NITRITE
         7.1. Introduction
         7.2. Name and chemical formula
         7.3. Physico-chemical properties
         7.4. Synthesis
         7.5. Analytical methods
               7.5.1. Quality control
                       7.5.1.1    Solid sodium nitrite
                       7.5.1.2    Sodium nitrite injection
                       7.5.1.3    Preparation of volumetric solutions
               7.5.2. Identification
                       7.5.2.1    Sodium
                       7.5.2.2    Nitrite
               7.5.3. Impurities
                       7.5.3.1    Preparation of sodium nitrite to test
                       7.5.3.2    Preparation of special reagents
                       7.5.3.3    Preparation of standard
                       7.5.3.4    Preparation of test
                       7.5.3.5    Preparation of monitor
                       7.5.3.6    Preparation of hydrogen sulfide test
                                  solution
                       7.5.3.7    Test procedure
         7.6. Shelf-life
         7.7. General properties
               7.7.1. Mode of action
               7.7.2. Other relevant properties
         7.8. Animal studies
               7.8.1. Pharmacokinetics
               7.8.2. Pharmacodynamics
               7.8.3. Toxicology

         7.9. Volunteer studies
               7.9.1. Pharmacokinetics
               7.9.2. Sodium nitrite poisoning
         7.10. Clinical studies
         7.11. Clinical studies - case reports
         7.12. Summary of evaluation
               7.12.1. Indications
               7.12.2. Contraindications
               7.12.3. Advised route and dosage
               7.12.4. Other consequential or supportive therapy
         7.13. Model information sheet
               7.13.1. Uses
               7.13.2. Dosage and route
               7.13.3. Precautions/contraindications
               7.13.4. Adverse effects
               7.13.5. Use in pregnancy and lactation
               7.13.6. Storage
         7.14. References

    8. 4-DIMETHYLAMINOPHENOL
         8.1. Introduction
         8.2. Name and chemical formula
         8.3. Physico-chemical properties
         8.4. Synthesis
         8.5. Analytical methods
               8.5.1. Identity
               8.5.2. Quantification
               8.5.3. Purity
               8.5.4. Methods for analysis of 4-DMAP in biological
                       samples
         8.6. Shelf-life
         8.7. General properties
         8.8. Animal studies
               8.8.1.  In vitro studies
                       8.8.1.1    Metabolism of 4-DMAP in the liver
                       8.8.1.2    Red cell metabolism of 4-DMAP
                       8.8.1.3    Toxic effects of 4-DMAP on
                                  erythrocytes
                       8.8.1.4    Toxic effects of 4-DMAP on isolated
                                  rat kidney tubules
                       8.8.1.5    Oxygen saturation and methaemoglobin
                                  formation
               8.8.2. Pharmacokinetics
               8.8.3. Pharmacodynamics
               8.8.4. Toxicology
                       8.8.4.1    Nephrotoxicity
                       8.8.4.2    Mutagenicity
         8.9. Volunteer studies
               8.9.1. Metabolism of 4-DMAP in the liver
               8.9.2. Metabolism of 4-DMAP in erythrocytes

               8.9.3. Adverse effects
         8.10. Clinical studies
         8.11. Clinical studies - case reports
         8.12. Summary of evaluation
               8.12.1. Indications
               8.12.2. Recommended routes and dosage
               8.12.3. Other consequential or supportive therapy
               8.12.4. Areas of use where there is insufficient
                       information to make recommendations
         8.13. Model information sheet
               8.13.1. Uses
               8.13.2. Dosage and route
               8.13.3. Precautions/contraindications
               8.13.4. Adverse effects
               8.13.5. Use in pregnancy and lactation
               8.13.6. Storage
         8.14. References

    9. METHYLENE BLUE AND TOLUIDINE BLUE
         9.1. Methylene blue
               9.1.1. Introduction
               9.1.2. Name and chemical formula of antidote
               9.1.3. Physico-chemical properties
               9.1.4. Synthesis
               9.1.5. Analytical methods
               9.1.6. Shelf-life
               9.1.7. General properties
               9.1.8. Animal studies
               9.1.9. Volunteer studies
               9.1.10. Clinical studies
               9.1.11. Clinical studies - case reports
               9.1.12. Summary of evaluation
                       9.1.12.1   Indications
                       9.1.12.2   Advised route and dosage
                       9.1.12.3   Precautions and contraindications
                       9.1.12.4   Adverse effects
                       9.1.12.5   Other consequential or supportive
                                  theory
               9.1.13. Model information sheet
                       9.1.13.1   Uses
                       9.1.13.2   Dosage and route of administration
                       9.1.13.3   Precautions and contraindications
                       9.1.13.4   Adverse effects
                       9.1.13.5   Use in pregnancy/lactation
                       9.1.13.6   Storage
               9.1.14. References
         9.2. Toluidine blue
               9.2.1. Introduction
               9.2.2. Name and chemical formula of antidote
               9.2.3. Physico-chemical properties
               9.2.4. Synthesis

               9.2.5. Analysis
                       9.2.5.1    Analysis of methaemoglobin
               9.2.6. Stability
               9.2.7. General properties
               9.2.8. Animal studies
                       9.2.8.1    Pharmacokinetics
                       9.2.8.2    Pharmacodynamics
                       9.2.8.3    Toxicology
               9.2.9. Volunteer studies
               9.2.10. Clinical studies
               9.2.11. Clinical studies - case reports
               9.2.12. Summary of evaluations
               9.2.13. Model information sheet
                       9.2.13.1   Indications
                       9.2.13.2   Side effects
                       9.2.13.3   Advised route and dose
                       9.2.13.4   Use in pregnancy and children
                       9.2.13.5   Storage
               9.2.14. References

    10. ANALYTICAL METHODS FOR CYANIDE ALONE AND IN
         COMBINATION WITH CYANIDE ANTIDOTES IN BLOOD

         10.1. Qualitative methods
               10.1.1. Detection in blood with a detector tube
                       10.1.1.1   Principle
                       10.1.1.2   Materials
                       10.1.1.3   Procedure
                       10.1.1.4   Specificity
               10.1.2. Spot test
                       10.1.2.1   Principle
                       10.1.2.2   Equipment
                       10.1.2.3   Chemicals
                       10.1.2.4   Reagents
                       10.1.2.5   Specimen collection
                       10.1.2.6   Procedure
                       10.1.2.7   Specificity
         10.2. Quantitative methods
               10.2.1. Gas chromatographic head space technique
                       10.2.1.1   Principle
                       10.2.1.2   Equipment
                       10.2.1.3   Chemicals
                       10.2.1.4   Solutions
                       10.2.1.5   Calibration standards
                       10.2.1.6   Specimen collection and sample
                                  preparation
                       10.2.1.7   Operational parameters for gas
                                  chromatography
                       10.2.1.8   Analytical determination
                       10.2.1.9   Calibration
                       10.2.1.10  Calculation of the analytical result

                       10.2.1.11  Reliability of the method
                       10.2.1.12  Detection limit
                       10.2.1.13  Specificity
               10.2.2. Microdiffusion technique
                       10.2.2.1   Principle
                       10.2.2.2   Equipment
                       10.2.2.3   Chemicals
                       10.2.2.4   Solvents and reagents
                       10.2.2.5   Calibration standards
                       10.2.2.6   Specimen
                       10.2.2.7   Procedure
                       10.2.2.8   Reliability of the method
                       10.2.2.9   Detection limit
                       10.2.2.10  Specificity
         10.3. References
    

    WORKING GROUP ON ANTIDOTES TO POISONING BY CYANIDE

     Members

    Professor C. Bismuth, Hôpital Fernand Widal, Clinique Toxicologique,
    Paris, France

    Professor M. von Clarmann, Poisons Centre, Toxicology Department, 11
    Med. Klinikrechts der Isar der Tecknischer Universität, Munich,
    Germany

    Dr A. van Dijk, Apotheek, Academisch Ziekenhuis, Utrecht, The
    Netherlands

    Professor M. Geldmacher von Mallinckrodt, Institut für
    Rechtsmedizia, Erlangen, Germany

    Dr A. Hall, Rocky Mountain Poison and Drug Center, Denver, Colorado,
    USA

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

    Dr T.C. Marrs, Department of Health, London, United Kingdom

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

    Dr A.C.G.M. Parren, Te Heerlen, The Netherlands

    Dr H. Persson, Poison Information Centre, Karolinska Sjukhuset,
    Stockholm, Sweden

    Dr U. Taitelman, Rambam Medical Center, Haifa, Israel

     Observers

    Dr J. Aubrun, Rhone-Poulenc, Courbevoie, France

    Dr A. Heath, Poisons Therapy Group, Department of Anaesthesia and
    Intensive Care, Sahlgren's Hospital, Gothenburg, Sweden

    Dr J. Henry, Poisons Unit, New Cross Hospital, London, United
    Kingdom

    Dr J.A. Vale, West Midlands Poisons Unit, Dudley Road Hospital,
    Birmingham, United Kingdom

     Secretariat

    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  (Chairman)

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

    Dr M.-Th. van der Venne, Health and Safety Directorate, Commission
    of the European Communities, Luxembourg

    PREFACE

         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. 
    These were included in volume 1 of this series.

         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 antidotes used in the treatment of
    poisoning by cyanide.

         The review and evaluation of antidotes used in the treatment of
    poisoning by cyanide was initiated at a joint meeting of the
    European Association of Poison Control Centres and Clinical
    Toxicologists (EAPCCT; formerly known as the European Association of
    Poison Control Centres), the IPCS, and the CEC, organized by the
    National Poison Information Centre of the Netherlands National
    Institute of Public Health and Environmental Hygiene and held at the
    University Hospital AZU, Utrecht, The Netherlands, 13-15 May 1987. 
    In preparation for this meeting, documents were drafted, using the
    proforma, on oxygen by Dr U. Taitelman, sodium thiosulfate by Dr H.
    Persson, hydroxocobalamin by Professor C. Bismuth, dicobalt edetate

    by Dr T.C. Marrs, sodium nitrite by Dr A. Hall, and
    4-dimethylaminophenol by Professor M. von Clarmann.  Also in
    preparation for the meeting, documents were drafted by Professor M.
    Geldmacher von Mallinckrodt on the analytical assessment of cyanide
    poisoning, by Dr A. van Dijk on the pharmaceutical aspects of
    cyanide antidotes, and by Professor A.N.P. van Heijst (formerly
    Director, Dutch National Poison Control Centre, Utrecht, the
    Netherlands) on the clinical aspects of cyanide antidotes.  The
    documents presented by each author were discussed at the meeting and
    participants gave their own experience and views.  Experience of
    industrial aspects of cyanide poisoning was presented by Dr A.C.G.M.
    Parren.

         The main meeting was followed by that of an IPCS/CEC working
    group, consisting of the authors of documents, the meeting
    rapporteur and a number of observers, at which a review was made of
    the comments on the documents and of the additional material
    presented at the main meeting.  Based on the available material, an
    evaluation was made of the different approaches to treatment of
    cyanide poisoning depending on the type of cyanide exposure
    (hydrogen cyanide, either alone or with carbon monoxide, cyanide
    salts or cyanogenic glycosides), the state of intoxication and
    number of patients, the location of the patient with respect to
    treatment facilities, and special situations (e.g., inherited
    metabolic and haemoglobin abnormalities).  The group concentrated on
    acute poisoning by cyanide, considering that there were insufficient
    data for evaluating approaches to treatment of chronic cyanide
    toxicity.  Nevertheless, it was considered that a review of chronic
    poisoning by cyanide, particularly in relation to cyanide ingestion
    from food, was needed.  It was agreed that traditional means of
    treatment of cyanide poisoning would have to be revised, and that
    any evaluation of approaches to treatment must also include
    antidotes for methaemoglobin-forming agents.  Concerning the
    analytical aspects, it was noted that there was particular
    difficulty in measuring the concentration of cyanide in blood if an
    antidote had already been administered, a problem that is being
    studied by a group of experts established under the auspices of the
    German Research Association Commission on Clinical Analytical
    Toxicology.  A number of new cyanide antidotes in various stages of
    research and development were discussed.  An editorial group
    consisting of Professor A.N.P. van Heijst (chairman of the meeting),
    Dr T.J. Meredith (rapporteur), Dr J.A. Haines (IPCS, chairman of the
    working group) and Dr J.-C. Berger (CEC) was established in order to
    prepare a consolidated monograph on cyanide antidotes.

         Draft documents were revised by their authors.  Those on
    methylene blue and toluidine blue were prepared by Dr Christina
    Alonzo (CIAT, Montevideo, Uruguay) and Dr T.C. Marrs, respectively. 
    Subsequently Dr J.A. Vick (Food and Drug Administration, USA), who
    was invited to the meeting but was unable to attend, prepared a
    draft document on experience with the use of amyl nitrite in

    treating cyanide poisoning in animals.  Professor C. Bismuth and
    Dr A. Hall drafted material on new antidotes under development for
    clinical trials, and Dr A.C.G.M. Parren drafted material on
    protective measures.

         The editorial group met twice in Utrecht on 22-23 October 1987
    and 20-22 July 1988.  Material was checked and rearranged,
    additional material was prepared for a number of the chapters and
    the overview chapter was drafted.  The efforts of all who helped in
    the preparation and finalization of this monograph are gratefully
    acknowledged.

    ABBREVIATIONS

    ATA            atmosphere absolute
    BE             base excess
    CNS            central nervous system
    CT             computer tomography
    4-DMAP         4-dimethylaminophenol
    EDTA           ethylenediaminetetraacetic acid
    G6PD           glucose-6-phosphate dehydrogenase
    Hb             haemoglobin
    HMPS           hexose monophosphate shunt
    INN            international non-proprietary name
    LDLo           lowest published lethal dose
    MLD            minimal lethal dose
    NADH           reduced nicotinamide adenine dinucleotide
    NADPH          reduced nicotinamide adenine dinucleotide phosphate
    OHB12        hydroxocobalamin
    LD50 	   Lethal Dose 50
    USP            United States Pharmacopoeia
    B12          Vitamin B12
    HbO2         Oxyhaemoglobin
    AV             atrioventricular    
    SNP            sodium nitroprusside
    VS             volumetric solution

    1.  Overview

    1.1  Historical Review

         The recognition of cyanide as a poison in bitter almonds,
    cherry laurel leaves, and cassava goes back to antiquity.  An
    inscription on an Egyptian papyrus in the Louvre Museum, Paris,
    refers to the "penalty of the peach," and Dioscorides in the first
    century A.D. was aware of the poisonous properties of bitter almonds
    (Sykes, 1981).

         The first description of cyanide poisoning was by Wepfer in
    1679 and dealt with the effects of the administration of extract of
    bitter almonds (Sykes, 1981).  Two fatal cases of poisoning in
    Ireland caused by drinking cherry laurel water, used as a flavouring
    agent in cooking and to dilute brandy, led to the experiments of
    Madden (1731).  He showed that cherry laurel water contains a
    poison; given orally, into the rectum, or by injection, it rapidly
    killed dogs.  It was not until 1786 that isolation of pure hydrogen
    cyanide (HCN) from the dye Prussian blue was achieved by Scheele
    (1786).  The mechanism of toxicity of cyanide was explored by
    Fontana (1795).  Cyanide was obtained from bitter almonds by
    Schrader (1802).  The introduction of cyanide as a medicament to
    treat coughs and lung diseases was suggested by Magendie (1817). 
    Indeed, it was not until 1948 that cherry laurel water was removed
    from the British Pharmacopoeia!  Attempts to antagonize the toxic
    effects of cyanide were reported by Blake (1839 and 1840). 
    Hoppe-Seyler (1876) reported that cyanide inhibits tissue oxidation
    reactions.

         Antagonism between amyl nitrite and prussic acid was mentioned
    by Pedigo (1888), and, as early as 1894, cobalt compounds were
    advocated by Antal (1894) as cyanide antagonists.  Sodium nitrite
    was used as an antidote in experimental cyanide poisoning by
    Mladoveanu & Gheorghiu (1929).

         A biochemical mechanism for cyanide antagonism was described by
    Chen et al. (1933, 1934).  They suggested using a combination of
    amyl nitrite, sodium nitrite and sodium thiosulfate, the latter
    compound serving as a sulfur donor for rhodanese (thiosulfate sulfur
    transferase).  Rhodanese accelerates cyanide detoxification by
    forming the metabolite thiocyanate.  This represented the
    development of one of the first antidotes based on scientific
    toxicological reasoning.  This combination of antidotes has stood
    the test of time, and still represents one of the most efficacious
    antidotal combinations for the treatment of cyanide intoxication.

         Interest in cobalt compounds was renewed by Mushett et al.
    (1952), who demonstrated in 1952 that hydroxocobalamin (vitamin
    B12a) combined with cyanide to form cyanocobalamin (vitamin
    B12).

         Paulet (1960) subsequently reported that cobalt EDTA was more
    effective as a cyanide antidote than the classic nitrite-thiosulfate
    combination.

    1.2  Potential Sources of Cyanide

    1.2.1  Industrial sources

         Hydrogen cyanide is used in the fumigation of ships, large
    buildings, flour mills, private dwellings, freight cars, and
    aeroplanes that have been infested by rodents or insects.  It is
    bound to a carrier, commonly diatomaceous earth, and blended with an
    odorous or irritating product as a warning marker.

         Cyanide salts are utilized in metal cleaning, hardening,
    ore-extracting processes, and electroplating.

         Halogenated cyanides (chloro-, bromo- and iodocyanide) in
    contact with water produce the non-toxic cyanic acid.  As a result
    of contact with strong acids, hydrogen cyanide is liberated.

         Nitriles are cyano-derivatives of organic compounds.  Acetonitrile
    is used as a solvent and is less toxic (LD50 = 120 mg/kg) than
    hydrogen cyanide (LD50= 0.5 mg/kg), but often contains toxic
    admixtures due to metabolism to inorganic cyanide.  While aliphatic nitriles
    metabolise to inorganic cyanide, the aromatic nitrile bond is stable
     in vivo. Acrylonitrile is the raw material used for the
    manufacture of plastics and synthetic fibres.  Contact with skin
    causes bullae formation.  Pyrolysis generates hydrogen cyanide. 
    Acrylonitrile and propionitrile are less toxic (LD50 = 35 mg/kg)
    than butyronitrile (LD50 = 10 mg/kg).  Trichloroacetonitrile
    (LD50  = 200 mg/kg) is used as an insecticide.  The aromatic
    nitriles, bromoxynil (LD50= 190 mg/kg) and ioxynil (LD50= 110
    mg/kg), are used as herbicides.

         Cyanamide, cyanoacetic acid, ferricyanide and ferrocyanide do
    not release cyanide.  They are therefore less toxic (LD50=
    1000-2000 mg/kg) than the cyanogenic compounds above, though they
    may cause toxicity by other means, e.g. cyanide in combination with
    alcohol.

    1.2.2  Non-industrial sources

         Fires and automobile pollution-control devices with
    malfunctioning catalytic converters (Voorhoeve et al., 1975)
    generate cyanide.  Natural substances, such as wool, silk, horse
    hair, and tobacco, as well as modern synthetic materials, such as
    polyurethane and polyacrylonitriles, release cyanide during
    combustion (Levine et al., 1978; Birky et al., 1979; Anderson &
    Harland, 1982; Clark et al., 1983; Alarie, 1985; Lowry et al., 1985)
    (Table 1).

    Table 1.  Hydrogen cyanide generated by pyrolysis

                                                        
                                       µg HCN per
              Material                 g material
                                                        
              paper                       1100
              cotton                       130
              wool                        6300
              nylon                        780
              polyurethane foam           1200
                                                        
              From:  Montgomery et al. (1975)

    1.2.3  Natural sources

         Cyanide is found in foodstuffs such as cabbage, spinach, and
    almonds, and as amygdalin in apple pips, peach, plum, cherry, and
    almond kernels.  In the kernels themselves, amygdalin seems to be
    completely harmless as long as it is relatively dry.  However, the
    seeds contain an enzyme that is capable of catalysing the following
    hydrolytic reaction when the seeds are crushed and moistened:

         C20H27NO11  +   2H2O   -->   2C6H12O6  +  C6H5CHO +  HCN

         amygdalin       glucose      benzaldehyde            hydrogen
                                                              cyanide

         The reaction is slow in acid but rapid in alkaline solution.

         Natural oil of bitter almonds contains 4% HCN.  American white
    lima beans contain 10 mg cyanide/100 g bean.  The dried root of
    cassava (tapioca) may contain 245 mg cyanide/100 g root.  The
    cyanide content in 100 g of cultivated apricot seeds has been found
    to be about 9 mg and that in wild apricot seeds more than 200 mg.

    1.2.4  Iatrogenic sources

         Cyanide is also formed during nitroprusside therapy, especially
    when it is prolonged, because tachyphylaxis sometimes requires the
    use of higher doses than the recommended maximum of 10 µg/kg per min
    (Smith & Kruszyna, 1974; MacRae & Owen, 1974; Piper, 1975; Atkins,
    1977; Anon, 1978). Cyanide metabolises to thiocyanate. Thiocyanates
    were used some years ago as
    antihypertensive agents and they saw wide use because they were very
    effective.  However, a variety of subacute toxic effects, including
    anorexia, fatigue, and gastrointestinal tract and CNS disturbances,
    led to their disfavour.

         Laetrile, amygdalin derived from apricot kernels, has been used
    as an anticancer agent, but it is now obsolete because a therapeutic
    effect could not be demonstrated in either retrospective or
    prospective studies.  Laetrile has caused fatal cyanide poisoning
    (Sadoff et al., 1978).

    1.3  Toxicity of Cyanide in Man

    1.3.1  Acute poisoning

         It is generally accepted that inhalation of approximately 50 ml
    (at 1.85 mmol/l) of hydrogen cyanide gas is fatal within minutes. 
    Poisoning from hydrogen cyanide is more frequently
    accidental than suicidal.  Thus accidental cyanide poisoning may
    occur in fumigators and chemists who use hydrogen cyanide during the
    course of their work (Chen et al., 1944).  In fires, a combination
    of HCN and carbon monoxide (CO) toxicity, as a result of inhalation of
    combustion products, may cause fatalities.

         Suicidal ingestion of cyanide salts most commonly occurs in
    personnel with occupational access to cyanide.  The ingestion of as
    little as 250 mg of an inorganic cyanide salt may be fatal (Peters
    et al., 1982).  However, death may be delayed for several hours
    following the ingestion of cyanide on a full stomach; a first-pass
    effect in the liver may also delay the onset of toxicity
    (Naughton, 1974).

    1.3.2  Chronic poisoning

         Chronic low-dose neurotoxicity have been suggested by
    epidemiological studies of populations ingesting naturally occurring
    plant glycosides (Blanc et al, 1985).  These glycosides are present
    in a wide variety of plant species, most notably the cassava plant,
    a major tropical foodstuff (Conn, 1973; Cook & Coursey, 1981;
    Ministry of Health, Mozambique, 1984).  Cassava has been associated
    with tropical ataxic neuropathy (Cook & Coursey, 1981).  Epidemic
    spastic paraparesis has been associated with a combination of a high
    cyanide and a low sulfur intake from diets dominated by
    insufficiently processed cassava and lacking protein supplementary
    food (Rosling, 1989).  A neurotoxicological role for cyanide has
    also been suggested in tobacco-associated amblyopia (Grant, 1980)
    and in amygdalin-associated peripheral neuropathy (Kalyanaraman et
    al., 1983).  Long-term cyanide intoxication has been shown to be
    associated both with thyroid gland enlargement and dysfunction in
    case reports and in cohort studies of individuals exposed
    occupationally (Blanc et al., 1985), through dietary intake (Cook &
    Coursey, 1981), and experimentally (El Ghawabi et al.,  1975).

    1.4  Mechanism of Toxicity

         Cyanide has a special affinity for the ferric ions that occur
    in cytochrome oxidase, the terminal oxidative respiratory enzyme in
    mitochondria.  This enzyme is an essential catalyst for tissue
    utilization of oxygen.  When cytochrome oxidase is inhibited by
    cyanide, histotoxic anoxia occurs as aerobic metabolism becomes
    inhibited.  In massive cyanide poisoning, the mechanism of toxicity
    is more complex.  It is possible that autonomic shock from the
    release of biogenic amines may play a role by causing cardiac
    failure (Burrows & Way, 1976).  Cyanide could cause both pulmonary
    arteriolar and/or coronary arterial vasoconstriction, which would
    result, either directly or indirectly, in pump failure and a
    decrease in cardiac output.  This theory is supported by the sharp
    increase in central venous pressure that was observed by Vick &
    Froelich (1985) at a time when the arterial blood pressure fell
    after the intravenous administration of sodium cyanide to dogs.  The
    observation that phenoxybenzamine, an alpha-adrenergic blocking
    drug, partially prevented these early changes (Vick & Froelich,
    1985) supports the concept of an early shock-like state not related
    to inhibition of the cytochrome oxidase system.  Inhalation of amyl
    nitrite, a potent arteriolar vasodilating agent, resulted in the
    survival of dogs in these experimental circumstances.  This could
    have been due to reversal of early cyanide-induced vasoconstriction
    with restoration of normal cardiac function (Vick & Froelich, 1985).

    1.5  Clinical Features

         The smell of bitter almonds in expired air is an important sign
    in cyanide poisoning.  However, many people are unable to perceive
    the odour of hydrocyanic acid (Kalmus & Hubbard, 1960).  The
    incidence of "non-smellers" is reported to be 18% among males and 5%
    among females (Kirk & Stenhouse, 1953; Fukumoto et al., 1957).

         Immediately after swallowing cyanide, very early symptoms, such
    as irritation of the tongue and mucous membranes, may be
    experienced.  A blood-stained aspirate may be observed if gastric
    lavage is performed.  Early symptoms and signs that occur after
    inhalation of HCN or the ingestion of cyanide salts include anxiety,
    headache, vertigo, confusion,  and hyperpnoea, followed by dyspnoea,
    cyanosis, hypotension, bradycardia, and sinus or AV nodal
    arrythmias.

         In the secondary stage of poisoning, impaired consciousness,
    coma and convulsions occur and the skin becomes cold, clammy, and
    moist.  The pulse becomes weaker and more rapid.  Opisthotonos and
    trismus may be observed.  Late signs of cyanide toxicity include
    hypotension, complex arrythmias, cardiovascular collapse, pulmonary
    oedema, and death.

         It should be emphasized that the bright-red coloration of the
    skin or absence of cyanosis mentioned in textbooks (Gosselin et al.,
    1984; Goldfrank et al., 1984) is seldom described in case reports of
    cyanide poisonings.  Theoretically this sign could be explained by
    the high concentration of oxyhaemoglobin in the venous return, but,
    especially in massive poisoning, cardiovascular collapse will
    prevent this from occurring.  Sometimes, cyanosis can be observed
    initially, while later the patient may become bright pink (Hilmann
    et al., 1974).

         The pathogenesis of pulmonary oedema could be due to several
    different mechanisms: (1) an intracellular metabolic process that
    could injure the alveolar and capillary epithelium directly,
    producing a capillary leak syndrome; (2) neurogenic pulmonary oedema
    or, (3) most likely, a direct effect on the myocardium leading to
    left ventricular failure and increased pulmonary venous pressure.

         The brain is obviously the key organ involved in cyanide
    poisoning and it has been shown that cyanide significantly increases
    brain lactate and decreases brain ATP concentrations (Olsen & Klein,
    1947).

    1.6  Laboratory Findings

    1.6.1  Lactic acidosis

         Since oxidative phosphorylation is blocked, the rate of
    glycolysis is markedly increased, which in turn leads to lactic
    acidosis.  The degree of lactic acidosis can be correlated with the
    severity of cyanide poisoning (Trapp, 1970; Naughton, 1974).

    1.6.2  Hyperglycaemia

         A reversible toxic effect occurs on the pancreatic beta-cells,
    which may occasionally give rise to an erroneous diagnosis of
    hyperglycaemic diabetic coma.

    1.6.3  Cyanide concentration in blood and plasma

         Before intravenous treatment with antidotes is commenced, it is
    necessary to collect a heparinized (not fluoride) blood sample for
    determination of cyanide concentration.  Results from samples
    collected after treatment are totally unreliable.  A quantitative
    test employing a detector tube (see chapter 10) can be used if the
    diagnosis is in doubt.  The blood can also be used for a
    quantitative test (see chapter 10), so that the severity of
    poisoning can be evaluated.  Therapeutic measures after antidotal
    treatment should be based on the clinical condition of the patient
    rather than on blood cyanide concentrations (Berlin, 1971; Vogel et
    al., 1981; Peters et al., 1982).  Since blood concentrations of up
    to 0.005-0.04 mg/l have been recorded in healthy non-smokers, and

    0.01-0.09 mg/l in smokers, only concentrations above these values
    were previously considered to be toxic (Vogel et al., 1981; Peters
    et al., 1982).  Lundquist et al., (1985) reported even lower
    concentration:  non-smokers 3.4 µg/l  (whole blood), 0.5 µg/l
    (plasma), 6.0 µg/l (erythrocytes); smokers 8.6 µg/l (whole blood),
    0.8 µg/l (plasma), 17.7 µg/l (erythrocytes).

         Fatal cyanide poisoning has been reported with whole blood
    concentrations of >3 mg/l and severe poisoning with 2 mg/l (Graham
    et al., 1977).  However, when cyanide enters the bloodstream, up to
    98% quickly enters the red blood cells where it becomes tightly
    bound.  A plasma-to-blood ratio as high as 1:10 has been reported
    and, as a consequence, the whole blood cyanide concentration may not
    accurately reflect tissue concentrations of cyanide.  Plasma levels
    of cyanide may be of greater significance because severe toxicity
    occurs in the presence of only modest concentrations (Vesey et al.,
    1976).  However, a serious drawback to the use of plasma cyanide
    determinations in the assessment of poisoning is the pronounced
    instability of cyanide in plasma (Lundquist et al., 1985).

    1.7  Biological Detoxification of Cyanide

         The major pathway of endogenous detoxification is conversion,
    by means of thiosulfate, to thiocyanate. Minor routes of elimination
    are excretion of hydrogen cyanide through the lungs and binding
    to cysteine or hydroxocobalamin.

    .Metabolic Detoxification of Cyanide;V02ANnew.BMP


    The detoxification of cyanide occurs slowly at the rate of
    0.017 mg/kg per min (McNamara, 1976).  A sulfurtransferase
    enzyme is needed to catalyse the transfer of a sulfur atom
    from the donor thiosulfate to cyanide.  The classical theory
    indicating that mitochondrial thiosulfate sulfurtransferase
    is the most important enzyme in this reaction is now in
    doubt because thiosulfate penetrates lipid membranes slowly
    and would, therefore, not be readily available as a
    source of sulfur in cyanide poisoning.  The modern concept assumes a
    greater role for the serum albumin-sulfane complex, which is the
    primary cyanide detoxification buffer operating in normal metabolism
    (Sylvester et al., 1983).  A further enzyme, beta-mercaptopyruvate
    sulfurtransferase, also converts cyanide to thiocyanate (Vesey et
    al., 1974).  This enzyme is found in the erythrocytes, but in human
    cells its activity is low.

    1.7.1  Thiocyanate toxicity

         The detoxification product of cyanide, thiocyanate, is excreted
    in the urine.  Thiocyanate concentrations are normally between
    1-4 mg/l in the plasma of non-smokers and 3-12 mg/l in smokers.  The
    plasma half-life of thiocyanate in patients with normal renal
    function is 4 h (Blaschle & Melmon, 1980), but in those with renal

    insufficiency it is markedly prolonged and these patients are
    therefore at increased risk of toxicity (Schulz et al.,  1978). 
    Thiocyanate levels exceeding 100 mg/l are thought to be associated
    with toxicity.  Thiocyanate toxicity is characterized by weakness,
    muscle spasm, nausea, disorientation, psychosis, hyper-reflexia, and
    stupor (Smith, 1973; Michenfelder & Tinker, 1977).  Lethal poisoning
    at concentrations greater than 180 mg/l has been reported (Healy,
    1931; Garvin, 1939; Russel & Stahl, 1942; Kessler & Hines, 1948;
    Domalski et al., 1953).  Haemodialysis is recommended as an
    effective means of removing thiocyanate (Marbury et al., 1982). 
    Dialysance values of 82.8 ml/min ( in vivo) and 102.3 ml/min
    ( in vitro) have been recorded (Pahl & Vaziri, 1982).  Little is
    known about the protein-binding characteristics of thiocyanate, and
    haemoperfusion may be more effective than haemodialysis.

    1.8  Protective Measures for Occupational Exposure

         Accidental exposure to cyanide, as either hydrogen cyanide or
    cyanide salts, will occur primarily in the occupational context, and
    appropriate preventive and protective measures need to be taken
    wherever cyanides are manufactured or used.  Many industrial accidents
    occur as a result of mixing cyanide salts and acids, and care
    should be taken when both are present on industrial premises. 
    As hydrogen cyanide may be generated during combustion of organic
    substances, fire fighters may also be exposed occupationally.

         The public may be affected in the case of a major industrial
    emergency, or of a transport accident, involving the release of
    cyanides.  It is essential for local authorities in areas where
    cyanides are used to have contingency plans that will enable them to
    respond effectively.  Adequate hospital facilities for treatment of
    casualties must be available.

         Proper maintenance  of plant, good operating practice, and
    industrial hygiene are essential for the prevention of cyanide
    poisoning.  Areas in the workplace where cyanides are used and
    containers for storage and transport of cyanide should be clearly
    marked.  Work schedules should ensure that there are at least two
    people in zones where cyanide could be released accidentally. There
    should be showers and first-aid kits in these areas.  Personnel
    without proper training should not be allowed in the plant.  Normal
    industrial and laboratory hygiene measures for personnel handling
    toxic materials, such as dirty and clean locker facilities and
    showers, should be provided.  Eating, drinking, and smoking should
    not be allowed in the work area where cyanides are used but in
    places specially reserved for these purposes.

         Each employee working at a plant or laboratory that handles
    cyanides, should receive instruction on the dangers of cyanides and
    be trained in appropriate first-aid measures, as should
    emergency-service personnel.  They should be aware of the hazards
    and informed about the possible routes of exposure (inhalation, skin
    absorption, ingestion).  Training should involve recognition of the
    symptoms and signs of cyanide poisoning and how to achieve safe

    removal of victims from the source of intoxication.  Personnel
    should also be able to guide a rescue or fire-fighting team to a
    trapped intoxicated person.  Rescue personnel should be able to put
    on protective clothing quickly in an emergency.  There should be
    regular instruction sessions covering procedures for handling
    cyanides and for rescue in case of accidents, as well as random
    alarm exercises.  First-aid training should include the essential
    measures to be taken before medical help arrives, which may need to
    be undertaken at the same time as removal of contaminated clothing
    and decontamination of exposed skin and eyes.  It should be realized
    that further uptake of cyanide into the blood may occur after
    showering because of continued skin absorption.

         Each plant handling cyanide should have its own medical staff
    trained in the emergency treatment of cyanide poisonings.  The
    atmospheric concentrations of hydrogen cyanide should be monitored
    in plants where the gas is used or may be generated.  Warning
    devices are available for this purpose and should be installed.  In
    certain circumstances in which cyanide is used, it is possible to
    add a warning gas, e.g., cyanogen chloride and chloropicrin have
    been added to hydrogen cyanide used as a fumigant (Cousineau & Legg,
    1935; Polson & Tattersall, 1969).

         Filter respirators should be carried at all times by employees
    working in zones where hydrogen cyanide may be released.  At high
    hydrogen cyanide concentrations, absorption occurs through the skin
    and impermeable butyl rubber protective clothing is required. 
    Oxygen breathing apparatus may be needed.

         In the case of an accident involving hydrogen cyanide there
    should be both an acoustic and a visual alarm for the plant, which
    may be activated by workers in zones where the gas is used.  Each
    worker should be aware of the emergency procedures to be followed
    and the protective clothing and equipment to be used.  If a large
    number of victims is involved or if there is a danger to the public,
    local authorities need to be warned, so that contingency plans are
    put into effect and hospitals alerted.

         For accidents at plants in remote areas where a qualified
    physician is not readily available and there are no hospital
    intensive care facilities, attending paramedical personnel should
    have the authority and training to perform the special resuscitation
    measures involved in treating cyanide poisonings, including rapid
    endotracheal intubation and techniques for obtaining intravenous
    access.

    1.9  Treatment

    1.9.1  Supportive treatment

         Although effective antidotes are available, general supportive
    measures should not be ignored and may be life-saving.

         According to Jacobs (1984), who reported his personal
    experience of 104 industrial poisoning cases, the use of specific
    antidotes was indicated only in cases of severe intoxication with
    deep coma, wide non-reactive pupils, and respiratory insufficiency
    in combination with circulatory insufficiency.  In patients with
    moderately severe poisoning, who had suffered only a brief period of
    unconsciousness, convulsions, vomiting, and cyanosis, therapy
    consisted of intensive care and intravenous sodium thiosulfate.  In
    cases of mild intoxication with dizziness, nausea, and drowsiness,
    rest and oxygen alone were used.

         Peden et al. (1986) described nine patients poisoned by
    hydrogen cyanide released by a leak from a valve.  Three of them
    were briefly unconscious but recovered rapidly after being moved
    from the area where they had been working.  The arterial whole-blood
    cyanide concentrations on admission were 3.5, 3.1 and 2.8 mg/l,
    respectively.  The cyanide concentrations in the other cases ranged
    between 2.6 and 0.93 mg/l.  All recovered with supportive therapy
    alone.

         Between 1970 and 1984, three other men were treated similarly; 
    two were transiently unconscious, and in these cases the cyanide
    concentrations 30 min after exposure were 7.7 and 4.7 mg/l.  The
    concentration in the other patient was 1.6 mg/l.  All three patients
    recovered without the use of cyanide antidotes.  Small numbers of
    comatose patients with potentially lethal blood concentrations on
    admission, and who recovered without cyanide antidotes, have been
    reported by Graham et al. (1977), Edwards & Thomas (1978), and Vogel
    et al. (1981).

         Even if a patient is unconscious, an antidote does not
    necessarily have to be administered immediately unless vital signs
    deteriorate.

         A patient exposed to hydrogen cyanide who reaches hospital
    fully conscious is only likely to require observation and
    reassurance.

    1.9.2  Antidotal treatment

    1.9.2.1  Oxygen

         It is difficult to understand how oxygen has a favourable
    effect in cyanide poisoning, because inhibition of cytochrome
    oxidase is non-competitive.  However, oxygen has always been
    regarded as an important first-aid measure in cyanide poisoning, and
    there is now experimental evidence that oxygen has specific
    antidotal activity.  Oxygen accelerates the reactivation of
    cytochrome oxidase and protects against cytochrome oxidase
    inhibition by cyanide (Takano et al., 1980).  Nevertheless, there
    are other possible modes of action and those that are clinically
    important have yet to be determined.

         Hyperbaric oxygen is recommended for smoke inhalation victims
    suffering from combined carbon monoxide and cyanide poisoning, since
    these two agents are synergistically toxic.  The use of hyperbaric
    oxygen in pure cyanide poisoning remains controversial.

    1.9.2.2  Sodium thiosulfate

         The major route of cyanide detoxification in the body is
    conversion to thiocyanate by rhodanese, although other
    sulfurtransferases, such as beta-mercaptopyruvate sulfurtransferase,
    may also be involved.  This reaction requires a source of sulfane
    sulfur, but endogenous supplies of this substance are limited. 
    Cyanide poisoning is an intramitochondrial process and an
    intravenous supply of sulfur will only penetrate mitochondria
    slowly.  While sodium thiosulfate may be sufficient alone in mild to
    moderately severe cases, it should be administered with other
    antidotes in cases of severe poisoning.  It is also the antidote of
    choice when the diagnosis of cyanide intoxication is not certain,
    for example in cases of smoke inhalation.  Sodium thiosulfate is
    assumed to be intrinsically nontoxic but the detoxification product
    formed from cyanide, thiocyanate, may cause toxicity in patients
    with renal insufficiency (see section 1.7).

    1.9.2.3  Amyl nitrite

         The administration of amyl nitrite by inhalation has been used
    for many years as a simple first-aid measure that generates
    methaemoglobin and which can be employed by lay personnel.  Its use
    was abandoned because the methaemoglobin concentration obtained with
    amyl nitrite is no more than 7% and it is thought that at least 15%
    is required to bind a potentially lethal dose of cyanide.  However,
    recent studies suggest that methaemoglobin formation plays only a
    small role in the therapeutic effect of amyl nitrite, and
    vasodilatation may be the most important mechanism of antidotal
    action.  Artificial respiration with amyl nitrite ampoules broken
    into an Ambu bag proved to be life-saving in dogs severely poisoned

    with cyanide.  Amyl nitrite may therefore be reintroduced as a
    first-aid measure.

    1.9.2.4  Sodium nitrite

         Nitrites generate methaemoglobin, which combines with cyanide
    to form the nontoxic substance cyanmethaemoglobin.  Methaemoglobin
    does not have a higher affinity for cyanide than does cytochrome
    oxidase, but there is a much larger potential source of
    methaemoglobin than there is of cytochrome oxidase.  The efficacy of
    methaemoglobin is therefore primarily the result of mass action.  A
    drawback of methaemoglobin generation is the resultant impairment of
    oxygen transport to cells and, ideally, the total amount of free
    haemoglobin should be monitored to ensure aerobic metabolism of the
    cells.  Methaemoglobin can be measured very quickly, but this in
    itself will not provide an accurate guide to the amount of
    haemoglobin available for oxygen transport because the
    cyanmethaemoglobin concentration is not taken into account.  
    Individuals deficient in glucose-6-phosphate dehydrogenase (G6PD)
    are at great risk from sodium nitrite therapy because of the
    likelihood of severe haemolysis, but the risk from amyl nitrite is
    likely to be less because only low plasma concentrations are
    achieved.  Excess methaemoglobinaemia may be corrected with either
    methylene or toluidine blue (see Chapter 9) or, preferably, where
    feasible, by exchange transfusion.

    1.9.2.5  4-Dimethylaminophenol (4-DMAP)

         4-DMAP generates a methaemoglobin concentration of 30-50%
    within a few minutes (Weger, 1968) and, theoretically, it should
    therefore be valuable as a first-aid measure.  However, the problems
    associated with methaemoglobin formation, as described above for
    nitrites, apply to 4-DMAP to an even greater extent.  Furthermore,
    it has very poor dose-response curve reproducibility.  Haemolysis as
    a result of 4-DMAP therapy has been observed in overdose as well as
    following a correct therapeutic dose.  Treatment with 4-DMAP is
    contraindicated in patients with G6PD deficiency.  Excess
    methaemoglobinaemia may be corrected with either methylene or
    toluidine blue (see section 1.9.2.8).

    1.9.2.6  Hydroxocobalamin (vitamin Bl2a)

         This antidote binds cyanide strongly to form cyanocobalamin
    (vitamin B12) and, compared to nitrite and 4-DMAP therapy, it has
    the great advantage of not interfering with tissue oxygenation.  The
    disadvantage of hydroxocobalamin as a cyanide antidote is the large
    dose required for it to be effective.  Detoxification of 1 mmol
    cyanide (corresponding to 65 mg KCN) needs 1406 mg hydroxocobalamin. 
    In most countries it is only commercially available in formulations
    of 1-2 mg per ampoule.  In some countries, e.g., France, a
    formulation is available that contains 4 g hydroxocobalamin powder

    that has to be reconstituted with 80 ml of a 10% sodium thiosulfate
    solution prior to use and administered intravenously in a minimum of
    220 ml of 5% dextrose.  Recorded side effects are anaphylactoid
    reactions and acne.  Some authors have reported a reduced antidotal
    effect as a result of mixing hydroxocobalamin and sodium thiosulfate
    in the same solution (Evans, 1964; Friedberg & Shukla, 1975). 
    Histological changes in the liver, myocardium, and kidney apparently
    induced by hydroxocobalamin  have been reported in animal
    experiments (Hoebel et al., 1980), but their relevance to man has
    not yet been established.  Transient pink discoloration of mucous
    membranes and urine is an unimportant and nontoxic side-effect.

    1.9.2.7  Dicobalt edetate

         This agent has been shown to be effective in the treatment of
    cyanide poisoning in man, and in the United Kingdom it is the
    current treatment of choice provided that cyanide toxicity is
    definitely present.  This is a strict criterion, because as a result
    of the manufacturing process some free cobalt ions are always
    present in solutions of dicobalt edetate.  Cobalt ions are toxic and
    the use of dicobalt edetate, in the absence of cyanide, will lead to
    serious cobalt toxicity.  There is evidence from animal experiments
    that glucose protects against cobalt toxicity and it is recommended
    that this be given at the same time as dicobalt edetate.  Serious
    adverse effects recorded include vomiting, urticaria, anaphylactic
    shock, hypotension, and ventricular arrhythmias (Hilmann et al.,
    1974; Naughton, 1974).

    1.9.2.8  Antidotes to methaemoglobin-forming agents

         Accurate determination of methaemoglobin and free haemoglobin
    concentrations in the presence of cyanide is difficult. 
    Nevertheless, excess methaemoglobinaemia does undoubtedly occur on
    occasions following the use of nitrites and 4-DMAP.  Excess
    methaemoglobin concentrations may be reduced by methylene or
    toluidine blue.  However, regeneration of haemoglobin will release
    cyanide back into the circulation, leading to a recurrence of
    toxicity.

    1.10  Summary of Treatment Recommendations

         The management of cyanide poisoning is determined by (i) the
    nature of exposure, i.e. hydrogen cyanide (with or without carbon
    monoxide), cyanide salts, aliphatic nitriles, cyanogenic glycosides;
    (ii) the severity of poisoning; (iii) the number of patients
    involved; (iv) the proximity of hospital facilities; (v) the
    presence of risk factors, e.g., G6PD deficiency.  Urgent specific
    antidotal therapy is not indicated unless the patient is in a deep
    coma, with dilated non-reactive pupils and deteriorating
    cardio-respiratory function.  A patient exposed to hydrogen cyanide

    confwho reaches hospital fully conscious requires observation and
    reassurance only.

         In order to assess the severity of cyanide poisoning, it is
    necessary to take a blood sample before the administration of
    antidotes.  Analytical results are otherwise unreliable.

    1.10.1  First aid and treatment measures at the site of the incident

         The doses given are for adults.  Model Information Sheets
    should be consulted for the pediatric dose and for the use of
    antidotes in special-risk groups, e.g., G6PD-deficient patients.

         The following should be undertaken:

         (a)  Trained personnel (wearing appropriate protective clothing
              and breathing apparatus if hydrogen cyanide or liquid
              cyanide preparations are involved) should

              *    terminate further exposure
              *    commence artificial ventilation with 100% oxygena
              *    administer 0.2-0.4 ml amyl nitrite via Ambu bag

         (b)  A physician (if immediately present on the scene) should

              *    terminate further exposure
              *    artificial ventilation with 100% oxygena
              *    administer 0.2-0.4 ml amyl nitrite via Ambu bag

         In cases of unequivocal moderate to severe poisoning, the above
    procedure should be followed by

                        50 ml of 25% sodium thiosulfate solution
                        (12.5 g) i.v. for 10 minutes

            and either   20 ml of 1.5% dicobalt edetate solution
                        (300 mg) i.v. for 1 minute

               or      10 ml of 40% hydroxocobalamin solution (4 g)
                        i.v. for 20 minutes

               or      10 ml of 3% sodium nitrite solution (300 mg)
                        i.v. for 5-20 minutes

               or        5 ml of 5% 4-DMAP solution (250 mg or
                        3-4 mg/kg) i.v. for 1 minute

                 

    a  Oxygen should be administered using a mask and a bag with a
         "non-return" valve to prevent inspiration of exhaled gases.

    1.10.2  Hospital treatmenta

         The doses given are for adults.  Model Information Sheets
    should be consulted for the pediatric doses and for the use of
    antidotes in special-risk groups, e.g., G6PD-deficient patients.

    1.10.2.1  Severe poisoning

         Patients in deep coma with dilated non-reactive pupils and
    deteriorating cardio-respiratory function (blood cyanide
    concentrations 3 to 4 mg/l) should be given

              *    artificial ventilation with 100% oxygenb

              *    cardio-respiratory support

    This should be followed by

                        50 ml of 25% sodium thiosulfate solution
                        (12.5 g) i.v. over 10 min

           and  either   20 ml of 1.5% dicobalt edetate solution
                        (300 mg) i.v. over 1 min

               or        10 ml of 40% hydroxocobalamin solution (4 g)
                        i.v. over 20 min

               or        10 ml of 3% sodium nitrite solution (300 mg)
                        i.v. over 5-20 min

               or        5 ml of 5% 4-DMAP solution (250 mg or
                        3-4 mg/kg) i.v. over 1 min

	a  Hospital physicians must establish whether specific antidotal
         therapy was administered at the time of the incident before
         further doses are administered, especially in the case of
         methaemoglobin-forming agents.

    b   Oxygen should be administered using a mask and a bag with a
        "non-return" valve to prevent inspiration of exhaled gases.					
						
    						
    1.10.2.2  Moderately severe poisoning

         Patients who have suffered a short-lived period of
    unconsciousness, convulsions, vomiting, and/or cyanosis (blood
    cyanide concentrations 2-3 mg/l) should be given   

              *    100% oxygen, but for no longer than 12-24 h
              *    50 ml of 25% sodium thiosulfate solution (12.5 g)
                   i.v. over 10 min
              *    observation in an intensive-care area

    1.10.2.3  Mild poisoning

         Patients with nausea, dizziness, drowsiness only (blood cyanide
    concentrations < 2 mg/l) should be given

              *    oxygen
              *    reassurance
              *    bed rest

         It should be noted that severely poisoned patients may
    occasionally fail to respond to the initial dose of a specific
    antidote.  Whilst repeat doses of hydroxocobalamin and/or sodium
    thiosulfate are unlikely to be associated with toxicity, expert
    advice should be sought before a repeat dose of any other specific
    antidote is administered.  Intensive supportive therapy is of
    paramount importance in these circumstances.

    1.11  Summary of Analytical Aspects

         There are many reliable methods for the detection and
    qualitative determination of cyanide in biological material in cases
    of suspected intoxication (see Chapter 10).  They can be used as
    "bedside methods" as well as for qualitative determination in cases
    of acute poisoning but only before antidotes are administered. 
    Interference results from the presence of thiosulfate,
    methaemoglobin, thiocyanates, and chelating agents during the course
    of whole-blood cyanide analysis.  For this reason, it may be more
    appropriate to measure plasma rather than whole-blood cyanide
    concentrations.  However, the pronounced instability of cyanide in
    plasma is a serious drawback (Lundquist et al., 1985).

         Quantitative analysis of cyanide in blood or serum before the
    administration of antidotes is a useful means of evaluating the
    severity of poisoning.  Evaluation of the efficacy of different
    antidotes will not be possible before accurate methods of analysis
    free from interference are developed.

         When methaemoglobin-generating agents (nitrites or 4-DMAP) are
    administered as antidotes in cyanide poisoning, it is necessary to
    maintain an adequate concentration of free haemoglobin in order to
    guarantee sufficient oxygen transport to allow aerobic tissue
    metabolism.

         Special instruments for rapid analysis of methaemoglobin in
    hospitals do not provide information about the amount of haemoglobin
    available for oxygen transport, because the multicomponent analysis
    is invalidated by the presence of a haemoglobin derivate
    (cyanmethaemoglobin).  Since there is no satisfactory means of
    quantifying cyanmethaemoglobin under these circumstances, therapy
    with methaemoglobin-generating agents cannot be monitored at present
    by laboratory methods.

    1.12  Proposed Areas for Research

         There are two areas of research where further work is needed as
    a matter of urgency:

         (a)  Analytical techniques currently available for the
              measurement of methaemoglobin do not permit accurate
              estimation of the amount of free haemoglobin available for
              oxygen transport, because cyanmethaemoglobin cannot be
              quantified.  A rapid and accurate technique for measuring
              methaemoglobin and cyanmethaemoglobin concentrations in
              conjunction is therefore needed to monitor the use of
              methaemoglobin-generating cyanide antidotes.

         (b)  Reliable quantitative analytical methods for cyanide in
              whole blood in the presence of one or more antidotes are
              needed.

         (c)  Determination of cyanide concentration in plasma or serum
              may be the best reflection of the tissue concentration of
              cyanide, since cyanide trapped in erythrocytes will not
              affect tissue utilization of oxygen.  However, cyanide has
              been shown to be very unstable in these body fluids.  A
              method to prevent this phenomenon is urgently needed.

         (d)  The intravenous injection of DMAP generates high
              concentrations of methaemoglobin within minutes.  However,
              the absorption kinetics of DMAP administered
              intramuscularly are not known with certainty, particularly
              in patients who are shocked with poor muscle perfusion. 
              The efficacy of intramuscular DMAP as a first-aid measure
              in cases of severe cyanide poisoning needs further
              evaluation.

         (e)  Hydroxocobalamin has recently been reported to cause
              histological changes in the liver, kidney, and myocardium
              of animals.  The relevance of these findings to man is not
              known and further investigation is required.

         (f)  Enzyme systems other than cytochrome oxidase may be
              inhibited.  This may be the cause for the symptomatology
              in acute severe cyanide poisoning.

    1.13  New Developments in Cyanide Antidotes

         Currently available cyanide antidotes have potentially
    undesirable adverse effects, and none has been successful in all
    cases of acute, severe cyanide poisoning.  Various agents for the
    treatment of cyanide poisoning are at the experimental stage of
    development.  However, these antidotes are not currently recommended
    for administration in cases of human poisoning.

    1.13.1  Nonspecific agents

         Based on animal studies, certain nonspecific agents, such as
    naloxone in huge doses (equivalent to 700 mg in a 70 kg human adult
    compared with a usual therapeutic dose of 0.4 to 10 mg) (Leung et
    al., 1986) or alpha-adrenergic blocking agents such as
    chlorpromazine (which has no beneficial effect when administered
    alone but variably enhances sodium nitrite and/or sodium thiosulfate
    efficacy) (Kong et al., 1983; Petterson & Cohen, 1985), have been
    suggested as adjunctive therapy.  However, at the moment, there is
    no accepted place for the use of these agents in human poisoning.

    1.13.2  Sodium pyruvate

         This agent re-establishes cellular respiration in tumour
    tissues inactivated by cyanide and has some efficacy in experimental
    animal poisoning.  It may promote cyanide detoxification through
    combination of the cyanide anion with a carbonyl radical, producing
    cyanohydrin (Pronczuk de Garbino & Bismuth, 1981).  Sodium pyruvate
    acts rapidly and is well distributed to tissues, but clinical trials
    in human cyanide poisoning have not been undertaken.

    1.13.3  Ifenprodil

         Ifenprodil is a 2-piperidine allonal derivative, which, in
    experimental poisoning, affords some protection including decreased
    respiratory distress, improved blood pressure, normalization of
    cardiac rhythm, and lessened electroencephalographic abnormalities. 
    The mechanism of action is thought to be a direct stimulation of
    mitochondrial respiratory function.  At present, ifenprodil is in
    the investigational stage and no human clinical trials have been
    proposed (Pronczuk de Garbino & Bismuth, 1981).

    1.13.4  Rhodanese

         Rhodanese (thiosulfate-cyanide sulfurtransferase) is the
    naturally-occurring cyanide-detoxifying enzyme (see section 1.7). 
    Although the availability of sulfane sulfur is the rate-limiting
    factor, studies in dogs have indicated that there is enough
    rhodanese present in the normal liver and muscle tissue to detoxify
    about 500 grams of cyanide.  When derived from hepatic tissue, the
    enzyme is unstable and requires an optimal pH for cyanide

    detoxification.  Bacterial enzyme, derived from cultures of
     Thiobacillus  denitrificans, is more stable and has been studied
    in experimental animals.  It is efficacious in experimental cyanide
    poisoning, but no human clinical applications have yet been proposed
    (Pronczuk de Garbino & Bismuth, 1981).

    1.13.5  Alpha-ketoglutaric acid

         The cyanide ion can react with carbonyl groups to form
    cyanohydrins, and this could represent an important detoxification
    reaction.  In rodents poisoned with cyanide and pretreated with
    various antidotes, alpha-ketoglutaric acid was more effective than
    either sodium nitrite or sodium thiosulfate, and nearly as effective
    as sodium nitrite and sodium thiosulfate in combination (Moore et
    al., 1986).  The combination of alpha-ketoglutaric acid with sodium
    nitrite plus sodium thiosulfate increased the cyanide LD50 from a
    mean of 6.7 mg/kg in control animals to 119.4 mg/kg, whereas the
    sodium nitrite/thiosulfate combination alone increased the mean
    LD50 to only 35.0 mg/kg (alpha-ketoglutaric acid alone increased
    the mean LD50 to 33.3 mg/kg).  No methaemoglobin induction was
    observed with alpha-ketoglutaric acid administration.  Some tremors
    were noted when this agent was administered alone.  Tremors did not
    occur when sodium thiosulfate was added to the treatment regimen,
    and the LD50 value was increased to 101.3 mg/kg with this
    combination (very close to the 19.4 mg/kg LD50 observed with
    addition of both sodium nitrite and sodium thiosulfate to
    alpha-ketoglutaric acid).  While these studies demonstrated only
    protective activity with prophylactic alpha-ketoglutaric acid
    administration, they raise the possibility of another potentially
    efficacious and safe antidote combination with sodium thiosulfate
    (Moore et al., 1986).  Alpha-ketoglutaric  acid, especially in
    combination with sodium thiosulfate, deserves further study.

    1.13.6  Stroma-free methaemoglobin solution

         Stroma-free methaemoglobin solution is prepared from outdated
    red blood cells by the removal of all cellular membranes (stroma)
    and induction of methaemoglobinaemia equivalent to 90% of the total
    haemoglobin with potassium ferricyanide.  Any excess potassium
    ferricyanide is then removed by dialysis against saline.  The
    resultant preparation does not contain the antigenic components that
    have been previously reported to cause renal failure and
    coagulopathies.  No rats given stroma-free methaemoglobin solution
    alone died or had any adverse reactions.  Concentrated stroma-free
    methaemoglobin solution (200-300 g/l) was an effective experimental
    antidote when administered 30 seconds after doses of cyanide up to 6
    times the LD90.  More dilute solutions were effective 90% of the
    time following cyanide administration up to 4 times the LD90. 
    None of the animals in these studies were given any supportive
    therapy.  In animals administered an amount of stroma-free
    methaemoglobin solution thought to be equivalent to the conversion

    of 1.5% of the endogenous haemoglobin, a 90% survival rate was noted
    when a cyanide LD100 was administered.  Spectroscopic examination
    of urine revealed cyanmethaemoglobin excretion (Ten Eyck et al.,
    1984, 1985, 1986).

         These studies suggest that methaemoglobin prepared exogenously
    may be an effective cyanide antidote.  Exogenously administered
    methaemoglobin would not be expected to interfere with oxygen
    transport and, unlike methaemoglobin-generating agents (Moore et
    al., 1987), could even be used in smoke-inhalation victims with
    elevated carboxyhaemoglobin levels.  Since no adverse effects have
    been noted, this agent may be a safe alternative to currently
    available cyanide antidotes.  However, no human studies have been
    undertaken and extensive animal toxicology experiments have not yet
    been reported.

    1.14  References

    Alarie Y (1985) The toxicity of smoke from polymeric materials
    during thermal decomposition. Am Rev Pharmacol Toxicol, 25: 325-347.

    Anderson RA & Harland WA (1982) Fire deaths in the Glasgow area.
    III. The role of hydrogen cyanide. Med Sci Law, 22: 35-37.

    Anon (1978) Controlled intravascular sodium nitroprusside treatment.
    Br Med J, 6140: 784-785.

    Antal J (1894) [Experimental studies on the treatment of cyanide
    poisoning.] Ung Arch Med, 3:117-128 (in German).

    Atkins D (1977) Cyanide toxicity following nitroprusside-induced
    hypotension. Can Anaesth Soc J, 24: 651-660.

    Berlin AM (1971) The treatment of cyanide poisoning in children.
    Pediatrics, 46: 793.

    Birky MM, Halpin BM, Caplan YH, Fisher RS, McAllister JM, & Dixon AM
    (1979) Fire fatality study. Fire Mater, 3: 211-217.

    Blake G (1839) Observations on the physiological effects of various
    agents introduced into the circulation, as indicated by the
    hemodynamometer. Edinburgh Med Surg J, 53: 35-49.

    Blake G (1840) Observations and experiments in the mode in which
    various poisonous agents act on the animal body. Edin Med Surg J 53:
    35-49.

    Blanc P, Hogan M, Malin K, Hryhorczuk D, Hessl S, & Bernard B (1985)
    Cyanide intoxication among silver reclaiming workers. J Am Med
    Assoc, 253: 367-371.

    Blaschle TE & Melmon KL (1980) Antihypertensive agents and the drug
    therapy of hypertension. In: Goodman LS & Gilma A ed. The
    pharmacological basis of therapeutics. New York, MacMillan
    Publishing Co., vol 6, pp 807-808.

    Burrows GE & Way JL (1976) Antagonism of cyanide toxicity by
    phenoxybenzamine. Fed Proc, 35: 533.

    Chen KK, Rose CL, & Clowes GHA (1933) Methylene blue, nitrites and
    sodium thiosulfate against cyanide poisoning. Proc Soc Exp Biol Med,
    31: 250-252.

    Chen KK, Rose CL, & Clowes GHA (1934) Comparative values of several
    antidotes in cyanide poisoning. Am J Med Sci, 188: 767.

    Chen KK, Rose CL, & Clowes GHA (1944) The modem treatment of cyanide
    poisoning. J Indiana Med Assoc, 37: 344-350.

    Clark CJ, Campbell D, & Reid WH (1983) Blood carboxyhaemoglobin and
    cyanide levels in fire survivors. Lancet, 1: 1332-1335.

    Conn EE (1973) Cyanogenic glucosides (Committee on Food Protection,
    Food and Nutrition Board, National Research Council). In: Toxicants
    occurring naturally in foods, 2nd ed. Washington, DC, National
    Academy of Sciences, pp 299-308.

    Cook RD & Coursey DB (1981) Cassava: a major cyanide-containing food
    crop. In: Vennesland B, Conn E, Knowles CJ, Westley J, & Wissing F
    ed. Cyanide in biology. New York, London, Academic Press, pp 11-28.

    Cousineau A & Legg FG (1935) Hydrocyanic acid gas and other toxic
    gases in commercial fumigation. Am J Public Health, 25: 277-294.

    Domalski CA, Kolb LC, & Hines EA (1953) Deleterious reactions
    secondary to thiocyanate therapy of hypertension. Proc Mayo Clin,
    28: 272-280.

    Edwards AC & Thomas IC (1978) Cyanide poisoning. Lancet, 1: 92.

    El Ghawabi SH, Gaafar MA, El Saharti AA, Ahmed SH, Malash KK, & Fares
    R (1975) Chronic cyanide exposure, a clinical, radio isotope and
    laboratory study. Br J Ind Med, 32: 215-219.

    Evans CL (1964) Cobalt compounds as antidotes for hydrocyanic acid.
    Br J Pharmacol Chemother, 23: 455-475.

    Fontana F (1795) Treatise on the venom of viper; on the American
    poisons; and on the Cherry Laurel and some other vegetable poisons.
    J Skinner (trans), London.

    Friedberg KD & Shukla UR (1975) The efficiency of aquocobalamin as
    an antidote in cyanide poisoning when given alone or combined with
    sodium thiosulfate. Arch Toxicol, 33: 103-113.

    Fukumoto Y, Nakajimo H, Uetake M, Matusyama A, & Yoshida T (1957) A
    study on the sense of smell with respect to potassium cyanide solution
    and its hereditary transmission. Jpn J Hum Genet, 2: 7-16.

    Garvin CF (1939) The fatal toxic manifestations of the thiocyanates.
    J Am Med Assoc, 112: 1125-1127.

    Goldfrank L, Bresnitz EA, Weisman RS, & Lewin NA (1984) The inhaled
    agents and other disorders of oxygen transport. In: Hanson W Jr ed.
    Toxic emergencies. New York, Churchill Livingstone, pp 204-211.

    Gosselin RE, Smith RP, & Hodge HC (1984) Cyanide. In: Clinical
    toxicology of commercial products. Baltimore, London, Williams &
    Wilkins Co., pp 123-130.

    Graham DL, Laman D, Theodore J, & Robin ED (1977) Acute cyanide
    poisoning complicated by lactic acidosis and pulmonary edema. Arch
    Intern Med, 137: 1051.

    Grant WM (1980) The peripheral visual system as a target. In: Spencer
    PS & Schaumberg HH ed. Experimental and clinical neurotoxicology.
    Baltimore, London, Williams & Wilkins Co., pp 77-91.

    Healy JC (1931) Therapeutics and toxicology of sulfocyanates. New Engl
    J Med, 205: 5481-5583.

    Hilmann B, Bardham, KD, & Bain JTB (1974) The use of dicobalt edetate
    (Kelocyanor) in cyanide poisoning. Postgrad Med J, 50: 171-174.

    Hoebel M, Engeser P, Nemeth L, & Pill J (1980) The antidote effect of
    thiosulfate and hydroxocobalamine in formation of nitroprusside
    intoxication of rabbits. Arch Toxicol, 46: 207-13.

    Hoppe-Seyler F (1876) [On fermentation processes and their
    relationship to the life of organisms.] PfQgers Arch gesamte Physiol
    Menschen Tieren, 12: 1-17 (in German).

    Jacobs K (1984) [Report on experience with the administration of
    4-DMAP in severe prussic acid poisoning. Consequences for medical
    practice.] Zentralbl Arbeitsmed, 34: 274-277 (in German).

    Kalmus H & Hubbard DJ (1960) The chemical senses in health and
    disease. Springfield, Illinois, Charles C. Thomas.

    Kalyanaraman UP, Kalyanaraman K, & Cullinan SA (1983) Neuropathy of
    cyanide intoxication due to "laetrile" (amygdalin). Cancer, 51:
    2126-2133.

    Kessler DL & Hines LE (1948) Hazards of thiocyanate therapy in
    hypertension. J Am Med Assoc, 138: 549-551.

    Kirk RI & Stenhouse NS (1953) Ability to smell solutions of
    potassium cyanide. Nature (Lond), 171: 698-699.

    Kong A, Shen A, Burrows G, Sylvester D, Ison GE, & Way JL (1983)
    Effect of chlorpromazine on cyanide intoxication. Toxicol Appl
    Pharmacol, 71: 407-413.

    Leung P, Sylvester DM, Chiou F, Way LI, Way EL, & Way JL (1986)
    Stereospecific effect of naloxone hydrochloride on cyanide
    intoxication. Toxicol Appl Pharmacol, 83: 525-530.

    Levine MS, Radford MPH, & Radford EP (1978) Occupational exposures
    to cyanide in Baltimore fire fighters. J Occup Med, 20: 53-56.

    Lowry WT, Juarez L, Petty CS, & Roberts B (1985) Studies of toxic
    gas production during actual structural fires in the Dallas area. J
    Forensic Sci, 30: 59-72.

    Lundquist P, Rosling H, & Sorbo B (1985) Determination of cyanide in
    whole blood, erythrocytes and plasma. Clin Ther, 31: 591-595.

    McNamara BP (1976) Estimation of the toxicity of hydrocyanid acid
    vapors in man. (Edgewood Arsenal Technical Report No. EB-TR-76023)
    (Army Department).

    MacRae WR & Owen M (1974) Severe metabolic acidosis following
    hypotension induced with sodium nitroprusside. Br J Anaesth, 46:
    795-797.

    Madden T (1731) Poisoning by laurel leaves. Philos Trans R Soc Lond
    Biol Sci, 37: 84-99.

    Magendie F (1817) Sur l'emploi de l'acide prussique dans le
    traitement de plusieurs maladies de poitrine, et particulièrement
    dans la phtisie pulmonaire. Ann Chim Phys, 6: 347-360.

    Marbury TC, Sheppard JE, Gibbons K, & Lee CC (1982) Combined antidotal
    and hemodialysis treatments for nitroprusside-induced cyanide
    toxicity. J Toxicol Clin Toxicol, 19: 475-482.

    Michenfelder JD & Tinker JH (1977) Cyanide toxicity and thiosulfate
    protection during chronic administration of sodium nitroprusside in
    the dogs: correlation with a human case. Anesthesiology, 47:
    441-448.

    Ministry of Health, Mozambique (1984) Mantakassa: an epidemic of
    plastic paraparesis associated with chronic cyanid intoxication in a
    cassava staple area of Mozambique. I.

    Epidemiology and clinical and laboratory findings in patients. Bull
    World Health Organ, 62: 477-488.

    Mladoveanu C & Gheorghiu P (1929) Le nitrite de soude comme antidote
    de l'empoisonnement experimental par le cyanure de potassium. C R
    Soc Biol (Paris), 102: 164-166.

    Montgomery R, Reinhart CF, & Terril JB (1975) Comments on fire
    toxicity. Comb Toxicol, 2: 179-212.

    Moore SJ, Norris JC, Ing KH, & Hume AS (1986) The efficacy of
    alpha-ketoglutaric acid in the antagonism of cyanide intoxication.
    Toxicol Appl Pharmacol, 81: 265-273.

    Moore SJ, Norris JC, Walsh DA, & Hume AS (1987) The antidotal use of
    methemoglobin-forming cyanide antagonists in concurrent carbon
    monoxide/cyanide intoxication. J Pharmacol Exp Ther, 242: 70-73.

    Mushett CW, Kelly KL, Boxer GE, & Rickards JC (1952) Antidotal
    efficacy of vitamin B12a (hydroxocobalamin) in experimental
    cyanide poisoning. Proc Soc Exp Biol Med, 81: 234-237.

    Naughton M (1974) Acute cyanide poisoning. Anesth Intensive Care, 4:
    351-356.

    Olsen NS & Klein RJ (1947) Effect of cyanide on the concentration of
    lactate and phosphates in brain. J Biol Chem, 167: 739.

    Pahl MV & Vaziri ND (1982) In vivo and in vitro hemodialysis studies
    of thiocyanate. J Toxicol Clin Toxicol, 19: 965-974.

    Paulet G (1960) Les chelatés de cobalt dans le traitement de
    l'intoxication cyanhydrique. Pathol Biol, 8: 255-266.

    Peden NR, Taha A, McSofiey PD, Bryden GT, Murdoch IB, & Anderson JM
    (1986) Industrial exposure to hydrogen cyanide: implications for
    treatment. Br Med J, 293: 538.

    Pedigo LG (1888) Antagonism between amyl nitrite and prussic acid.
    Trans Med Soc Virginia, 19: 124-131.

    Peters CG, Mundy JVB, & Rayner PR (1982) Acute cyanide poisoning.
    Anaesthesia, 37: 582-586.

    Petterson JC & Cohen SD (1985) Antagonism of cyanide poisoning by
    chloro-promazine and sodium thiosulfate. Toxicol Appl Pharmacol, 81:
    265-273.

    Piper J (1975) Use and toxicity of nitroprusside. New Engl J Med,
    292: 1081-1082.

    Polson CJ & Tattersall RN (1969) Clinical toxicology. London,
    Pitman, pp 128-159.

    Pronczuk de Garbino J & Bismuth C (1981) Propositions thérapeutiques
    actuelles en cas d'intoxication par les cyanures. Toxicol Eur Res,
    III(2): 69-76.

    Rosling H (1989) Cassava associated neurotoxicity in Africa. In:
    Proceedings of the 5th International Congress of Toxicology. Volans
    GN, Sims J, Sullivan FM, & Turner P ed. Brighton, Taylor & Francis, pp
    605-614.

    Russel WO & Stahl WC (1942) Fatal poisoning from potassium
    thiocyanate treatment of hypertension. J Am Med Assoc, 119:
    1177-1181.

    Sadoff L, Fuchs J, & Hollander J (1978) Rapid death associated with
    laetrile ingestion. J Am Med Assoc, 239: 1532.

    Scheele CW (1786) Chemical essays. Beddves T (trans) Reprint 1966,
    London, Dawson.

    Schrader JCC (1802) Des expériences de M. Schrader sur l'acide
    prussique continu dans les plantes. J Phys Chim Hist Nat Arts, 56:
    224-226.

    Schulz V, Dohring W, & Rathsac (1978) [Thiocyanate poisoning in
    antihypertensive therapy with sodium nitroprusside.] Klin
    Wochenschr, 56: 355-361 (in German).

    Smith RP (1973) Cyanate and thiocyanate: Acute toxicity. Proc Soc
    Exp Biol Med, 142: 1041-1044.

    Smith RP & Kruszyna H (1974) Nitroprusside produces cyanide
    poisoning via a reaction with hemoglobin. J Chem Exp Ther, 191:
    557-563.

    Sykes AH (1981) Early studies on the toxicology of cyanide. In:
    Vennesland B, Conn EE, Knowles CJ, Westley J, & Wissing F ed.
    Cyanide in biology. New York, London, Academic Press, pp 1-9.

    Sylvester DM, Hayton WL, Morgan RL, & Way JL (1983) Effects of
    thiosulfate on cyanide pharmacokinetics in dogs. Toxicol Appl
    Pharmacol, 69: 265.

    Takano T, Miyzaki Y, Nashimoto I, & Kobayashi K (1980) Effect of
    hyperbaric oxygen on cyanide intoxication: in situ, changes in
    intracellular oxidation reduction. Undersea Biomed Res, 7: 191-197.

    Ten Eyck RP, Schaerdel AD, Lynett JE, Marks DH, Patrissi GA,
    Ottinger WE, & Stansell MJ (1984) Stroma-free methemoglobin solution
    as an antidote for cyanide poisoning: a preliminary study. Clin
    Toxicol, 21: 343-358.

    Ten Eyck RP, Schaerdel AD, & Ottinger WE (1985) Stroma-free
    methemoglobin solution: an effective antidote for acute cyanide
    poisoning. Am J Emergency Med, 3: 519-523.

    Ten Eyck RP, Schaerdel AD, & Ottinger WE (1986) Comparison of
    nitrite treatment and stroma-free methemoglobin solution as
    antidotes for cyanide poisoning in a rat model. Clin Toxicol, 23:
    477-487.

    Trapp WG (1970) Massive cyanide poisoning with recovery. Can Med
    Assoc J, 102: 3517.

    Vesey CJ, Cole PV, Linnell JC, & Wilson J (1974) Some metabolic
    effects of sodium nitroprusside in man. Br Med J, 2: 140-142.

    Vesey CJ, Cole PV, & Simpson PJ (1976) Cyanide and thiocyanate
    concentration following sodium nitroprusside infusion in man, Br J
    Anaesth, 48: 651-659.

    Vick JA & Froelich HL (1985) Studies on cyanide poisoning. Arch Int
    Pharmacodyn, 273: 314-322.

    Vogel SN, Sultan TR, & Ten Eyck RP (1981) Cyanide poisoning. Clin
    Toxicol, 18: 367-383.

    Voorhoeve RJH, Patel CKW, Trimble LE, & Kerl RJ (1975) Automobile
    pollution control devices with malfunctional catalytic converters.
    Science, 190: 149-151.

    Weger N (1968) [Aminophenols as antidotes to prussic acid.] Arch
    Toxikol, 24: 49-50 (in German).

    2.  OXYGEN

    2.1  Introduction

         Oxygen consumption is indispensable for human life.  All organs
    of the body will undergo successive dysfunction, permanent damage,
    and then death if oxygen consumption falls below that necessary for
    metabolic needs.  A decrease in oxygen consumption may be due to
    inadequate oxygen delivery to the cells or to an inability of the
    cells to utilize oxygen.

         The ambient atmospheric pressure at sea level is defined as one
    atmosphere absolute (1 ATA), equivalent to an air pressure of
    1.03 kg per sq. cm., i.e., 101 kPa or 760 mm of mercury.  Molecular
    oxygen constitutes about 21% of atmospheric air.  Hence the partial
    pressure of oxygen at sea level is 0.21 ATA.

         Hyperoxia is the artificial elevation of the partial pressure
    of oxygen in the body and can be produced by increasing the partial
    pressure of oxygen from 0.21 to 1 ATA (normobaric oxygen therapy) or
    above 1 ATA (hyperbaric oxygenation).  Normobaric oxygenation is
    commonly used in many conditions associated with decreased delivery
    of oxygen to tissues, such as cardiac and pulmonary failure and
    cardiopulmonary resuscitation.

         Both normobaric and hyperbaric oxygen therapy have an antidotal
    effect on carbon monoxide (CO), hydrogen sulfide (H2S), and
    cyanide (CN) poisoning.  However, hyperoxia may induce oxygen
    toxicity related to the length of exposure and the partial pressure
    of oxygen employed.  The use of oxygen therapy is therefore limited
    by oxygen toxicity.

    2.2  Name and Chemical Formula of Antidote

         Molecular oxygen is a diatomic molecule and is properly named
    dioxygen (formula O2, relative molecular mass 16).  More than
    99.9% of atmospheric oxygen consists of molecules containing the
    16O isotope.  Trace quantities of 17O and 18O exist in
    atmospheric air.

    2.3  Physico-chemical Properties of Molecular Oxygen
              Critical temperature                 154.575 K
              Critical density                       0.4361 g/cm3
              Critical pressure                     50.14 ATA
              Boiling point                         90.188 K
              Melting point                         54.361 K
              Solubility
                   in water, 25 °C                   2.8%
                   in ethanol                       22.6%
                   in plasma                  approx 3.0%
                   in whole blood                   20.0%

    2.4  Synthesis

         Oxygen is obtained on a large scale by the liquefaction of air. 
    Modern manufacturing processes produce the gas at a concentration 
    much higher than 99.0% (v/v) of oxygen, this being the lowest
    permissible concentration in medical oxygen allowed in the European
    Pharmacopoeia.  Contamination with carbon monoxide or carbon dioxide
    is therefore very slight.

    2.5  Analytical Methods

    2.5.1  Quality control procedures

    2.5.1.1  Tests

         For the tests, deliver the sample to be examined at a rate of
    4 l/h.

          Carbon monoxide Carry out the test using 7.5 l of the
    substance to be examined and 7.5 l of argon R for the blank.  The
    difference between the volumes of sodium thiosulfate (2 mmol/l) used
    in the titrations should not be more than 0.4 ml (5 ppm v/v).

         For the following three tests, pass the sample to be examined
    through the appropriate reagent contained in a hermetically closed
    flat-bottomed glass cylinder (with dimensions such that 50 ml of
    liquid occupies a height of 12 cm to 14 cm) fitted with: (a) a
    delivery tube terminated by a capillary 1 mm in internal diameter
    and reaching to within 2 mm of the bottom of the cylinder; (b) an
    outlet tube.  Prepare the reference solutions in identical
    cylinders.

         Acidity or alkalinity

          Test solution Pass 2.0 litres of the sample to be examined
    through a mixture of 0.1 ml of hydrochloric acid (0.01 mol/l) and
    50 ml of CO2-free water R.

          Reference solution (a)  50 ml of CO2-free water R.

          Reference solution (b)  To 50 ml of CO2-free water R add
    0.2 ml of hydrochloric acid (0.01 mol/l).

         To each solution add 0.1 ml of a 0.02% m/v solution of methyl
    red R in alcohol (70% v/v).  The intensity of the colour of the test
    solution should be between those of reference solutions (a) and (b).

          Carbon dioxide Pass 1.0 litre through 50 ml of barium
    hydroxide solution R (the solution to be used must be clear).  Any
    turbidity in the solution after passage of the gas should not be
    more intense than that in a reference solution prepared by adding

    1 ml of a 0.11% m/v solution of sodium hydrogen carbonate R in
    CO2-free water R to 50 ml of barium hydroxide solution R
    (300 ppm v/v).

          Oxidizing substances Place in each of two cylinders 50 ml
    of freshly prepared potassium iodide and starch solution R, and add
    0.2 ml of glacial acetic acid R.  Protect the cylinders from light. 
    Pass 5.0 l of the substance to be examined into one of the
    cylinders.  The test solution should remain colourless when compared
    with the blank.

    2.5.1.2  Assay for oxygen

         Use a gas burette (see Fig. 1) of 25 ml capacity in the form of
    a chamber having at its upper end a tube graduated in 0.2% divisions
    between 95 and 100, and isolated at each end by a tap with a conical
    barrel.  The lower tap is joined to a tube with an olive-shaped
    nozzle and is used to introduce the gas into the apparatus.  A
    cylindrical funnel above the upper tap is used to introduce the
    absorbent solution.  Wash the burette with water and dry.  Open the
    two taps.  Connect the nozzle to the source of the sample to be
    examined and set the flow rate to 1 l/min.  Flush the burette by
    passing the substance to be examined through it for 1 min.  Close
    the upper tap of the burette and immediately afterwards the lower
    tap.  Rapidly disconnect the burette from the source of the sample
    to be examined and give a half turn to the upper tap to eliminate
    any excess pressure in the burette.  Keeping the burette vertical,
    fill the funnel with a limited amount of freshly prepared mixture of
    21 ml of a 56% m/v solution of potassium hydroxide R and 130 ml of a
    20% m/v solution of sodium dithionite R.  Open the upper tap slowly. 
    The solution absorbs the oxygen and enters the burette.  Allow to
    stand for 10 min without shaking.  Read the level of the liquid
    meniscus on the graduated part of the burette.  This figure
    represents the percentage v/v of oxygen.

    2.5.2  Methods for identification

         (a)  Place a glowing splint of wood in the substance to be
              examined.  The splint bursts into flame.

         (b)  Shake with alkaline pyrogallol solution R.  The sample to
              be examined is absorbed and the solution becomes dark
              brown.

    FIGURE 01

    2.5.3  Methods for analysis of the antidote in biological samples

    2.5.3.1  In the gas phase

         (a)  Chemical volumetric methods based on absorption of oxygen
              by chemical reagents (such as alkaline pyrogallol) are
              still in use.  The skill of the operator is a major factor
              in the accuracy of the results.

         (b)  Methods based on physical properties of oxygen:

              *    paramagnetic susceptibility;
              *    thermomagnetic (magnetic wind);
              *    differential pressure (Quincke);
              *    magnetic auto-balance (Faraday);
              *    electron capture;
              *    ultraviolet absorption;
              *    mass spectrometry.

    2.5.3.2  In solution

         Polarographic measurement is a sensitive method for measuring
    the partial pressure of oxygen in solution.  The Clark electrode is
    commonly used for biological samples.  This has a cathode made of a
    platinum wire, while the anode is Ag/AgCl in phosphate buffer with
    added KCl.

    2.5.4  The saturation of haemoglobin by oxygen

         The degree of saturation of haemoglobin by oxygen (HbO2 %)
    can be measured by photometric procedures.  This method is
    particularly useful for the non-invasive monitoring of patients
    (oximeters, pulse oximeters).

    2.6  Storage Conditions

         Oxygen should be stored under pressure in a suitable metal
    container of a type permitted by the safety regulations of the
    national authority.  Valves and taps should not be lubricated with
    oil or grease.

    FIGURE 02

         The containers for medical oxygen are coded with a white colour
    at the top according to ISO-32-19 (Fig 2) and carry the indication
    "Medical Oxygen" carved in the iron material of the container.  To
    prevent connection with other medical gases, the containers are
    provided with an international standardized "Pin Index Safety
    System" ISO-407.  This system is the same for medical and for
    industrial oxygen so that in emergencies industrial oxygen can also
    be used.

         Regular control of gas pressure is necessary to prevent
    shortage of oxygen in emergency situations.

         Although the duration of storage will not alter the quality, it
    is recommended that the gas should not be stored for more than six
    months.

    2.7  General Properties

         In most circumstances, supplemental oxygen is indicated if
    tissue hypoxia is imminent.  In the case of cyanide poisoning,
    however, cytochrome oxidase activity is inhibited and tissue
    utilization of oxygen prevented.  Even so, animal experiments
    (Takano et al., 1980) suggest that inhibition of cytochrome oxidase
    activity by cyanide is prevented, and recovery accelerated, in the
    presence of oxygen.

    2.8  Animal Studies

    2.8.1  Pharmacokinetics

         The bioavailability of oxygen depends on the following factors:

         (a)  the thickness of the alveolar membrane; as a result of
              oedema fluid in the interstitial spaces of the membrane
              oxygen cannot readily diffuse into the blood;

         (b)  haemoglobin configuration:

              (i)  oxygen transport by the blood to the tissues can be
                   disturbed by methaemoglobin-forming antidotes, e.g.,
                   sodium nitrite or 4-dimethylaminophenol, often used
                   in the treatment of cyanide poisoning;

              (ii) oxygen transport is also hampered in carbon monoxide
                   poisoning, which occurs in combination with cyanide
                   poisoning following smoke inhalation.

         (c)  Oxygen utilization by tissues is prevented by the
              inhibition of cytochrome oxidase activity in cyanide
              poisoning.

    2.8.2  Pharmacodynamics

         Isom & Way (1982) studied the reduction of cytochrome oxidase
    prepared from brains and livers of mice poisoned with potassium
    cyanide.  They examined several groups of animals and compared
    exposure to air and to oxygen at 0.95 ATA, both with and without
    sodium nitrite and sodium thiosulfate treatment.  Cytochrome oxidase
    was inhibited by 75% within 2 min of the injection of 5 KCN mg/kg. 
    The inhibition of cytochrome oxidase was similar whether the animals
    inhaled air or 0.95 ATA O2, but the reactivation phase of
    cytochrome oxidase was faster in oxygen-breathing animals.  They
    also found that, at a lower dose (4 KCN mg/kg), enzyme activity was
    increased from 22% to 66% by increasing the partial pressure of
    oxygen from 0.11 to 0.95 ATA O2.  Oxygen shifted the dose-response
    curve of brain cytochrome oxidase inhibition by KCN to the right: 
    the dose of KCN producing 50% inhibition was 24 mg/kg in
    air-breathing animals and 55 mg/kg in animals breathing oxygen.  The
    authors also found that liver rhodanese activity was 15 times higher
    than brain rhodanese activity and that treatment with sodium nitrite
    and sodium thiosulfate protected liver, but not brain, cytochrome
    oxidase at high doses of KCN.

         Way et al. (1972) found that oxygen alone antagonized cyanide
    toxicity.  However, they found no significant benefit in increasing
    the partial pressure of oxygen from 1 to 4 ATA O2.  The authors
    demonstrated that oxygen potentiates the effect of thiosulfate and
    that the best results were obtained using hyperbaric oxygen together
    with nitrite and thiosulfate.

         Ivanov (1959) reported that hyperbaric oxygen re-established
    normal EEG activity in animals poisoned with cyanide.

         Takano et al. (1980) measured the reduction of pyridine
    nucleotide  in vivo as an indicator of cytochrome oxidase
    inhibition.  They found a protective effect of oxygen at 1 and 2 ATA
    O2 when the doses of KCN injected into rabbits were within the
    range that produced inhibition of cytochrome oxidase in animals
    breathing air.  At much higher doses, the protective effect of O2
    at 1 or 2 ATA became progressively weaker.

         Paulet (1955) described the protective effect of oxygen on the
    chemoreceptors and on the clinical manifestations in dogs poisoned
    by cyanide.

         Cope & Tenn (1961) showed that infusion of KCN solution (2 g/l)
    to anaesthetized dogs breathing air produced atrioventricular block. 
    Oxygen breathing reversed this phenomenon and re-established normal
    sinus rhythm.

    2.8.3  Toxicology

         Oxygen toxicity was discovered in 1878 by Paul Bert, who
    observed convulsions in animals treated with air at 15-20 ATA.  He
    then noted that when pure oxygen was used instead of air, one-fifth
    of the pressure sufficed to produce the convulsions.  He concluded
    that oxygen toxicity is directly correlated to the product of
    partial pressure of oxygen and administration period.

         Pulmonary oxygen toxicity was first described in 1897 by
    Lorrain Smith, who noted that mice exposed to oxygen at pressures
    above 1 ATA died from respiratory failure and that there was great
    variability both within and between animal species in the
    susceptibility to oxygen damage.

         The initial stage of acute pulmonary oxygen toxicity is termed
    the exudative phase and is characterized by a perivascular and
    interstitial inflammatory response that includes damage to the
    capillary endothelium with exudation of oedema
    fluid,polymorphonuclear leucocytes and, eventually, macrophages. 
    The oedema is associated with widening of the interalveolar septa
    and a thickened air-blood tissue barrier.  Alveolar haemorrhage
    develops and there is fibrin deposition in the alveoli (hyaline
    membranes).  Loss of alveolar lining cells (type I lining cells) can
    also occur during this phase.  As the condition progresses, it
    reaches a stage referred to as the proliferative phase,
    characterized by resolution of the inflammatory exudate and
    increased cellularity of the interstitium due to proliferation of
    macrophages and fibroblasts.  There is also proliferation of the
    cuboidal type II alveolar lining cells that serve to
    re-epithelialize the denuded alveolar basement membrane.  By the
    time a typical laboratory animal, such as the rat, is near death, a
    substantial number of pulmonary capillaries have been destroyed and
    embolization of some arterioles will have occurred.  Depending on
    the species, however, the destruction of type I alveolar lining
    cells and the proliferation of the type II cells may not be
    extensive.

         The bronchial tree is damaged by oxygen toxicity:  a
    necrotizing bronchiolitis has been described in neonatal mice by
    Ludwin et al. (1974).  The bronchiolitis consists of degeneration
    and necrosis of bronchiolar mucosal epithelium involving both
    ciliated as well as non-ciliated (Clara) cells.  Early change,
    occurring within 72 h in neonatal mice, consist of loss of cilia,
    cellular swelling associated with mitochondrial swelling, oedema of
    the bronchiolar wall, and leucocytic infiltration of the bronchiolar
    connective tissue.  These changes are followed by necrosis of the
    epithelial cells with subsequent sloughing.  Necrosis of the mucosal
    cells is maximal at 6 to 7 days.  Cold-blooded animals are
    relatively resistant to oxygen toxicity.

         Guinea-pigs, dogs and rats have a median time to death (LT50)
    in the range of 65-80 h while breathing 1 ATA oxygen.  Monkeys have
    an LT50 of about 100 h, while chicks survived for 28 days at
    1 ATA.  In general, there is a great interspecies variability in
    susceptibility to hyperoxia.

    2.8.3.1  Mechanism of injury

         Free radicals (superoxide anion O-2 and hydroxyl radical
    OH€) are produced by reduction of oxygen and these and hydrogen
    peroxide (H2O2) are responsible for most of the toxic effects of
    oxygen.  The free radicals react with cellular components and cause
    oxidative damage by oxygen radicals of the unsaturated fatty acids,
    which are abundant in the membranes of cells, and the sulfhydryl
    groups present in many enzymes and cellular proteins.  The cells are
    protected against oxygen radicals by enzymes which convert the
    radicals to less toxic molecules and to molecular oxygen.  Haber &
    Weiss (1934) described the production of hydroxyl free radicals from
    the superoxide anion and hydrogen peroxide.

         Hyperoxia also causes arteriolar vasoconstriction, this being
    unrelated to the free radical mechanism.

    2.9  Volunteer Studies of Pulmonary Oxygen Toxicity

         Clark & Lambertsen (1971) studied normal humans exposed to
    2 ATA O2.  Symptoms began within 3-8 h in the form of mild
    tracheal irritation and increased in intensity throughout the
    exposure time.  After 8-10 h, symptoms were characterized by
    uncontrollable coughing, dyspnoea at rest, and a constant
    tracheobronchial burning sensation.  Decreased vital capacity was
    the most constant and sensitive sign of pulmonary oxygen toxicity. 
    Decreased carbon monoxide diffusion capacity and increased
    alveolar-arterial O2 difference were also observed.

         Singer et al. (1970) compared two groups of patients treated
    with 0.4 ATA O2 or 1 ATA O2 for up to 24 h and found no
    immediate or delayed dysfunction.  He concluded that humans can
    tolerate 1 ATA O2 for up to 24 h without detectable pulmonary
    dysfunction.  Comroe et al. (1945) showed that inhalation of oxygen
    at 0.5-1.0 ATA O2 for more than 24 h decreased vital capacity. 
    Clark & Lambertsen (1971) established oxygen pressure versus time
    tolerance curves.  It appears that no dysfunction is observed if up
    to 0.5 ATA O2 is inhaled for many days and that the use of 1 ATA
    O2 is safe for up to 24 h.

    2.10  Clinical Studies of Oxygen Toxicity

         All human cells are affected by oxygen toxicity, but the three
    organs most susceptible to acute oxygen toxicity are the eyes in the
    newborn, and the lung and central nervous system at all ages.

    2.10.1  Eyes

         The eyes of the newborn and, especially, those of the premature
    newborn, are very sensitive to oxygen toxicity, which produces
    retrolental fibroplasia.

         Patz (1968a,b) described retrolental fibroplasia which occurred
    in premature babies exposed to 0.30-0.45 ATA O2 for 10 days. 
    Initially, in the presence of high concentrations of oxygen, there
    is vasoconstriction of arteries, arterioles and venules with
    necrosis of the immature retinal vessels (referred to as the
    vaso-obliterative phase).  Following oxygen-induced
    vaso-obliteration, there is a vaso-proliferation of new capillaries
    in the inner layer of the retina.  The proliferation of capillaries
    extends into the vitreous behind the lens and eventually results in
    retinal detachment, which represents the final phase.

    2.10.2  Central nervous system (CNS)

         Central nervous system oxygen toxicity is characterized by
    grand-mal seizures.  The manifestation of acute CNS toxicity in
    humans consists not only of grand-mal seizures but also of focal
    motor seizures, constriction of the visual fields, deafness,
    hyperacuity, changes of mood, and visual and auditory
    hallucinations.  CNS oxygen toxicity in humans is only observed
    during hyperbaric oxygen exposure.

         Grand-mal convulsions have been described in humans exposed to
    pressures above 2.8 ATA O2 for periods exceeding 2 h.  Central
    nervous system toxicity is rare if the O2 pressure is less than
    2.5 ATA and if oxygen and air are used intermittently, e.g., 10 min
    of air and 30 min of oxygen, as long as the total hyperbaric
    exposure time does not exceed 2 h.

    2.11  Clinical Studies - Case Reports

         In most cases reported in the literature the evidence for
    cyanide poisoning is lacking because blood concentrations of cyanide
    are not reported.  Moreover, the majority of cases of cyanide
    poisoning are treated with oxygen in combination with other
    antidotal therapy.  The individual role of oxygen as a cyanide
    antidote in man is therefore difficult to assess.  The following
    cases have been reported.

    2.11.1  Patients treated alone with supportive therapy and who
            survived

         *    An analytical chemist (48 years old) was found unconscious
              with unrecordable blood pressure and cardiac arrest.  The
              blood cyanide level was 5.8 mg/l (Edwards & Thomas, 1978).

         *    A three-year-old child was treated with laetrile enemas. 
              On admission the child was cyanotic, lethargic, and
              unresponsive.  The blood cyanide level was 2.14 mg/l
              (Ortega & Creek, 1978).

         *    A biochemist, aged 31 years, was found in deep coma with a
              metabolic acidosis and a blood cyanide concentration of
              2.3 mg/l (Vogel et al., 1981).

         *    Nine people were poisoned by hydrogen cyanide as a result
              of a leak from a valve.  Six of them lost consciousness
              briefly.  Blood cyanide concentrations were 2.8 to
              7.7 mg/l (Peden et al., 1986).

    2.11.2  Hyperbaric oxygen therapy in cyanide poisoning

         Trapp (1970) reported acute cyanide poisoning successfully
    treated with hyperbaric oxygen, while Litovitz et al. (1983)
    reported the unsuccessful use of hyperbaric oxygen.

         Trapp & Lepawsky (1983) reported five cases of acute cyanide
    poisoning successfully treated by hyperbaric oxygen in the Vancouver
    General Hospital, Canada.

    2.11.3  Cyanide poisoning due to smoke inhalation

         Four out of five smoke inhalation victims with mean blood
    cyanide levels of 1.62 mg/l (62 µmol/l) are reported to have
    survived after receiving hyperbaric oxygen therapy combined with the
    administration of sodium nitrite and sodium thiosulfate (Hart et
    al., 1985).

    2.12  Summary of Evaluation

         1.  In acute cases of poisoning by inhalation of hydrogen
    cyanide, either alone or in combination with carbon monoxide,
    termination of exposure and evacuation of the patient from
    contaminated areas is indicated most urgently and should be
    performed by the rescuers using appropriate protective equipment. 
    Measures to support respiration, circulation, and cardiac function
    should be commenced, if necessary.  Patients should receive pure
    humidified oxygen.

         Following smoke inhalation, patients frequently develop acute
    pulmonary failure due to the irritant effect of the smoke.  They may
    benefit from the use of continuous positive airway pressure while
    breathing spontaneously or positive end-expiratory pressure if
    ventilated artificially.

         Other antidotes for cyanide poisoning, especially sodium
    thiosulfate, should be administered intravenously as soon as
    possible.  Blood samples,  which must be taken before the
    administration of  antidotes, should be sent for toxicological
    analysis.

         2.  In acute cyanide poisoning by ingestion, decontamination of
    the stomach should be performed as soon as possible after or during
    first-aid therapy.

         3.  The final decision about the use of hyperbaric oxygen
    therapy depends on the availability of chambers and the ability to
    treat.

         4.  Oxygen has always been regarded as an important first-aid
    measure even though, it is difficult to understand, on a theoretical
    basis, how oxygen has a favourable effect in cyanide poisoning
    because inhibition of cytochrome oxidase is non-competitive.  There
    is now experimental evidence that oxygen has specific antidotal
    activity.  Oxygen accelerates the reactivation of cytochrome oxidase
    and protects against cytochrome oxidase inhibition by cyanide. 
    Nevertheless, there are other possible modes of action and those
    which are clinically important have yet to be determined.

         5.  Hyperbaric oxygen is recommended for smoke inhalation
    victims suffering from combined carbon monoxide and cyanide
    poisoning, since these two agents are synergistically toxic.  The
    role of hyperbaric oxygen in pure cyanide poisoning remains
    controversial.

    2.13  Model Information Sheet

    2.13.1  Uses

         Oxygen is indicated for the treatment of patients poisoned by
    cyanide, either alone or in combination with carbon monoxide
    following smoke inhalation.

    2.13.2  Dosage and route

         In cases of severe poisoning (respiratory insufficiency and/or
    deteriorating vital signs), artificial ventilation with 100% oxygen
    is indicated but for no longer than 12-24 h, after which the
    inspired oxygen concentration should be reduced.  Similarly in
    moderately severe cases, 100% oxygen is indicated but again for no
    longer than 12-24 h.  For mild cases 40% oxygen is indicated.

    2.13.3  Precautions/contraindications

         In order to prevent rebreathing of hydrogen cyanide from the
    expired air of a patient, it is advisable to include a non-return
    valve when a bag and mask is used.

    2.13.4  Adverse effects

         The alveolar membranes of the lung are damaged by 100% oxygen
    if this is administered for more than 24 h.

    2.13.5  Use in pregnancy and lactation

         No contraindication.

    2.13.6  Storage

         Store under pressure in a suitable metal container of a type
    permitted by the safety regulations of the national authority. 
    Valves and taps should not be lubricated with oil or grease.

         The containers for medical oxygen are coded with a white colour
    at the top according to ISO-32-19 (Fig. 2) and carry the indication
    "Medical Oxygen" in the iron material of the container.  To prevent
    connection with other medical gases the containers are provided with
    an international standardized "Pin Index Safety System" ISO-407. 
    This system is the same for medical and for industrial oxygen so
    that in emergencies industrial oxygen can also be used.

         Regular control of gas pressure is necessary to prevent
    shortage of oxygen in emergency situations.

         Although the duration of storage will not alter the quality it
    is recommended that the gas should not be stored for more than six
    months.

    2.14  References

    Clark JM & Lambertsen CJ (1971) Rate of development of pulmonary
    O2 toxicity in man during O2 at 2.0 ATM. ABS. J Appl Physiol,
    30: 730-752.

    Clark CJ, Campbell D, & Reid WH (1981) Blood carboxyhaemoglobin and
    cyanide levels in fire survivors. Lancet, 1: 1332-1335.

    Comroe JH, Dripps RD, Dumke PR, & Deming M (1945) Oxygen toxicity:
    The effect of inhalation of high concentrations of oxygen for
    twenty-four hours on normal men at sea level and at a simulated
    altitude of 18,000 feet. J Am Med Assoc, 128: 710-717.

    Cope C & Tenn M (1961) The importance of oxygen in the treatment of
    cyanide poisoning. J Am Med Assoc, 175: 109-112.

    Edwards AC & Thomas ID (1978) Cyanide poisoning. Lancet, 1: 92-93.

    Haber F & Weiss JJ (1934) The catalytic decomposition of hydrogen
    peroxide by iron salts. Proc R Soc Lond, A147: 332-351.

    Hart GB, Strauss MB, Lennon PA, & Whitcroft DD (1985) Treatment of
    smoke inhalation by hyperbaric oxygen. J Emergency Med, 3: 211-215.

    Isom GE & Way JL (1982) Effects of oxygen on the antagonism of
    cyanide intoxication - cytochrome oxidase,  in vitro. Toxicol Appl
    Pharmacol, 74: 57-62.

    Ivanov JP (1959) The effect of elevated oxygen pressure on animals
    poisoned with potassium cyanide. Pharmacol Toxicol (USSR), 22:
    476-479.

    Litovitz TL, Larkin RF, & Myers RAM (1983) Cyanide poisoning treated
    with hyperbaric oxygen. Am J Emergency Med, 1: 94-101.

    Ludwin SK, Northway WH Jr, & Bensch KG (1974) Oxygen toxicity in the
    newborn. Necrotizing bronchiolitis in mice exposed to 100% oxygen.
    Lab Invest, 31: 421-435.

    Ortega MA & Creek JE (1978) Acute cyanide poisoning following
    administration of laetrile enemas. J Pediatr, 93: 1059.

    Patz A (1968a) New role of ophthalmologist in prevention of
    retrolental fibroplasia. Arch Ophthalmol (Chicago), 78: 565-568.

    Patz A (1968b) Comment on oxygen therapy. Arch Ophthalmol (Chicago),
    79: 507.

    Paulet G (1955) Valeur et méchanisme d'action d'intoxication
    cyanhydrique. Arch Int Physiol, 63: 340.

    Peden NR, Taha A, McSorley PD, Bryden GT, Murdoch IB, & Anderson JM
    (1986) Industrial exposure to hydrogen cyanide: Implications for
    treatment. Br Med J, 293: 538.

    Singer MM, Wright F, Stanley LK, Roc BB, & Hamilton WK (1970) Oxygen
    toxicity in man. A prospective study in patients after open heart
    surgery. New Engl J Med, 283: 1473.

    Takano T, Miyzaki Y, Nashimoto I, & Kobayashi K (1980) Effect of
    hyperbaric oxygen on cyanide intoxication: in situ changes in
    intracellular oxidation reduction. Undersea Biomed Res, 7: 191-197.

    Trapp WG (1970) Massive cyanide poisoning with recovery:  A
    Boxing-Day story. Can Med Assoc J, 102: 517.

    Trapp WG & Lepawsky M (1983) 100% survival in five life-threatening
    acute cyanide poisoning victims treated by a therapeutic spectrum
    including hyperbaric oxygen. Proceedings of the 1st European
    Conference on Hyperbaric Medicine Amsterdam, The Netherlands p 45.

    Vogel SN, Sultan TR, & Ten Eyck R (1981) Cyanide poisoning. Clin
    Toxicol, 18: 367-383.

    Way JL, End E, Sheehy MH, De Miranda P, & Feitknecht UF (1972)
    Effect of oxygen on cyanide intoxication. IV. Hyperbaric oxygen.
    Toxicol Appl Pharmacol, 22: 415-421.

    3.  SODIUM THIOSULFATE

    3.1  Introduction

    3.1.1  Indications

         The use of sodium thiosulfate as an antidote has been recorded
    in the literature on poisoning due to cyanide, mustard gas, nitrogen
    mustard, bromate, chlorate, bromine, iodine, cisplatin, and certain
    drugs (Dactinomycin, Mechlorethamine, Mitomycin) when extravasated. 
    There are also some references to its effect in iodate and
    hypochlorite toxicity. The principal role of sodium thiosulfate,
    though, lies in the treatment of cyanide poisoning.

    3.1.2  Rationale for the choice of the antidote

         The effect of sodium thiosulfate as an antidote in cyanide
    poisoning is well documented, and was first demonstrated by Lang
    (1895).  Some authors believe it to be a relatively slow-acting
    antidote, though others have demonstrated that it acts more rapidly
    than was previously thought, enabling the conversion of cyanide to
    thiocyanate (Krapez et al., 1981).  Thiosulfate helps to detoxify
    cyanide in the presence of the enzyme rhodanese.  However, rhodanese
    is an intramitochondrial enzyme and thiosulfate has limited ability
    to penetrate cell and mitochondrial membranes.  Thus, distribution
    of thiosulfate is almost exclusively extracellular (Cardozo &
    Edelman, 1952), while its antidotal action has been thought to take
    place intracellularly.  This view is now being re-examined in the
    light of recent experimental evidence (see section 3.8).

    3.1.3  Risk groups

         No special-risk groups can be identified as regards the use of
    sodium thiosulfate.  However, it should be noted that there is an
    apparently reduced ability to convert cyanide to thiocyanate in some
    diseases, e.g., toxic amblyopias (in particular tobacco amblyopia)
    and Leber's hereditary optic atrophy (Wilson, 1965; Darby & Wilson,
    1967).  Abnormally low rhodanese activity in the liver has been
    described in two patients with Leber's hereditary optic atrophy
    (Grant, 1986).

    3.2  Name and Chemical Formula of Antidote

         International non-proprietary name: Natrii thiosulfas; Sodium
    thiosulfate (thiosulphate); Thiosulfate de sodium; Natrium
    thiosulfuricum; Natriumthiosulfat (Hager, 1977).

         CAS number: 10102-17-7 for sodium thiosulfate, pentahydrate
    (NIOSH, 1986); 7772-98-7 for sodium thiosulfate, anhydrous, (NIOSH,
    1986).

         IUPAC name:  Sodium thiosulfate, pentahydrate

         Manufacturer:  Readily available in many countries.

         Commercial names:  Commercially available as sodium thiosulfate
         or equivalent in many countries.

         Formula:  Na2S2O3.5H2O (Martindale, 1989)

         Relative molecular mass:  248.2 (Martindale, 1989)

         Specification of chemical salt used:  sodium thiosulfate
    contains not less that 99.0% and not more than the equivalent of
    101.0% of Na2S2O3.5H2O (European Pharmacopoeia, 1980);
    transparent, colourless crystals (European Pharmacopoeia, 1980);
    colourless, odourless, (or almost odourless) monoclinic prismatic
    crystals, or a coarse crystalline powder with a saline taste
    (Martindale, 1989).

    3.3  Physico-chemical Properties

    3.3.1  Melting point, boiling point

         Sodium thiosulfate dissolves in its own water of
    crystallisation at about 49 °C (European Pharmacopoeia, 1980;
    Martindale, 1989;).  It loses all its water at 100 °C and decomposes
    at higher temperatures (Windholz, 1983).  Above 200-300 °C, it
    decomposes to sulfate and pentasulfide (Kirk-Othmer, 1969; Hager,
    1977).  When it is heated to the point of decomposition, fumes of
    sulfur oxides are emitted (Sax, 1984; PoisIndex, 1987).

    3.3.2  Solubility in vehicle for administration

         Highly soluble in water (2 parts sodium thiosulfate in 1 part
    water) (Martindale, 1982; Windholz, 1983).

    3.3.3  Optical properties

         Inactive with respect to polarized light; absorbs ultraviolet
    light at wavelengths less than 230 nm (information from the National
    Corporation of Swedish Pharmacies).

    3.3.4  Acidity

         A 10% solution in water has a pH of 6.0-8.4 ( European
    Pharmacopoeia, 1980; Martindale, 1989).  The pH of injectable
    thiosulfate (0.15 g/ml) is 8.2-8.8 (information from the National
    Corporation of Swedish Pharmacies).

    3.3.5  pKa

         pKa1: 1.46-1.74 (an approximate value) (IUPAC, 1969).

    3.3.6  Stability

         Efflorescent in warm (>30 °C) dry air.  Slightly hygroscopic
    (delinquesent) in moist air (Windholz, 1983).  Store in an airtight
    container (Martindale, 1989).

         Aqueous solutions have limited stability due to a tendency to
    decompose slowly by the following reactions:

         Na2S2O3  -->  Na2SO3 + S (neutral or acidic
    solutions)

         Na2S2O3 + H2O  -->  Na2SO4+H2S (alkaline
    solutions)

         The first reaction is accelerated by acids and the second by
    air or oxygen.  Aqueous solutions of sodium thiosulfate decompose
    more rapidly on heating.  Storage with limited access to air and
    light in a cool environment increases stability (Kirk-Othmer, 1969;
    Martindale, 1989; Windholz, 1983).

         Injectable thiosulfate stored in ampoules for three years shows
    no significant change in composition.

    3.3.7  Refractive index, specific gravity

         No value for refractive index (injectable thiosulfate
    [0.15  g/ml]) is available.  The specific gravity for injectable
    thiosulfate (0.15 g/ml) is 1.076 (information from the National
    Corporation of Swedish Pharmacies).

    3.3.8  Loss of weight on drying

         The substance loses 35.5-37.0% of weight when it is dried at
    105 °C (Pharmacopoea Nordica, 1964; Hager, 1977).

    3.3.9  Excipients

         For injectable thiosulfate (0.15 g/ml): sodium phosphate
    dodecahydrate (Na2HPO4.12H2O) 1.21% (information from The
    National Corporation of Swedish Pharmacies; see also section 3.6).

    3.3.10  Incompatibility

         Incompatibility with iodine, acids, lead, mercury and silver
    salts (Windholz, 1983) and with salts of heavy metals, oxidizing
    agents, and acids has been indicated.  If sodium thiosulfate is

    triturated with chlorates, nitrates, or permanganates, an explosion
    may occur (Martindale, 1989).

    3.3.11  Other information

         Solutions are sterilized by autoclaving.  A 2.98% solution is
    iso-osmotic with serum (Martindale, 1989).

        Table 2.  Physical properties of sodium thiosulfate pentahydrate

                                                                       

         Property                                              Value
                                                                       

         refractive index, nd20                                 1.4886

         density, d425                                          1.750

         heat of solution in water at 25 °C (kcal/mol)       - 11.3
         heat of formation (kcal/mol)                        -621.9
         heat of fusion (kcal/mol)                             11.85
         specific heat at 21 °C (cal/g)                         0.346
         cryoscopic constant                                   42.6
         dissociation pressure (mm Hg)
            at 20 °C                                            6.0
            at 35 °C                                           18.0
         vapour pressure of saturated solution (mmHg)
            at 33 °C                                           10.0
            at 57 °C                                           42.0
            at 90 °C                                          233.0
                                       20
         density of aqueous solution, d20, (Na2S2O3, by weight)
            10%                                                 1.0847
            20%                                                 1.1760
            30%                                                 1.2762
            40%                                                 1.3851
                                                                       

         From:  Kirk-Othmer (1969)
    
    3.4  Synthesis

         Sodium thiosulfate can be produced commercially by a number of
    methods, of which the four principal processes may be classified as
    follows (Kirk-Othmer, 1969).

         (a)  Air oxidation of sulfides, hydrosulfides, and polysulfides
              of alkali and alkaline earth metals.

         (b)  Recovery from waste liquors resulting from the production
              of sulfur black and other sulfur dyes.

         (c)  Reaction of sulfur dioxide and sulfides.

         (d)  Reaction of sulfur and sulfites.

         Possible contaminants include chlorides, sulfates, sulfites,
    and heavy metals.

    3.5  Analytical Methods

    3.5.1  Quality control procedures for sodium thiosulfate
           (European Pharmacopoeia, 1980)

         Tests for chlorides, sulfates, sulfites, sulfides, and heavy
    metals.  The pH of a 10% solution is 6.0-8.4; clear and colourless
    (European Pharmacopoeia, 1980).

    3.5.2  Methods for identifying sodium thiosulfate
           (European Pharmacopoeia, 1980).

         (a)  Decolorization of iodinated potassium iodide solution R.

         (b)  In combination with silver nitrate solution R2, a white
              precipitate forms, which rapidly turns yellowish and then
              black.

         (c)  In combination with hydrochloric acid, a precipitate of
              sulfur is formed and a gas is evolved which gives a blue
              colour to starch iodate paper R.

         (d)  Reaction on sodium.

    3.5.3  Assay

         Dissolve 0.500g in 20ml of water and titrate with 0.1 N iodine,
    using 1 ml of starch solution R, added towards the end of the
    titration, as indicator.

         1 ml of 0.1 N iodine is equivalent to 24.82 mg of Na2S2O3.5H2O
    (European Pharmacopoeia, 1980).

    3.5.4  Methods for analysis of sodium thiosulfate in
    biological
           samples

         Gast et al. (1952) and Dixon (1962) described a method for the
    measurement of thiosulfate, based on reduction by iodine, but this
    is a nonspecific means of estimation.

         Sörbo & Ohman (1978) described a method for the determination
    of thiosulfate in urine.  After the removal of interfering
    compounds, including endogenous thiocyanate by ion exchange,
    thiosulfate is converted to thiocyanate in the presence of cyanide
    and cupric ions.  The thiocyanate formed is concentrated by ion
    exchange, eluted with an acid solution of ferric ions, and the
    ferric thiocyanate complex is determined colorimetrically.

         A method for determining biological thiols at the picomole
    level, based upon the conversion of thiols to fluorescent
    derivatives by reaction with monobromobimane and the separation of
    the derivatives by reverse-phase high-performance liquid
    chromatography, was described by Newton et al. (1981).  A
    modification of this method was used by Shea et al. (1984) for the
    determination of thiosulfate levels in plasma and urine.

         The methylene blue method (Ivankovich et al., 1983) is
    considered specific for thiosulfate and sensitive to 1 µg/ml.  The
    analyses are undertaken directly on heparinized plasma or urine. 
    Potassium iodide, potassium bromide, and monobasic potassium
    phosphate are added to the sample.  Then potassium borohydride (in
    sodium hydroxide) and acetone are added with stirring, followed by
    ferric sulfate (with water) and  N,N-dimethyl- p-phenylenediamine
    sulfate in sulfuric acid.  A blue colour develops and absorbance is
    measured at 665 nm.

    3.6  Shelf-life

         According to studies at the National Corporation of Swedish
    Pharmacies, which has followed the stability of injectable
    thiosulfate (0.15 g/ml) (Sweden) for three years, no significant
    changes occurred during the period of observation.

         A solution without, or with only small amounts of, sodium
    phosphate dodecahydrate is unstable.  The National Corporation of
    Swedish Pharmacies has found the composition described in
    section 3.3.9 to be the most suitable.  For information concerning
    storage, see section 3.13.6.

    3.7  General Properties

    3.7.1  Mechanism of antidotal activity

         The major route of detoxification of cyanide in the body is
    conversion to thiocyanate.  This reaction requires a source of
    sulfane sulfur (divalent sulfur bonded to another sulfur) and is
    catalysed by sulfur transferases.  It has been suggested that there
    is a physiological pool of cyanide-reactive sulfane sulfur bound to
    albumin that might act as a buffer against endogenous cyanide
    production (Westley et al., 1983; Way et al., 1984).  Thiosulfate is
    present in the body in only small quantities, derived mainly from
    cystine and other mercapto compounds.  The physiological reserves
    available for detoxifying cyanide are therefore limited (Schulz et
    al., 1979b).

                                Rhodanese
                   Na2S2O3 + CN-   -->   SCN- + Na2SO3.

         The sulfurtransferases catalyse reactions where sulfane sulfur
    is involved.  Rhodanese is the sulfurtransferase which has been
    studied most extensively.  Rhodanese (thiosulfate: cyanide
    sulfurtransferase; EC 2.8.1.1.) catalyses the transfer of sulfane
    sulfur directly to cyanide.  It is distributed throughout the body,
    the highest concentrations being found in the liver, and is located
    mainly in the matrix of mitochondria (Westley et al., 1983).

         The existence of a thiocyanate oxidase that could oxidize
    thiocyanate back to cyanide (Goldstein & Rieders, 1953) has been
    questioned.  However, this is now attributed to artifactual
    formation of HCN during assay (Vesey, 1979).

         Sodium thiosulfate contains the necessary divalent sulfur donor
    bound to another sulfur and it is the main sulfur donor for
    rhodanese in the conversion of cyanide to thiocyanate.  Whereas
    rhodanese is available in excess in the body, the lack of a suitable
    sulfur donor is the rate-limiting factor for this route of
    detoxification in cyanide poisoning.  This is the rationale for the
    administration of sodium thiosulfate in cyanide poisoning so that
    the endogenous detoxification capacity of the body is enhanced.

    3.7.2  Other biochemical/pharmacological profiles

         As described above, sodium thiosulfate may be used for
    indications other than cyanide poisoning, utilizing other properties
    of this substance.  These are not dealt with in this document.

    3.8  Animal Studies

    3.8.1  Pharmacokinetics

         When high doses of thiosulfate are injected into mammals, the
    greater part is eliminated unchanged by renal excretion but a
    certain amount is oxidized to sulfate.  This latter fraction
    increases as the dose of thiosulfate decreases.  The oxidation of
    thiosulfate to sulfate occurs in the liver by a two-step enzymatic
    pathway.  Studies by Gilman et al. (1946) demonstrated that
    intravenously injected thiosulfate is rapidly distributed in the
    extracellular fluid space and that its renal excretion occurs by
    glomerular filtration.  Further animal experiments have shown that
    tubular transport may also take place (Sörbo, 1972).

         Thiosulfate is both secreted and reabsorbed in man and dog,
    according to Bucht (1949) and Foulks et al. (1952).  Clearance of
    thiosulfate is low, but at high levels clearance equals the
    glomerular filtration rate.  This means that at high plasma levels
    of thiosulfate, secretion Tm (transfer maximum) is similar to
    reabsorption Tm, whereas at low plasma levels both filtered and
    secreted thiosulfate are reabsorbed and thus there is a diminished
    clearance value for thiosulfate.

         The volume of distribution, as determined in seven dogs 
    weighing 8.5-14.4 kg was, on average, 3 1 (Cardozo & Edelman, 1952). 
    Kinetic studies have shown that there is a cationic site on
    rhodanese for the anionic sulfur donor (Westley et al., 1983).  Most
    of an injected dose of thiosulfate is excreted unchanged. 
    Thiosulfate is thought to permeate slowly through cell membranes
    (Himwich & Saunders, 1948; Sörbo, 1962).

         According to Crompton et al. (1974), thiosulfate can utilize
    the dicarboxylate carrier to enter the mitochondria, as shown in
    experiments with rat liver mitochondria.  This system is specific
    for divalent anions.

         It has been shown by Szczepkowski et al. (1961) that when using
    labelled thiosulfate the two atoms of sulfur have different fates
    during the course of metabolism in animals.  In rats, the inner
    sulfur atom is eliminated very quickly in the form of sulfate while
    the outer atom is transformed into sulfate much more slowly,
    probably going through a number of intermediate stages.

         When an experimental animal is injected with thiosulfate
    containing 35S in its sulfane position exclusively, the whole of
    this can be found labelled in the plasma as quickly as a sample can
    be obtained (Schneider & Westley, 1969).

         Experiments on dogs (Michenfelder & Tinker, 1977; Schulz et
    al., 1979b) have shown that the capacity of the endogenous reserves
    of thiosulfate to detoxify cyanide is exceeded if sodium
    nitroprusside is administered as a continuous infusion at a rate of
    more than 0.5 mg/kg per h.  When the experimental animals received
    doses greater than 0.5 mg/kg per h, their blood cyanide
    concentrations rose continuously.  Experimental animals receiving
    the same dosage under otherwise identical conditions, but with the
    additional infusion of thiosulfate at six times (w/w) the sodium
    nitroprusside dosage, showed no abnormal signs.  The urinary volume
    in the thiosulfate-treated dogs, over a 48 h period, was
    approximately twice that of the untreated animals, presumably due to
    the increased rate of formation of thiocyanate and the resultant
    osmotic diuresis.

         Similar results were obtained in experiments on rabbits (Höbel
    et al., 1978).

    3.8.2  Pharmacodynamics

         After the induction of acute sodium nitroprusside (SNP)
    intoxication in rabbits, bolus injections of thiosulfate and
    hydroxocobalamin (B12a), at SNP/antidote molar ratios of 1:5, were
    equally effective in reducing the early signs and severity of the
    metabolic acidosis (Pill et al., 1980).  During the subsequent
    observation period, the base excess with B12a as an antidote was
    found to be lower than with thiosulfate.  When the two antidotes
    were given in parallel with a highly toxic dose of SNP, sodium
    thiosulfate proved to be superior to B12a.  The authors suggested
    that for clinical purposes, SNP should always be administered in
    combination with thiosulfate (1:5).

         One molecule of sodium nitroprusside contains five cyanide
    ions.  Thiosulfate should therefore be given in a molar ratio of at
    least 5:1, which corresponds to a dose of four parts by weight of
    sodium thiosulfate to one of SNP.  Schulz et al. (1979b) suggested
    that since thiosulfate is rapidly metabolized and eliminated from
    the body, it is better to administer it in  excess by continuous
    infusion.

         In studies by Ivankovich et al. (1980), dogs were given KCN in
    a constant infusion (0.1 mg/kg per min).  One group of animals
    (n = 5) was given an infusion of sodium thiosulfate (12 mg/kg per h)
    intravenously 10 min prior to and during KCN infusion;  another
    group (n = 5) was given an intravenous bolus injection of
    thiosulfate (150 mg/kg).  The infusion increased the amount of
    cyanide needed to cause death and the bolus injection increased the
    protection from lethality even further.  It was shown that
    thiosulfate alone is capable of providing complete protection
    against both cyanide and cyanide-forming compounds when administered
    simultaneously with these compounds as a continuous infusion.

    Furthermore, thiosulfate treatment resulted in no noticeable adverse
    haemodynamic or respiratory effects when given as either a bolus or
    a constant infusion.  When high plasma concentrations of thiosulfate
    are available, the detoxification mechanism is rapid enough to
    provide adequate protection.  Since thiosulfate is rapidly
    eliminated by the kidneys, this high plasma level of thiosulfate is
    best maintained by constant infusion.  The only deleterious effect
    of such a constant infusion may be a lowering of plasma volume,
    since thiosulfate acts as an osmotic diuretic at this dosage; 
    however, this is rarely important clinically.  The authors stated
    that true detoxification of cyanide was achieved with thiosulfate
    alone and that thiosulfate appears to be the agent of choice,
    resulting in the lowest cyanide concentrations in tissues and blood
    and the fewest side effects.

         Dogs given thiosulfate (75 mg/kg) intravenously 5 min before
    the start of an infusion of SNP (1.5 mg/kg) had significantly lower
    plasma and red cell cyanide concentrations, and significantly higher
    plasma thiocyanate concentrations, than controls (Krapez et al.,
    1981).  These changes were associated with only minimal disturbance
    of tissue oxygenation.  The authors suggested that a bolus dose of
    sodium thiosulfate (75 mg/kg) is an effective antidote with
    negligible toxicity.  This study demonstrates that thiosulfate acts
    more rapidly than had been thought previously.  The maintenance of
    low blood cyanide concentrations, coupled with the rapid increase in
    plasma thiocyanate concentrations and unimpaired tissue oxygenation
    in those animals that received thiosulfate, strongly suggests that
    cyanide was converted to thiocyanate as quickly as it was released
    from the nitroprusside.  In this study, no synergism was found
    between sodium thiosulfate and hydroxocobalamin.  Investigations by
    Evans (1964) suggested a negative interaction between thiosulfate
    and hydroxocobalamin, but others (Hall & Rumack, 1987) believe there
    is antidotal synergy.

         In a study by Vesey et al. (1985), a bolus dose of sodium
    thiosulfate (150 mg/kg, i.e., 12 times the stoichiometric amount
    theoretically required to "neutralize" the cyanide from the SNP
    dose) was given to dogs at the end of a 60-min infusion of SNP
    (3 mg/kg, i.e., near lethal dose).  Compared with the controls,
    there was an impressive reduction in the mean half-lives of plasma
    cyanide (25.1 min as opposed to 74.1min) and red blood cell cyanide
    concentrations (22.4 min as opposed to 203.6 min).  Cyanide toxicity
    may be delayed after SNP administration due to continued breakdown
    and release of HCN.  The authors suggested that red blood cells act
    as a site for cyanide detoxication.  Thiosulfate enhances the rate
    of HCN metabolism and also limits its peripheral distribution in
    dogs (Sylvester et al., 1981).

         Chen et al. (1934) showed that sodium thiosulfate detoxified
    sodium cyanide at up to 3 times the minimal lethal dose (MLD). 
    Sodium nitrite did so at up to 4 times the MLD and a combination of
    the two at up to 20 times the MLD.

         Differing doses of thiosulfate were given intraperitoneally to
    mice at different times after the injection of sublethal or lethal
    doses of cyanide (Schubert & Brill, 1968).  When thiosulfate was
    administered to mice 5 min after cyanide, the time for recovery from
    cyanide toxicity was shortened considerably.  Rats given thiosulfate
    10 min after cyanide (when inhibition of liver cytochrome oxidase
    was maximal) invariably recovered 5 to 10 min later instead of the
    30 to 40 min normally required without treatment.

         A pharmacokinetic analysis of cyanide distribution and
    metabolism with and without intravenous sodium thiosulfate was
    conducted in pretreated mongrel dogs (Sylvester et al., 1983).  The
    mechanism of thiosulfate protection appeared to be extremely rapid
    formation of thiocyanate in the central compartment, which therefore
    limited the amount of cyanide distributed to sites of toxicity. 
    Thiosulfate increased the rate of conversion of cyanide to
    thiocyanate over 30-fold.

    3.8.3  Toxicology

         According to NIOSH (1986), the intravenous LD50 in the mouse
    is 2350 mg/kg, whereas the intravenous lowest published lethal dose
    (LDLo) in the dog is 3000 mg/kg (Dennis & Fletcher, 1966).  When
    dogs were given 3000 mg/kg of sodium thiosulfate pentahydrate
    intravenously (Dennis & Fletcher, 1966), the following effects
    developed rapidly:  metabolic acidosis, hypoxaemia, hypernatraemia,
    and changes in the ECG and in arterial and venous pressures.  In
    these experiments, an immediate and rapid rise in the serum sodium
    concentration would be expected, since the sodium content in sodium
    thiosulfate pentahydrate is about 24 mEq/3000 mg.  Moreover, the
    dogs that survived the injection showed a marked diuresis, which
    would be expected from the large osmotic dose administered.  The
    authors suggested that sodium thiosulfate pentahydrate (1500 mg/kg)
    given intravenously at a constant rate over a 30-min period is
    tolerated well.

         During chronic sodium nitroprusside (SNP) administration, the
    simultaneous infusion of thiosulfate may present problems because of
    enhanced plasma thiocyanate accumulation and the danger of
    hypovolaemia (Michenfelder & Tinker, 1977).  Vesey et al. (1985)
    suggest that it would be sufficient to give a bolus dose of sodium
    thiosulfate only if the SNP dose/dose-rate is excessive.

         There appears to be no information concerning the
    teratogenicity or mutagenicity of sodium thiosulfate.

    3.9  Volunteer Studies

         Ivankovich et al. (1983) studied the available thiosulfate pool
    and the pharmacokinetics of administered thiosulfate in healthy
    volunteer subjects.  Plasma thiosulfate concentrations, sampled from
    the volunteers were 11.3 (± 1.1) mg/l (n = 26) and urine thiosulfate
    concentrations were 2.8 (± 0.2) mg/l (n = 24).  Bile contained 137.2
    (± 29.5) mg/l thiosulfate (n = 6, sampled during cholecystectomy). 
    Thiosulfate (150 mg/kg) was injected intravenously into 5 normal
    male volunteers.  Plasma thiosulfate concentrations after 5 min were
    1012 (± 88.5) mg/l.  The half-life of the distribution phase was
    23 min and that of the elimination phase 182 min.  The calculated
    VD was 151 ml/kg body weight.  Urine concentration, clearance, and
    rate of thiosulfate excretion increased markedly after injection. 
    Total excretion was 42.6 (± 3.5)% of the injected dose at 180 min,
    although urinary excretion did not increase much during the
    elimination phase;  at 18 h after injection it was 47.4 (± 2.4)%. 
    Bile thiosulfate excretion did not change after thiosulfate
    injection and bile excretion of thiosulfate accounted for less than
    0.1% of total thiosulfate excretion.  This study demonstrated that
    the plasma concentration of thiosulfate in normal males is about
    10 mg/l, and that excretion amounts to approximately 3 mg/day,
    compared with findings of Sörbo & Ohman (1978) who discovered a
    renal excretion value of 31.7 (± 12.8) mmol/24 h (7.9 (± 3.2) mg/l
    per 24h).  The normal endogenous production of thiosulfate can
    therefore be considered to be small and the ability to produce
    increased amounts in response to poisoning is likely to be limited. 
    The VD is 150 ml/kg and a 70 kg man would therefore have a total
    extracellular thiosulfate content of 125 mg.  Similar human VD
    values were found by Cardozo & Edelman (1952).  "Therapeutic" doses
    of thiosulfate (150 mg/kg according to these authors) would elevate
    plasma concentrations about 100 times.  Such high concentrations may
    be necessary to increase the intracellular concentration and enable
    rhodanese to detoxify cyanide at the mitochondrial membrane if
    indeed this is the site of action of thiosulfate (see above).

         Schulz (1984) and Schulz et al. (1982) found a serum half-life
    for thiosulfate of about 15 min during SNP therapy.  According to
    Gladtke (1966), the elimination half-life of sodium thiosulfate is
    about 40 min (children aged 4 months to 14 years).

         Absorption of sodium thiosulfate after oral administration is
    poor (Martindale 1989).  The oral toxicity is low and single doses
    of 5-18 g have only a laxative action.  Nausea and vomiting have
    also been reported (Sörbo, 1972; Poisindex, 1987).  It has also been
    shown that thiosulfate injected intravenously is rapidly distributed
    in the extracellular fluid space.  Its renal excretion occurs by
    glomerular filtration and secretion (Bucht, 1949; Foulks et al.,
    1952; Sörbo, 1972).

    3.10  Clinical Studies

         The use of sodium thiosulfate as a single antidote in cyanide
    poisoning has been evaluated in only a few studies.

         Schulz et al. (1982) studied cyanide toxicity resulting from
    sodium nitroprusside (SNP) in therapeutic use with and without
    sodium thiosulfate.  Cyanide was analysed using the method of Asmus
    and Garschagen (see chapter 10), and concentrations were expressed
    as nmol cyanide/ml erythrocyte.  Thiocyanate concentrations were
    also measured; following the addition of ferric chloride,
    thiocyanate reacted with elemental chlorine to form cyanogen
    chloride.  The remainder of the measurement was carried out as for
    cyanide.  Concentrations were expressed as nmol thiocyanate/ml
    plasma throughout.  Thiosulfate was measured in plasma using the
    method of Gast et al. (1952) and Dixon (1962).  Seventy patients
    (aged 17-78 years) were studied.

         In 51 patients, SNP was given for short periods, while in 19
    patients, SNP was given over longer periods of up to 2 weeks,
    usually in combination with sodium thiosulfate infusion.  In seven
    of these 19 patients, 1 g sodium thiosulfate was given intravenously
    as a bolus injection during SNP treatment in order to study the
    kinetics.  The drugs were infused quickly in a ratio of 50 mg SNP to
    500 mg sodium thiosulfate.  The threshold dose for the release of
    free cyanide into the bloodstream in this study was 2 µg SNP/kg per
    min.  It was calculated that 5 µg/kg per min for 10 h, 10 µg/kg per
    min for 4 hr or 20 µg/kg per min for 1.5 h would cause potentially
    toxic levels of cyanide.  Sodium thiosulfate by infusion stopped
    accumulation of cyanide and the elevated cyanide levels declined,
    whereas thiocyanate levels increased.  The simultaneous infusion of
    thiosulfate and SNP prevented accumulation of cyanide.  According to
    Saunders & Himwich (1950), the optimum  in vitro molar
    cyanide/thiosulfate ratio for the rhodanese reaction is 1:3.  As
    none of the patients in this study showed clinical signs of
    toxicity, an assessment of treatment efficacy was made on the basis
    of analytical data.

         Shea et al. (1984) found that sodium thiosulfate (12 g/m2)
    could be given safely to humans over 6 h provided that cardiac and
    renal functions were normal.  Similarly, 2 g/m2 per h for 12 h has
    been given without side-effects (Howell et al., 1982).  The
    elimination half-life determined by Shea et al. (1984) using a
    one-compartment kinetic model was approximately 80 min.

         Schulz et al. (1979a) studied the kinetics of thiocyanate in
    healthy subjects and in patients with renal failure.  In healthy
    subjects, the elimination half-life of thiocyanate (oral
    administration) was 1 to 5 days (mean 3 days).  The average
    half-life for patients with renal failure was 9 days, with
    elimination constants increasing in proportion to the creatinine

    clearance.  These findings have special relevance to the use of
    sodium nitroprusside in the treatment of patients with renal
    insufficiency.

    3.11  Clinical Studies - Case Reports

         Numerous cases have been reported where sodium thiosulfate has
    been used in conjunction with other antidotes in the treatment of
    cyanide poisoning.  The use of sodium thiosulfate alone, though, has
    been reported only rarely.

         A baby, weighing 4.4 kg, developed hypertension in the neonatal
    period and was treated with SNP by infusion (2-5 µg/kg per min). 
    After 30 h the blood pressure fell and the child became acidotic. 
    The erythrocyte cyanide concentration was 400 nmol/ml
    (life-threatening intoxication).  Infusion of sodium thiosulfate in
    a dose of 100 mg/kg promptly caused the cyanide level to fall to
    one-tenth of its maximum value.  It was concluded that the mixed
    infusion of SNP and thiosulfate is always advisable, especially in
    small children where the thiosulfate pool is small (Schulz & Roth,
    1982).

         A man aged 42 years was treated with SNP and sodium thiosulfate
    in combination and observations were made to calculate appropriate
    doses.  The patient developed toxic cyanide levels (erythrocyte
    cyanide concentration, 3.6 µg/ml) during long-term infusion of SNP,
    and was treated successfully with sodium thiosulfate (the
    erythrocyte cyanide concentration level dropped to 0.5 µg/ml within
    7 h).  It was suggested that thiosulfate should be given at a dose
    at least four times that of SNP but, as some excess of thiosulfate
    is desirable, it was suggested that 300-400 mg sodium thiosulfate be
    given together with 60 mg SNP by continuous infusion (Schulz et al.,
    1979b).

         A boy aged 14 years was given SNP during surgery until the
    desired clinical effect was achieved.  After 5 h the patient
    developed circulatory shock but responded promptly to the
    administration of sodium thiosulfate (150 mg/kg) given over 15 min
    (Perschau et al., 1977).

         A man aged 30 years ingested an unknown amount of a
    cyanide-containing insecticide in a glass of beer.  He was found
    unconscious, apnoeic, and cyanotic 30 min later.  Resuscitative
    measures were commenced and he was given sodium thiosulfate
    intravenously.  He regained consciousness but remained mute for 12
    days and developed choreiform movements.  On examination 16 years
    later everything was normal with the exception of mildly dysarthric
    speech and some minor motor disturbances.  A computer tomography
    (CT) scan showed bilateral symmetrical infarction of the globus
    pallidus and infarction of the left cerebellar hemisphere (Finelli,
    1981).

         During a fire in an apartment the mother jumped out through the
    window with a baby while twin brothers, aged 2.5 years, tried to get
    out through the front door.  They were subsequently found
    unconscious just inside the door and remained so on admission to
    hospital.  Both children were severely acidotic with pH values of
    6.77 and 6.9, respectively.  Initial treatment consisted of 100%
    oxygen, controlled ventilation, and the intravenous administration
    of sodium thiosulfate (approximately 400 mg/kg body weight).  A few
    hours later hyperbaric oxygen was also employed.  Carbon monoxide
    concentrations were low, being 5.7% and 1.4%, and cyanide
    concentrations in blood were 1.15 mg/l and 1.1 mg/l.  One twin
    remained more severely ill than the other, but both children were
    discharged after 3 weeks without sequelae.

         A man aged 28 years ingested an unknown amount of a cyanide
    solution in an attempt at suicide.  On admission to hospital,
    shortly after ingestion, he was talkative and anxious with an
    intense smell of bitter almonds.  He was hyperventilating slightly
    and had a mild metabolic acidosis (BE -6).  He was given 15 g of
    sodium thiosulfate by slow intravenous infusion.  The patient did
    not become unconscious but instead became calm and mentally clear. 
    The acidosis reverted without further treatment.  The blood cyanide
    concentration was 39.7 µmol/l (9.9 mg/l) and that of thiocyanate,
    150 µmol/l (0.88 mg/l).

         A man aged 32 years ingested an unknown amount of potassium
    cyanide solution in a suicidal attempt.  The patient became
    unconscious after 10 min and was brought to hospital within 40 min
    after ingestion.  He was by then deeply unconscious, slightly
    cyanotic, and unresponsive to pain.  The patient was treated
    immediately with oxygen, 8 g sodium thiosulfate intravenously, and
    gastric lavage.  There was striking improvement after 30 min and
    Kelocyanor was then given, after which the patient became completely
    conscious.  The cyanide concentration in blood was 3.7 mg/l.

         A 54-year-old man ingested 3 g potassium cyanide and arrived in
    hospital unconscious and with respiratory arrest.  The patient was
    resuscitated and 12 g of sodium thiosulfate was given intravenously. 
    There was a moderately severe acidosis (pH 7.31, BE -11) and another
    3 g of sodium thiosulfate was therefore given.  After 30 min the
    patient breathed spontaneously, after 90 min he moved his
    extremities and eyes, and after 6 h he was fully alert.  No sequelae
    were observed.  The cyanide concentration in blood was 0.26 mg/l.

    3.12  Summary of Evaluation

    3.12.1  Indications

         Sodium thiosulfate is indicated in poisoning from cyanide,
    chlorate, bromate, bromine, iodine, cisplatin, mustard gas, and
    nitrogen mustard.  However, this monograph deals only with the use
    of sodium thiosulfate as an antidote in cyanide poisoning

    3.12.2  Route of administration

         In cyanide poisoning, sodium thiosulfate should be given
    intravenously (absorption is poor after oral administration) as a
    bolus injection or by infusion over at least 10 min.  When used to
    prevent cyanide poisoning during sodium nitroprusside therapy, it
    may be given either simultaneously by continuous infusion or,
    alternatively, as a slow bolus injection.

    3.12.3  Dose

         The recommended initial dose for adults in established cyanide
    poisoning is 8 to 12.5 g (Chen et al., 1944; Chen & Rose, 1952), or
    0.2 g/kg body weight (Sörbo, 1972).  This dose is based on
    individual cases where doses of this size have proven effective. 
    Experimental data and theoretical considerations support these
    recommendations, though true validation is lacking.

         For children relatively higher doses are generally recommended. 
    For children with normal haemoglobin concentrations, a dose of
    approximately 410 mg/kg body wt has been suggested (Berlin, 1970)
    and many handbooks suggest doses in the range 300-500 mg/kg body
    weight.  It should be noted that in those sources which make these
    recommendations, sodium thiosulfate is used in combination with
    other antidotes, especially sodium nitrite.

         The risk of cyanide poisoning in patients undergoing treatment
    with sodium nitroprusside is well documented.  Sodium thiosulfate
    has been found to be ideal in this situation, and it has been
    recommended that the w/w ratio for SNP and sodium thiosulfate should
    be at least 1:4 (Schulz et al., 1979b) and preferably, to obtain an
    excess of thiosulfate, 1:5-6.  The antidote may be given either by
    continuous infusion, simultaneously with SNP (Schulz et al., 1982),
    or by bolus injection.

    3.12.4  Other consequential or supportive therapy

         The capacity of sodium thiosulfate to enhance the
    detoxification of cyanide in the body has been established in
    animals and man.  As an antidote in cyanide poisoning, sodium
    thiosulfate alone, together with oxygen and necessary supportive
    therapy, is probably sufficient in mild to moderately severe cases.

    It is also valuable in doubtful cases of poisoning, where it may
    have both therapeutic and diagnostic value.  In severe poisoning,
    sodium thiosulfate should be given together with other antidotes,
    with which it acts synergistically.

    3.13  Model Information Sheet

    3.13.1  Uses

         Sodium thiosulfate is indicated for use in cyanide poisoning.

    3.13.2  Dosage and route of administration (cyanide poisoning)

         The initial dose in  adults is (8 to)12.5 g of sodium
    thiosulfate given as an intravenous bolus injection/infusion over
    10 (to 15) min.  Alternatively, the total initial dose can be
    calculated as 150-200 mg/kg body weight.  Additional doses may be
    indicated according to the clinical course.

         The initial dose in  children is 400 (300-500) mg/kg body
    weight given intravenously as indicated above.

         To prevent cyanide intoxication during SNP therapy, sodium
    thiosulfate  should be given either by simultaneous infusion of a
    dose 5-6 times exceeding (w/w) the SNP dose or, alternatively, a
    bolus injection may be employed.

    3.13.3  Precautions and contraindications

         There are no specific contraindications.  The toxicity of
    sodium thiosulfate is low and toxic effects should not be expected
    unless doses far exceed those recommended.  In patients with renal
    insufficiency, dialysis can be considered for the more rapid
    elimination of thiocyanate (during long-term treatment).

    3.13.4  Adverse effects

         Adverse effects are mild and of minor importance compared to
    the risks associated with cyanide poisoning.  Rapid injection of a
    hyperosmolar sodium thiosulfate solution has caused nausea and
    vomiting (Ivankovich et al., 1983).  Hypotension has been reported,
    due probably to the formation of thiocyanate, which is known to have
    hypotensive properties (Done, 1961).  Other side effects attributed
    to excess thiocyanate production are nausea, headache, and
    disorientation.  When thiosulfate was injected into dogs (Vesey et
    al., 1985) no side effects were seen other than transient
    hypotension.  Diuretic effects and osmotic disturbances are possible
    side effects (Martindale, 1989).

    3.13.5  Use in pregnancy/lactation

         These aspects are seldom discussed in the literature and are of
    little relevance in this context.  In a life-saving situation, the
    dosage recommended above should not be modified in the case of
    pregnancy or lactation.

    3.13.6  Storage

         Injectable thiosulfate should be stored in ampoules.  Storage
    over three years does not cause any significant change in
    composition.  The solid substance may be stored in an airtight
    container for five years without change.

    3.14  References

    Berlin CM (1970) The treatment of cyanide poisoning in children.
    Pediatrics, 46:  793-796.

    Bucht H (1949) On the tubular secretion of thiosulfate and
    creatinine under the influence of caronamide. Scand J Clin Lab
    Invest, 1: 270-276.

    Cardozo RH & Edelman IS (1952) The volume of distribution of sodium
    thiosulfate as a measure of the extracellular fluid space. J Clin
    Invest, 31: 280-290.

    Chen KK & Rose CL (1952) Nitrite and thiosulfate therapy in cyanide
    poisoning. J Am Med Assoc, 149: 113-119.

    Chen KK, Rose CL, & Clowes GHA (1934) Comparative values of several
    antidotes in cyanide poisoning. Am J Med Sci, 188: 767-781.

    Chen KK, Rose CL, & Clowes GHA (1944) The modern treatment of
    cyanide poisoning.  J Indiana Med Assoc, 37: 344-350.

    Crompton M, Palmieri F, Capano M, & Quagliariello E (1974) The
    transport of thiosulphate in rat liver mitochondria. FEBS Lett, 46:
    247-250.

    Darby PW & Wilson J (1967) Cyanide, smoking, and tobacco amblyopia.
    Observations on the cyanide content of tobacco smoke. Br J
    Ophthalmol, 51: 336-338.

    Dennis DL & Fletcher WS (1966) Toxicity of sodium thiosulfate
    (NCS-45624), a nitrogen mustard antagonist, in the dog. Cancer
    Chemother Rep, 50: 255-257.

    Dixon K (1962) Spectrophotometric determination of sodium
    thiosulfate in body fluids by use of an iodine-amylose complex. Clin
    Chim Acta, 7: 453-458.

    Done AK (1961) Cyanide antidotes. In clinical pharmacology of
    systemic antidotes. Clin Pharmacol Ther, 2: 765-768.

    European Pharmacopoeia (1980) 2nd ed. Sainte-Ruffine, France,
    Maisonneuve S.A.

    Evans CL (1964) Cobalt compounds as antidotes for hydrocyanic acid.
    Br J Pharmacol, 23: 455-475.

    Finelli PF (1981) Changes in the basal ganglia following cyanide
    poisoning. J Comput assisted Tomography, 5: 755-756.

    Foulks J, Brazeau P, Koelle ES, & Gilman A (1952) Renal secretion of
    thiosulfate in the dog. Am J Physiol, 168: 77-85.

    Gast JH, Arai K, & Aldrich FL (1952) Quantitative studies on urinary
    thiosulfate excretion by healthy human subjects. J Biol Chem, 196:
    875-884.

    Gilman A, Philips S, & Koelle ES (1946) The renal clearance of
    thiosulfate with observations on its volume of distribution. Am J
    Physiol, 146: 348-357.

    Gladtke E (1966) [The thiosulfate space in the child.] Supplements
    to the Archiv für Kinderheilkunde. Stuttgart, Ferdinand Enke Verlag
    (in German).

    Goldstein F & Rieders F (1953) Conversion of thiocyanate to cyanide
    by an erythrocytic enzyme. Am J Physiol, 173: 287-290.

    Grant WM (1986) Toxicology of the eye, 3rd ed. Springfield,
    Illinois, Charles C.  Thomas.

    Hager HHJ (1977) [Hager's manual of pharmaceutical practice.]
    Berlin, Heidelberg, New York, Springer-Verlag.

    Hall AH & Rumack BH (1987) Hydroxocobalamin/sodium thiosulfate as a
    cyanide antidote. J Emergency Med, 5: 115-121.

    Himwich WA & Saunders JP (1948) Enzymatic conversion of cyanide to
    thiocyanate. Am J. Physiol, 153: 348-354.

    Höbel M, Kreye VAW, & Pill J (1978) Effect of sodium nitroprusside
    alone and in combination with sodium thiosulfate on the acid-base
    balance, and on thiocyanate and iron plasma levels in the rabbit.
    Klin Wochenschr, 56(suppl 1): 147-152.

    Howell SB, Pfeifle CE, Wung WE, Ohlsen RA, Lucas WE, Yon JL, & Green
    M (1982) Intraperitoneal cisplatin with systemic thiosulfate
    protection. Ann Intern Med, 97: 845-851.

    IUPAC (1969) Dissociation constants of inorganic acids and bases in
    aqueous solution. Oxford, United Kingdom, International Union of
    Pure and Applied Chemistry.

    Ivankovich AD, Braverman B, Kanuru RP, Heyman HJ, & Paulissian R
    (1980) Cyanide antidotes and methods of their administration in
    dogs: a comparative study.  Anaesthesiology, 52: 210-216.

    Ivankovich AD, Braverman B, Stephens TS, Shulman M, & Heyman HJ
    (1983) Sodium thiosulfate disposition in humans: Relation to sodium
    nitroprusside toxicity.  Anesthesiology, 58: 11-17.

    Kirk-Othmer (1969) Encyclopedia of chemical technology, 2nd ed. New
    York, John Wiley & Sons, vol 20.

    Krapez JR, Vesey CJ, Adams L, & Cole PV (1981) Effects of cyanide
    antidotes used with sodium nitroprusside infusions: Sodium
    thiosulfate and hydroxocobalamin given prophylactically to dogs. Br
    J Anaesth, 53: 793-804.

    Lang S (1895) [Prussic acid detoxification.] Arch Exp Pathol
    Pharmakol, 36: 75-99 (in German).

    Martindale (1989) In: Reynolds JEF ed. The extra pharmacopoeia, 29th
    ed. London, The Pharmaceutical Press, p 855.

    Michenfelder JD & Tinker JH (1977) Cyanide toxicity and thiosulfate
    protection during chronic administration of sodium nitroprusside in
    the dog: Correlation with a human case. Anesthesiology, 47: 441-448.

    Newton GL, Dorian R, & Fahey RC (1981) Analysis of biological
    thiols: Derivatization with monobromobimane and separation by
    reverse-phase high performance liquid chromatography. Anal Biochem,
    114: 383-387.

    NIOSH (1986) Register of toxic effects of chemical substances. 
    Washington, DC, National Institute for Occupational Safety and
    Health.

    Perschau RA, Modell JH, Bright RW, & Shirley PD (1977) Suspected
    sodium nitroprusside-induced cyanide intoxication. Anesth Analg,
    56(4): 533-537.

    Pharmacopoea Nordica (1964) Editio Svecica Stockholm,
    Apotekarsocietetens förlag.

    Pill H, Engeser P, Hobel M, & Kreye VAW (1980) Sodium nitroprusside:
    Comparison of the antidotal effect of hydroxocobalamin and sodium
    thiosulfate in rabbits. Dev Toxicol Environ Sci, 8: 423-426.

    PoisIndex (1987) Microfiche data base, 53rd ed. Denver, Colorado,
    Micromedex Inc.

    Saunders JP & Himwich WA (1950) Properties of the transulfurase
    responsible for conversion of cyanide to thiocyanate. Am J Physiol,
    163: 404-409.

    Sax NI (1984) Dangerous properties of industrial materials, 6th ed.
    New York, Van Nostrand Reinhold Co.

    Schneider JF & Westley J (1969) Metabolic interrelations of sulphur
    in proteins, thiosulphate and cystein. J Biol Chem, 244: 5735-5744.

    Schubert J & Brill WA (1968) Antagonism of experimental cyanide
    toxicity in relation to the  in vivo activity of cytochrome
    oxidase. J Pharmacol Exp Ther, 162: 352-359.

    Schulz V (1984) Clinical pharmacokinetics of nitroprusside, cyanide,
    thiosulphate and thiocyanate. Clin Pharmacokinet, 9: 239-251.

    Schulz V & Roth B (1982) Detoxification of cyanide in a new-born
    child. Klin Wochenschr, 60: 527-528.

    Schulz V, Bonn R, & Kindler J (1979a) Kinetics of elimination of
    thiocyanate in 7 healthy subjects and in 8 subjects with renal
    failure. Klin Wochenschr, 57: 243-247.

    Schulz V, Bonn R, Kammerer H, Kriegel R, & Ecker N (1979b)
    Counteraction of cyanide poisoning by thiosulphate when
    administering sodium nitroprusside as a hypotensive treatment. Klin
    Wochenschr, 57: 905-907.

    Schulz V, Gross R, Pasch T, Busse J, & Loeschcke G (1982) Cyanide
    toxicity of sodium nitroprusside in therapeutic use with and without
    sodium thiosulphate. Klin Wochenschr, 60: 1393-1400.

    Shea M, Koziol JA, & Howell SB (1984) Kinetics of sodium
    thiosulfate, a cisplatin neutralizer. Clin Pharmacol Ther, 35:
    419-425.

    Sörbo B (1962) Enzymatic conversion of cyanide to thiocyanate. In:
    Brodie BB & Erdos EG ed. Proceedings of the First International
    Pharmacological Meeting, Stockholm, 1961. Volume VI: Metabolic
    factors controlling duration of drug action. New York, Pergamon
    Press, pp 121-128.

    Sörbo B (1972) The pharmacology and toxicology of inorganic sulfur
    compounds. In: Senning A ed. Sulfur in organic and inorganic
    chemistry. New York, Marcel Dekker, vol 2, pp 156-158.

    Sörbo B & Ohman S (1978) Determination of thiosulphate in urine.
    Scand J Clin Lab Invest, 38: 521-527.

    Sylvester DM, Sander C, Hayton WL, & Way JL (1981) Alteration of the
    pharmaco-dynamics of sodium cyanide by sodium thiosulphate. Proc
    West Pharmacol Soc, 24: 135.

    Sylvester DM, Hayton WL, Morgan RL, & Way JL (1983) Effects of
    thiosulfate on cyanide pharmacokinetics in dogs. Toxicol Appl
    Pharmacol, 69: 265-271.

    Szczepkowski TW, Skarzynski B, & Weber M (1961) The metabolic state
    of thiosulphate. Nature (Lond), 189: 1007-1008.

    United States Dispensatory (1973) 27th ed. Philadelphia, Toronto,
    J.B.  Lippincott Co.

    Vesey CJ (1979) Letter to the editor. Clin Toxicol, 14: 307-309.

    Vesey CJ, Krapez JR, Varley JG, & Cole PV (1985) The antidotal
    action of thiosulfate following acute nitroprusside infusion in
    dogs. Anesthesiology, 62: 415-421.

    Way JL, Tamulinas CB, Leung P, Ray L, Nizamani S, Sylvester D, Way
    JL, & Chiou F (1984) Pharmacologic and toxicologic basis of cyanide
    antagonism. Proc West Pharmacol Soc, 27: 149-153.

    Westley J, Adler H, Westley L, & Nishida C (1983) The
    sulfurtransferases. Fundam Appl Toxicol, 3: 377-382.

    Wilson J (1965)  Leber's hereditary optic atrophy: a possible defect
    of cyanide metabolism. Clin Sci, 29: 505-515.

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

    4.  Hydroxocobalamin

    4.1  Introduction

         Hydroxocobalamin is one of a number of antidotes which may be
    used in the treatment of cyanide poisoning.  It acts as a chelating
    agent for cyanide.  Hydroxocobalamin is commonly used in conjunction
    with sodium thiosulfate (Na2S2O3) (Trousse anticyanure,
    Laboratoire Anphar-Rolland).  Another (unregistered) product is made
    by the Pharmacie Centrale des Hôpitaux de Paris.  It consists of 5 g
    hydroxycobalamin in 100 ml water.

    4.2  Name and Chemical Formula of Antidote

         Hydroxocobalamin (OHB12) is a natural form of vitamin B12
    found in humans (vitamin B12a).  This chemical is described in the
    British Pharmacopoeia (1980), the Italian Pharmacopoeia (1985), the
    Pharmacopoée Française (1988), and the United States Pharmacopeia
    (1985).

         CAS number:  13 422-51-0

         Formula:  (dimethyl-5,6-benzimadazolyl) hydroxocobamide

                           C62H89CoN13O15P

    FIGURE 03

         Structure:  OHB12 contains a hydroxyl group attached to a
                     cobalt atom.

         Relative molecular mass:  1346.47.

         Contains not less than 95.0% and not more than 102.0% of
    C62H89CoN13O15P, calculated on a dried weight basis (United States
    Pharmacopeia, 1985)

         Commercial names:  The Anphar-Rolland Cyanide Antidote Kit
    (containing OHB12 and Na2S2O3).  Other proprietary formulations
    currently available contain only 1 mg/ml.  Another (unregistered)
    product is produced by the Pharmacie Centrale des Hôpitaux de Paris. 
    It consists of 5 g hydroxycobalamin in 100 ml water.

    4.3  Physico-chemical Properties

    4.3.1  Characteristics

         Dark red crystals or a red powder
         Hygroscopic
         Odourless.

    4.3.2  Melting-point

         Becomes brown at 200 °C
         Not melted at 300 °C.

    4.3.3  Solubility in vehicles for administration

         Solubility in water: 1 part OHB12 in 50 parts water. 
    (Martindale, 1989).

    4.3.4  Optical properties

         Not relevant.

    4.3.5  Acidity

         The pH of a 0.2% solution in water is greater than 7.5
    (Pharmacopée Française, 1988).  The pH of a 2% solution in water is
    8.0 - 10.0   (USP°XXI).

    4.3.6  Stability in light

         Must be protected from light.

    4.3.7  Thermal stability

         Must be stored at +4 °C (Anphar-Rolland).

         Decreased activity has been observed with increasing
    temperature.

         Hydroxocobalamin solutions are sterilized by filtration.

    4.3.8  Interference with other compounds

         Complexes with basic substances.  Incompatible with reducing
    agents such as ascorbic acid, saccharose, sorbitol, and other B
    vitamins (Mizoule, 1966).  It has been suggested by Evans (1964)
    that hydroxocobalamin interacts with sodium thiosulfate to form a
    new compound, a thiosulfato-cobalamin, which can no longer (or less
    firmly) fix the cyanide ion.

    4.4  Synthesis

         Routes of synthesis                (confidential)

         Manufacturing processes            (confidential)

         OHB12 (4 g) is lyophilized as a freeze-dried form and mixed
    before use with one vial containing 8 g sodium thiosulfate.  The
    resulting solution is administered in 220 ml of 5% dextrose solution
    by intravenous infusion over 20 min.

    4.5  Analytical Methods

    4.5.1  Identification of hydroxocobalamin

    4.5.1.1  UV spectroscopy

         The visible absorption spectrum of the solution, prepared as
    for pH-dependent cobalamins, is maximal at 426 (±2)nm, 5l6 (±2)nm,
    and 550 (±2)nm (United States Pharmacopeia, 1985).

         OHB12 absorbs light at three other wavelengths:  274 nm,
    351 nm, and 523 nm (British Pharmacopoeia, 1980; Pharmacopèe
    Française, 1988).

    4.5.1.2  Colorimetric method

         Fuse a mixture of 1 mg of OHB12 and 50 mg of potassium
    pyrosulfate in a porcelain crucible.  Cool, break up the mass with a
    glass rod, add 3 ml of water, and boil until dissolved.  Add 1 drop
    of phenolphthalein TS and sodium hydroxide (2 mol/l) drop by drop
    until a pink colour appears.  Add 0.5 g of sodium acetate, 0.5 ml of
    acetic acid (1 mol/l), and 0.5 ml of nitroso R salt solution (1 in

    100): a red or orange-red colour should appear immediately.  Add
    0.5 ml of hydrochloric acid and boil for l min:  the red or
    orange-red colour should persist (United States Pharmacopeia, 1985).

    4.5.2  Quality controls

         Information not available.

    4.5.3  Raw materials

         Foreign pigments, acidic impurities, acetate, chloride, and
    sulfates should be lower than the following limits:

         *    Other cobalamins should constitute not more than 3.0% when
              determined by chromatography (diethylaminoethylcellulose
              column then carboxymethylcellulose column; and measurement
              of absorption at 361 nm) (British Pharmacopoeia, 1980).

         *    Coloured impurities < 5.0% with descending paper
              chromatography and measurement of absorption at 361 nm.

         *    Acidic impurities < 3.0% with a
              diethylaminoethylcellulose column and measurement of
              absorption at 351 and 361 nm.

         *    pH-dependent cobalamins between 95.0% and 102.0%,
              calculated on a dry weight basis using a
              spectrophotometric method (550 nm) with two solutions of
              OHB12.  One of these solutions, in a pH 4.0 buffer,
              serves as the blank, the other solution should be in a pH
              9.3 buffer (United States Pharmacopeia, 1985).

         *    Cyanocobalamin < 5.0%, calculated on a dry weight basis
              and determined with a cobalamin radiotracer assay using a
              cyanocobalamin tracer reagent.

    4.5.4  Finished galenic form

         *    OHB12 is in freeze-dried form

         *    After reconstitution with thiosulfate:

              -    pH 10-12
              -    the liquid is a clear, deep-red solution.

    4.5.5  Measurement

    4.5.5.1  In raw materials and in finished form

         *    Spectrophotometric measurement at 351 nm (British
              Pharmacopoeia, 1980; Pharmacopée Française, 1988)

         *    Radioimmunoassay with labelled cyanocobalamin (United
              States Pharmacopeia, 1985).

    4.5.5.2  In biological samples

         Differential spectroscopy at two wavelengths, 351 nm (maximum
    absorbance for OHB12) and 361 nm (maximum absorbance for
    cyanocobalamin), is used to measure OHB12 and cyanocobalamin
    levels in plasma and in urine (Baud et al., 1987).

    4.6  Shelf-life

         Keep in the dark, in a refrigerator at +4 °C, for no more than
    three years.

    4.7  General Properties

         Kaczka et al. (1950) have demonstrated in  in vitro studies
    that the reaction of OHB12 with cyanide results in the
    displacement of a hydroxyl group by a cyano group to form
    cyanocobalamin, which is then excreted in the urine.  Thus one
    molecule of OHB12 binds one molecule of cyanide.  Hydroxocobalamin
    has been shown to be an effective antidote for experimental cyanide
    poisoning in mice (Mushett et al., 1952), guinea-pigs (Posner et
    al., 1976a), baboons (Posner et al., 1976b), and dogs (Rose et al.,
    1965).

         The antidotal effect of OHB12 is enhanced by the use of
    thiosulfate in the same solution.  Thiosulfate provides sulfur
    radicals, which complex with cyanide to form thiocyanate using the
    endogenous cyanide detoxification mechanism rhodanese (thiosulfate
    sulfurtransferase).  However, the action of thiosulfate may be too
    slow to prevent death in serious cases of cyanide poisoning when
    administered alone.  The combination of OHB12 and thiosulfate was
    shown to be more effective by Mizoule (1966), who studied it
    extensively in experimental animals;  other authors have reported
    similar results (Motin et al., 1970; Luher et al., 1971; Yacoub et
    al., 1974; Jouglard et al., 1974; Friedberg & Shukla, 1975; Racle et
    al., 1976; Bismuth et al., 1984).

    4.8  Animal Studies

    4.8.1  Pharmacokinetics

         No data are available.

    4.8.2  Pharmacodynamics in the presence of the toxin

         The antidotal action of OHB12 has been studied in animals,
    firstly in the case of medium-term intoxication (Haguenoer et al.,
    1975), and secondly in the prophylactic treatment of cyanide
    intoxication (Posner et al., 1976a).

         In the study by Haguenoer et. al. (1975), five groups of Wistar
    male rats (mean weight, 330g) were used as follows:

         *    group 1 received injections of sodium chloride (0.9%) in
              water (control group)

         *    group 2 received 1 ml of 50 mg/ml acetonitrile

         *    group 3 received 1 ml of 50 mg/ml acetonitrile plus 1 ml
              of 5 mg/ml OHB12 (6 h later)

         *    group 4 received 1 ml of 1.12 mg/ml potassium cyanide
              (KCN)

         *    group 5 received 1 ml of 1.12 mg/ml KCN plus 1 ml of
              5 mg/ml OHB12 (6 h later).

         In all, 28 intraperitoneal injections were given over a 6-week
    period.  Rats treated with OHB12 (groups 3 and 5) excreted less
    free CN- in the urine than the others (groups 2 and 4); the
    excretion of combined CN- was higher for groups 3 and 5.  Urinary
    excretion was higher in the case of KCN than acetonitrile.  However,
    a significant proportion of acetonitrile and cyanhydric acid (formed
    from acetonitrile and KCN) was excreted by the lungs, and a large
    amount of thiocyanate was excreted in the faeces.

         In the second study (Posner et al., 1976a), nine groups of
    healthy baboons (weighing 20 to 30 kg) were anaesthetized and
    received nitroprusside by infusion, either alone (four animals) or
    in combination with OHB12 (22.5 mg B12 per 1 mg nitroprusside)
    (five baboons).

         The infusion was continued for up to 2 h or until 500 mg of
    nitroprusside had been administered.  Cyanide concentrations
    increased and the animals developed a severe metabolic acidosis. 
    OHB12 infused simultaneously with nitroprusside significantly
    reduced the increase in cyanide concentrations and eliminated the
    development of metabolic acidosis.

         In a third study (Höbel et al., 1980), sodium nitroprusside
    (SNP) and OHB12 were infused over 4 h in conscious rabbits in
    molar ratios of 1:4, 1:5, or 1:8.  Control animals received OHB12
    only.  Sodium thiosulfate was infused with SNP at molar ratios of
    1:4, 1:5, and 1:10.  SNP provoked a severe acidosis which was not
    corrected by the lowest dose of OHB12: three deaths occurred among

    ten animals.  When the 1:5 ratio was administered, the acidosis was
    less marked, but three out of seven animals succumbed.  The highest
    dose (1:8) prevented acidosis, but not death, in three out of eight
    animals.  All doses of OHB12 caused histological changes in the
    liver, myocardium, and kidney.  By contrast, sodium thiosulfate was
    completely effective as an antidote in this study and did not give
    rise to histological changes.

    4.8.3  Toxicology

         OHB12 has weak intrinsic toxicity and may be administered
    safely in large doses (Mizoule, 1966).

    4.8.3.1  Acute toxicity

         The acute toxicity of OHB12 has been tested in mice and
    rabbits.  A solution containing 0.4-0.75 g or l g/kg was
    administered to Swiss strain mice (0.20 ml per 20 mg body weight)
    without any toxic effects either immediately or a few days later. 
    Doses of OHB12 (50 and 100 mg/kg body weight) were injected
    intravenously into five rabbits (weighing 2-3 kg) without causing
    side effects either immediately or 3 days later (Mizoule, 1966).

    4.8.3.2  Sub-acute and chronic toxicity

         The sub-acute and chronic toxicity of hydroxocobalamin has been
    tested in rats and rabbits.  Twelve young male Wistar rats (weight
    43-53 g) received three subcutaneous injections weekly for a period
    of 82 days.  Six of these rats were treated with 50 mg OHB12
    injections, while six received injections of sodium chloride (0.9%)
    in water.  No significant differences were reported in the behaviour
    of the rats or in the haematological results.  Rabbits received one
    injection of OHB12 (50 mg/kg) every other day for a period of 10
    days.  No toxicity was reported (Mizoule, 1966).

    4.9  Volunteer Studies

         Data not available.

    4.10  Clinical Studies

         After intravenous infusion, OHB12 is bound sparingly to a
    specific globulin transport plasma protein, transcobalamin II
    (Herbert & Sullivan, 1964).  Lesser amounts are bound to the storage
    protein transcobalamin I (a glycoprotein) and to transcobalamin III
    (an intermediate glycoprotein).  A small amount may be free or very
    loosely bound to protein.  In the fasting state, the majority of
    circulating vitamin B12 is bound to transcobalamin, and it is
    stored principally in the liver and bone marrow.

         OHB12 has a very short half-life (5 min), being rapidly
    metabolized and excreted (Vesey & Cole, 1981; Cottrell et al.,
    1982).  Up to 50% of an administered dose of OHB12 is excreted
    unchanged in the urine, which becomes red-orange in colour following
    a dose of 4 g OHB12 (Goodman & Gilman, 1975).

    4.11  Clinical Studies - Case Reports

         Only a few studies in man have included measurement of free and
    total cyanide concentrations.  These include 12 cases of acute
    cyanide poisoning (Baud et al., 1986; Danel et al., 1986a);
    several cases of sodium nitroprusside poisoning in which OHB12 was
    used prophylatically (Cottrell et al., 1982), and one case of
    combined cyanide and carbon monoxide poisoning (Baud et al., 1987).

         Baud et al. (1986) reported the case of a 55-year-old worker,
    exposed to propionitrile by inhalation and dermal exposure, who was
    treated with a combination of OHB12 and thiosulfate.  The patient
    rapidly lost consciousness: 25 min after exposure he was comatose
    and unresponsive with a pulse rate of 90 beats/min and a blood
    pressure of 17.3/9.3 kPa (130/70 mmHg).  Laboratory investigations
    revealed no abnormalities except for a base deficit of 14-16 mmol/l
    on arterial blood gas analysis.  The serum lactate concentration was
    10 mmol/l.  The patient received 4 g OHB12 and 8 g thiosulfate
    (Trousse anticyanure, Laboratoire Anphar Rolland) intravenously in
    just under 30 min.  During infusion of the antidote, the patient
    regained consciousness and became oriented.  Complete reversion of
    CNS depression and normalization of vital signs were observed during
    the next hour.

    Table 3.  Blood cyanide and thiocyanate concentrations

                                                              

                Time after          Cyanide       Thiocyanate
               intoxication
                                                              

                    2 h            5.71 mg/l     undetectable
                   2.5 h           0.93 mg/l       21.1 mg/l
            (at the end of the
            antidote infusion)

                    7 h            1.00 mg/l    baseline level

                 

    a  Personal communication by V. Danel, L. Barret, and J.L. Debru
         Regional Hospital Centre and University, Grenoble, France
         concerning four hydrogen cyanide intoxications treated with
         OHB12 and sodium hyposulfite.

         Baud et al. (1989) studied the toxicokinetics of OHB12 and
    cyanocobalamin (CNB12),which were measured simultaneously in cases
    of cyanide intoxication treated with OHB12.  Plasma cobalamin
    levels were measured using a differential spectrophotometric assay. 
    Whole blood cyanide levels were measured, using a
    microdiffusion/colormetric assay, in samples taken before
    administration of OHB12 by infusion (5 g over 30 min).

                                                                                  

    Source of cyanide         Bromo   Sodium   Mercurous    Smoke inhalation
                             cyanide  cyanide   cyanide
    Route of absorption       lung    orally    orally   lung  lung  lung  lung
                                                                                  

    Initial blood cyanide      13       260      217      22    123   129   208
    (µmol/l)

    Peak OHB12 (µmol/l)       325        15       83     514     74   131     0

    Peak CNB12 (µmol/l)        39       275      226       0    215   221   259
                                                                                  
    
         After completion of the OHB12 infusion, low levels of OHB12
    were observed concurrently with high levels of CNB12 in patients
    with high whole blood cyanide levels, but not in those with low
    cyanide concentrations.  These findings suggest that OHB12 acts
    rapidly as a cyanide antidote.

         Cottrell et al. (1982) studied two groups of patients in whom
    cyanide toxicity was induced by nitroprusside.  Group I received
    nitroprusside alone, while group II received nitroprusside and
    OHB12.  Red cell and plasma cyanide concentrations were,
    respectively, 83.44% (±23.12) and 3.51 (±1.01) mg per 100 ml after
    administration of nitroprusside alone and 33.18% (±17.29) and 2.18%
    (±0.65) mg per 100 ml after administration of nitroprusside with
    OHB12.  OHB12 infusion resulted in lower blood cyanide
    concentrations and base deficit than in the untreated group.  The
    dose of nitroprusside used in each group did not differ
    significantly.

         Following inhalation of fumes from a firm making plastics, a
    man became poisoned by both cyanide and carbon monoxide (Baud et
    al., 1989).  On admission, the whole-blood cyanide concentration was
    3.6 mg/l.  He immediately received 4 g OHB12 and 8 g sodium
    thiosulfate by intravenous infusion.  A second injection had to be
    administered 3 h after admission because of clinical relapse, again
    with a satisfactory response.

    4.12  Summary of Evaluation

    4.12.1  Indications

         *    In the treatment of pernicious anaemia and vitamin B12
              deficiency states by intramuscular administration.

         *    In cyanide poisoning as an antidote:

              a)   as therapeutic treatment after exposure to
                   acetonitrile, propionitrile, or potassium or sodium
                   cyanides by oral, inhalational, or dermal routes;

              b)   as prophylactic treatment during sodium nitroprusside
                   infusion.

    4.12.2  Advised route and dosage

         OHB12 is usually given intravenously in a 5% solution of
    dextrose in water.  The normal adult dose is 4 g OHB12, but the
    dose may be increased in massive poisoning.

    4.12.3  Practical advice

         *    To maintain normal blood pH and isotonicity, the OHB12
              must be injected in more than 220 ml of 5% dextrose.

         *    OHB12 must not be administered with reducing or basic
              substances (see section 4.3.8).

         *    When OHB12 is administered as a prolonged infusion,
              aluminium foil should be used to protect it from light.

    4.12.4  Side effects

         *    The most common side effect is an orange-red discoloration
              of the skin, mucuous membranes, and urine that lasts about
              12 h but is not consequential.

         *    Allergic reactions to OHB12 have been reported (Dally &
              Gaultier, 1976).

    4.13  Model Information Sheet

    4.13.1  Uses

         Hydroxocobalamin is used for the treatment of pernicious
    anaemia and vitamin B12 deficiency states.  In the treatment of
    cyanide intoxication, hydroxocobalamin is indicated in two
    circumstances.  Firstly, it is used for the treatment of exposure to
    hydrogen cyanide, inorganic cyanide salts, and acetonitrile and

    propionitrile.  Secondly, hydroxocobalamin is used prophylactically
    to prevent cyanide intoxication as a result of sodium nitroprusside
    given by infusion.

    4.13.2  Dosage and route

         Hydroxocobalamin is given intravenously in a 5% dextrose
    solution.  The usual adult dose is 4 g, which may be increased in
    cases of massive cyanide poisoning.

    4.13.3  Precautions/contraindications

         Hydroxocobalamin should be diluted in more than 220 ml of 5%
    dextrose solution.  It must not be administered with reducing
    substances or basic agents.  In the case of prolonged infusion, the
    solution must be protected from light.  The possibility of an
    anaphylactoid reaction to hydroxocobalamin should be borne in mind.

    4.13.4  Adverse effects

         The most common side effect is an orange-red discoloration of
    the skin, mucous membranes, and urine that lasts for approximately
    12 h.  Allergic reactions have been reported.  Histological
    abnormalities in the liver, myocardium, and kidney of rabbits with
    nitroprusside intoxication treated with hydroxocobalamin have been
    described (Höbel et al., 1980).

    4.13.5  Use in pregnancy and lactation

         These aspects are seldom discussed in the literature, and are
    of little relevance in this context.  In a life-saving situation,
    the dosage recommended above should not be modified during pregnancy
    or lactation.

    4.13.6  Storage

         Hydroxocobalamin must be kept in the dark, in a refrigerator at
    4 °C, for not more than 3 years.

    4.14  References

    Baud FJ, Bismuth C, Astier A, Djeghour H, & Aubriot D (1986)
    Toxicocinétique des cyanures et des thyocyanates lors d'un accident
    professionel au propionitrile. Arch Mal Prof Méd Trav Sécur Soc,
    47(2): 85-86.

    Baud FJ, Astier A, Toffis V, & Barriot P (1987) Plasma kinetics of
    hydroxocobalamin and cyanocobalamin in a treated cyanide poisoning.
    Paris, Hôpital Fernand Widal (Unpublished data).

    Baud FJ, Toffis V, Certain B, Muzynski J, Astier A, & Bismuth C
    (1989) Hydroxocobalamine antidote des cyanures. Vérification
    cinétique de sa transformation en cyanocobalamine dans l'organisme
    intoxiqué. Thérapie, 44: 469.

    Bismuth C, Cantineau JP, Pontal P, Baud F, Garnier R, Poulos L, &
    Bolo A (1984) Priorité de l'oxygénation dans l'intoxication
    cyanhydrique: A propos de 25 cas. J Toxicol Méd, 4(2): 107-121.

    British Pharmacopoeia (1980) London, Her Majesty's Stationery
    Office.

    Cottrell JE, Casthely P, Brodie JD, Patel K, Klein A, & Turndorf H
    (1982) Prevention of nitroprusside-induced toxicity with
    hydroxocobalamin. New Engl J. Med, 298: 809-811.

    Dally S & Gaultier M (1976) Choc anaphylactique dû à
    l'hydrocobalamine. Nouv Presse Méd, 30: 1917.

    Evans CL (1964) Cobalt compounds as antidotes for hydrocyanic acid.
    Br J Pharmacol, 23: 455-475.

    Friedberg KD & Shukla UR (1975) The efficiency of aquocoblamine as
    an antidote in cyanide poisoning when given alone or combined with
    sodium thiosulfate. Arch Toxicol, 33: 103-113.

    Goodman LS & Gilman A ed. (1975) The pharmacological basis of
    therapeutics, 5th ed. New York, MacMillan Publishing Co., Inc., p
    1334.

    Haguenoer JM, Dequidt J, & Jacquemont MC (1975) Intoxications
    expérimentales par l'acétonitrile. 4ème note: Influence de
    l'hydroxocobalamine sur l'intoxication à moyen terme. J Eur Toxicol,
    2: 113-121.

    Herbert V & Sullivan LW (1964) Activity of coenzyme B12 in man.
    Ann NY Acad Sci, 112: 855-870.

    Höbel M, Engeser P, Nemeth L, & Pill J (1980) The antidote effect of
    thiosulphate and hydroxocobalamin in formation of nitroprusside
    intoxication of rabbits. Arch Toxicol, 46: 207-213.

    Italian Pharmacopoeia (1985) 9th ed. Rome, Libreria della Stato.

    Jouglard J, Nava G, Botta A, Michel-Manuel C, Poyen B, & Mattei JL
    (1974) A propos d'une intoxication aiguë par le cyanure de potassium
    traitée par l'hydroxocobalamine. Marseille Méd, 12: 617-624.

    Kaczka EA, Wolf DE, Kuehl FA Jr, & Folkers K (1950) Vitamin B12:
    reactions of cyano-cobalamin and related compounds. Science, 112:
    354-355.

    Luher F, Dusoleil P, & De Montgros J (1971) Action de
    l'hydroxocobalamine à dose massive dans l'intoxication aiguë au
    cyanure: A propos d'un cas. Arch Mal Prof, 32: 683-686.

    Martindale (1989) In: Reynolds JEF ed. The extra pharmacopoeia, 29th
    ed. London, The Pharmaceutical Press, pp 1260-1261.

    Mizoule J (1966) Etude de l'action antidote de l'hydroxocobalamine à
    l'égard de l'intoxication cyanhydrique. Paris, Université de Paris
    (Thèse de Doctorat en Pharmacie).

    Motin J, Bouletreau P, & Rouzioux JM (1970) Intoxication
    cyanhydrique grave traitée avec succès par l'hydroxocobalamine. J
    Méd Strasbourg, 1: 717-722.

    Mushett CW, Kelly KL, Boxer EG, & Rickards JC (1952) Antidotal
    efficacy of vitamin B12a (hydroxocobalamin) in experimental
    cyanide poisoning. Proc Soc Exp Biol Med, 81: 234-237.

    Pharmacopée Française (1988) 10th ed. Paris, La Commission nationale
    de Pharmacopée.

    Posner MA, Tobey RE, & McElroy H (1976a) Hydroxocobalamin therapy of
    cyanide intoxication in guinea-pigs. Anaesthesiology, 44(2):
    157-160.

    Posner MA, Rodkey FL, & Tobey RE (1976b) Nitroprusside induced
    cyanide poisoning:  Antidotal effect of hydroxocobalamin.
    Anaesthesiology, 44(4): 330-335.

    Racle JP, Chausset R, & Dissait F (1976) L'intoxication cyanhydrique
    aiguë. Aspects biologiques, toxicologiques, cliniques et
    thérapeutiques. Rev Méd Clermont-Ferrand, 5: 371-382.

    Rose CL, Worth RM, & Chen KK (1965) Hydroxocobalamin and acute
    cyanide poisoning in dogs. Life Sci, 4: 1785-1789.

    United States Pharmacopeia (1985) 21st ed. Rockville, Maryland,
    United States Pharmacopeial Convention, Inc.

    Vesey GJ & Cole PV (1981) Cyanide antagonist (letter). Can Anaesth
    Soc J, 28: 290.

    Yacoub M, Faure J, Morena H, Vincent M, & Faure H. (1974)
    L'intoxication cyanhydrique aiguë:  Données actuelles sur le
    métabolisme du cyanure et le traitement par l'hydroxocobalamine.
    J Eur Toxicol, 7: 22-29.

    5.  DICOBALT EDETATE (KELOCYANOR)

    5.1  Introduction

         The ability of cyanide to form complexes with cobalt is an
    example of the propensity of cyanide to form stable complexes with
    many transition metals (Sharpe, 1976).  It is a property of cyanide
    that has been known since the last century, but there is still room
    for disagreement about the precise nature of the complex formed  in
     vivo when cobalt is used to treat cyanide poisoning.   In vitro,
    a complex of one cobalt and five cyanide atoms is formed on addition
    of potassium cyanide to cobalt (II) salts (Cotton & Wilkinson,
    1966).

    FIGURE 04

         However, in the studies carried out  in vivo by Evans (1964),
    it was suggested that inorganic cobalt salts could bind six moles of
    cyanide for each mole of cobalt.  Evans further proposed that there
    was a two stage reaction, six moles of cyanide reacting with one
    mole of cobalt, forming first the cobaltocyanide ion, and then the
    cobalticyanide ion.

                      Co2+ + 6 CN-  -->  [Co (CN)6] 4-
                                             |
                                             |--> e-
                                             |
                                             v
                                         [Co (CN)6] 3-

         The cobalticyanides are not very toxic, much less so than the
    same amount of free cyanide, and have an LD50 in the region of
    1 g/kg.

         Although it was known that cobalt and cyanide formed stable
    complexes, it was generally thought that inorganic cobalt salts were
    too toxic for use in humans as cyanide antidotes.  In particular,
    those most investigated, the cobalt (II) salts, were known to be
    toxic to the heart, liver, and kidney (Speijers et al., 1982).  For
    a long time, during which the preferred treatment for cyanide
    poisoning was sodium nitrite and sodium thiosulfate, interest in the
    anticyanide activity of cobalt compounds dwindled.  A rekindling of
    interest in cobalt compounds, especially organic complexes, was
    brought about by the discovery by Mushett et al. (1952) that the
    complex organocobalt vitamin, hydroxocobalamin, possessed
    anticyanide activity in mice.  Experimental work, notably by Paulet
    (1957, 1958), led to the introduction of dicobalt edetate
    (Kelocyanor).

    5.2  Name and Chemical Formula

         Kelocyanor is the proprietary name for a solution containing
    dicobalt edetate and glucose.  Dicobalt edetate has not been
    crystallized.  The structural formula was believed to be that of the
    monocobalt salt of a monocobalt edetate anion (Fig. 4), but there is
    reason to think that this does not fully represent the structure of
    the active ingredient of Kelocyanor (see below).  The CAS number is
    36499-65-7, the INN is dicobalt edetate, and the IUPAC name is
    dicobalt ethylenediamine- NNN'N'-tetraacetate.  Other names are
    cobalt edetate, cobalt tetracemate and cobalt EDTA.

    FIGURE 05

    For reasons discussed in the text, this may only be an approximation
    to the true structure of the material in Kelocyanor.

    5.3  Physico-chemical Properties

         The usual preparation of dicobalt edetate is Kelocyanor.a
    Injectable Kelocyanor is a clear violet solution, produced in clear
    glass 20-ml ampoules, of pH 4.0-7.5.  It contains 0.196-0.240
    g/100 ml free cobalt and 1.35-1.65 g/100 ml dicobalt edetate, as
    well as 4 g glucose per ampoule.  The refractive index of the
    solution in the ampoule is 1.3638.

         The optical rotation of Kelocyanor is + 9.25°: the optical
    rotation of a dextrose solution of the same dextrose concentration
    as Kelocyanor is +11°.  Therefore the net optical rotation
    attributable to the cobalt edetate is -1.75°.  This would give a
    specific rotation for dicobalt edetate of -116.7°.  The relative
    molecular mass of dicobalt edetate is 408.  No information is
    available on the specific gravity.

    5.4  Synthesis

         Dicobalt edetate is made by adding cobalt carbonate to sterile
    water and then adding ethylenediaminetetraacetic acid.  After the
    addition of glucose, the resultant solution is cleared with
    activated charcoal.  Sterility and pyrogen tests are performed to
    the requirements of the European Pharmacopoeia (1980).  The volume
    of solution in the ampoules is checked using method B (British
    Pharmacopoeia, 1980).  The material is not checked for the absence
    of activated charcoal, but this would be expected to be removed
    during filtration through a membrane with a pore diameter of
    0.22 µm.

    5.4.1  Source of materials

    5.4.1.1  Cobalt carbonate

         The current supplier to both the French and British drug
    companies is Prolabo, BP No 200, 75526 Paris, Cedex 11, France.  The
    specification is as follows:

              appearance:              fine wine-coloured powder
              identification:          cobalt carbonate confirmed
              loss on drying:          about 14%
              assay:                   about 50% cobalt

                 

    a  L'Arguenon, Société d'Etudes et de Recherches Biologiques,
         53 Rue Villiers de l'Isle Adam, 75020 Paris, France, UK
         suppliers, Lipha Pharmaceuticals Ltd, Harrier House, Yiewsley,
         Middlesex UB7 7QG.

    5.4.1.2  Ethylenediaminetetraacetic acid (EDTA)

         The current supplier is BDH Ltd, Poole, Dorset, BH12 4NN,
    England.  The material complies with the US National Formulary (see
    USNF, 1985).

    5.4.1.3  Glucose

         This complies with the European Pharmacopoeia.

    5.5  Analytical Methods

    5.5.1  Free cobalt

         In duplicate, pipette 5.0 ml of the solution in the ampoule
    into a 250- ml conical flask.  Add 100 mg of murexide compound
    mixture and 100 ml water.  Stir using a magnetic stirrer.  Add
    dilute ammonia drop by drop until the solution just turns yellow
    (usually 1 drop).  Titrate with disodium edetate (0.02 mol/l) to a
    violet  end-point.  During the titration, the pH of the solution
    falls drastically and it turns red.  It is necessary to add 1 drop
    of dilute ammonia to restore the yellow colour.  Check the end point
    of the titration with 1 drop of ammonia.

         To prepare the murexide mixture, grind 10 mg murexide and
    990 mg sodium chloride in a pestle and mortar.  The dilute ammonia
    solution consists of 5.5 ml concentrated ammonia diluted to 100 ml
    with water.

         The following equation is used to calculate the concentration
    of free cobalt (g/100 ml):

                  T  x  1.179  x  F  x  20
                                          
                             1000

       where   T  =  titre (ml)
               F  =  factor for disodium edetate at 0.02 mol/l
               1 ml disodium edetate (0.02 mol/l) equiv. 1.179 mg cobalt

    5.5.2  Dicobalt edetate

         In duplicate, pipette 5.0 ml of the solution in the ampoules
    into a 250-ml conical flask.  Add 0.65 g potassium cyanide, stir,
    and allow to stand for 5 min.  Add 85 ml of water, 10 ml lead
    acetate solution, and 1 ml glacial acetic acid.  Stir for 15 min,
    using a magnetic stirrer.  Add 100 mg xylenol orange compound
    mixture and titrate slowly with disodium edetate (0.02 mol/l) to a
    clear yellow end-point (T1).  Carry out a blank titration,
    omitting the sample solution (T2).

         Xylenol orange compound mixture is prepared by grinding 10 mg
    xylenol orange with 990 mg potassium nitrate in a pestle and mortar. 
    The lead acetate solution consists of 1.5 g lead acetate dissolved
    in water and make up to 100 ml.  The following equation is used to
    calculate the concentration of  dicobalt edetate (g/100 ml):

                  (T2-T1)  x  F  x  4.06
                                        
                           25

       where   T1  =  sample titre
               T2  =  blank titre
               F  =  factor for disodium edetate (0.02 mol/l)

    5.5.3  Analysis in biological fluids

         Since dicobalt edetate has not been fully characterized, it is
    difficult to say whether analysis would be meaningful.  Methods are
    available for the analysis of cobalt in biological fluids by
    spectrophotometry or atomic absorption flame photometry (Thiers et
    al., 1955; Mulford, 1966; Christian, 1969; Lewis et al., 1985).

         Dicobalt edetate interferes with the Feldstein & Klendshoj
    (1954) method of analysis of biological fluids for cyanide
    (Ballantyne & Marrs, 1987).  See chapter 10 for further details.

    5.6  Stability and Shelf-life

         The shelf-life is 3 years at 25 °C, and the material should be
    stored in the dark.  The drug has been stored for up to 3 years at
    room temperature 20-25, 30 and 37 °C (Lipha Pharmaceuticals,
    1987)a,  the batches having been tested at regular intervals
    (12 months maximum) for compliance with the finished product
    specification with respect to appearance, extractable volume, pH,
    and free cobalt and dicobalt edetate content.  The only deviation
    noted was that the colour of the solution became lighter at 30 °C
    and above.   All the other variables remained within the
    specification.  If stored in the light, the drug bleaches in about
    one month.  The nature of the chemical change that brings about the
    loss of colour is not known.

                 

    a  Personal communication from Lipha Pharmaceuticals Ltd, West
         Drayton, Middlesex, United Kingdom.

    5.7  General Properties

         Cyanide blocks intracellular respiration by binding to
    cytochrome oxidase.  Cobalt forms a stable complex with cyanide, the
    effect being direct and peripheral (Mercker & Bastian, 1959). 
    Although some cobalt compounds are reputed to inhibit methaemoglobin
    reductase (Hagler & Coppes, 1982), it is unlikely that
    methaemoglobinaemia contributes meaningfully to the anticyanide
    action of dicobalt edetate.

    5.8  Animal Studies

    5.8.1  Pharmacokinetics

         Formal pharmacokinetic studies have not been carried out on
    dicobalt edetate.  Some of the effects observed with cobalt
    compounds, including dicobalt edetate, are probably centrally
    mediated (Bartelheimer, 1962a).  Thus, it is likely that after
    Kelocyanor injection, cobalt or dicobalt edetate crosses the
    blood-brain barrier.  In the mouse, the cyanide-cobalt complex is
    excreted in the urine (Frankenberg & Sörbo, 1975).

    5.8.2  Pharmacodynamics

    5.8.2.1  Efficacy in animals

         The introduction of dicobalt edetate followed from the work of
    Paulet (1957, 1958, 1960a,b, 1961, 1965) and Paulet et al. (1960),
    who studied the antidotal effects and toxicity of various cobalt 
    compounds.  The aim of studies on cobalt antidotes at this time was 
    to find a compound that retained the ability of cobalt to bind
    cyanide, yet lacked its toxic effects.  The compounds investigated
    by  Paulet (1960a,b) included certain cobalt (II) salts, namely the 
    chloride, glutamate, and gluconate.  Paulet (1960a) also studied the
    cobalt chelates, dicobalt edetate, cobalt histidine, and disodium
    monocobalt edetate and found that the salts, as well as all except
    the last named of the chelates, possessed appreciable anticyanide
    activity.  Dicobalt edetate and cobalt histidine were the most
    promising, and in one study dicobalt edetate seemed preferable
    (Paulet, 1960a, 1961).  This study was undertaken in anaesthetized
    dogs infused with sodium cyanide (0.1 mg/kg per min):  both dicobalt
    edetate and cobalt histidine were capable of resuscitating the dogs
    in secondary apnoea, but the dose of cobalt histidine required was
    considerably higher than that of dicobalt edetate.  In the same
    reports, the effective doses of the two compounds were compared with
    their respective intraperitoneal LD50 values in mice, the result
    favouring dicobalt edetate.  However, if the dog results had been
    tabulated against the rat intravenous LD50 values (Tauberger &
    Klimmer, 1963), cobalt histidine might have been preferred, this
    antidote having a much higher intravenous LD50 than dicobalt
    edetate.

         Evans (1964) investigated the stoichiometry of the reaction
    between dicobalt edetate or other cobalt compounds on the one hand,
    and cyanide on the other.  He studied their antidotal effects using
    various doses of both toxicant and treatment in both mice and
    rabbits, designing his experiments in such a way that it was
    possible to calculate the cyanide/cobalt molar ratios above which
    the antidote would no longer be efficacious.  He found that, while
    most cobalt (II) salts would be effective against six moles of
    cyanide/mole cobalt,  dicobalt edetate only possessed efficacy at
    molar ratios of up to two.   Whilst on a mass basis, dicobalt
    edetate is about half as toxic as cobalt (II) salts, on a molar
    basis it is not conspicuously less toxic than cobalt acetate.  It
    had been thought that dicobalt edetate  was the monocobalt (II) salt
    of a monocobalt edetate anion, one cobalt being completely
    unavailable for cyanide binding by virtue of its complexing with
    EDTA, the other being fully available.  In fact the study of Evans
    (1964) suggested that one cobalt is completely  unavailable whilst,
    of the coordination sites of the second, only two  are available to
    bind cyanide.  In that case, the structure given in  Fig 4 does not
    fully describe the structure of dicobalt edetate.  It seems likely,
    however, that the second cobalt atom is required for anticyanide
    activity, since disodium monocobalt edetate, which is a compound of
    very low toxicity (Eybl et al., 1959), is almost ineffective as a
    cyanide antidote (Paulet, 1958).  It may be that the  same property,
    ionizability, is responsible for cyanide binding and for cobalt
    toxicity.  In that case, complexing cobalt would decrease the
    therapeutic effect in parallel with the toxicity.  That this is not
    the case is suggested by, firstly, the lack of toxicity of
    hydroxocobalamin, which is nevertheless a successful cyanide
    antidote  and, secondly, the many years of successful use of
    dicobalt edetate.

    5.8.2.2  Comparison of dicobalt edetate with other compounds

         A number of research groups have compared cobalt edetate with
    other cyanide antidotes, from the point of view both of antidotal
    efficacy (Table 4) and also of toxicity.  For example, using dogs
    and rabbits, Paulet  (1960a, 1961) compared the cobalt chelate with
    sodium nitrite and found the former superior.  He also carried out a
    comparison with both elements of the sodium nitrite/thiosulfate
    treatment, again demonstrating that the cobalt compound was more
    effective.  Terzic & Milosevic (1963) compared sodium nitrite and
    dicobalt edetate.  They studied the effective doses of the two
    compounds, as well as their LD50 values, and concluded that cobalt
    edetate was superior.  Unfortunately their study was conducted in
    mice:  these animals especially, but also other small laboratory
    animals, are extremely insensitive to methaemoglobin-producing
    chemicals by virtue of the very high activity of NADH-linked
    methaemoglobin reductase in murine erythrocytes (Calabrese, 1983). 
    This difference renders extrapolation of these results to man almost
    impossible.  Two research groups have attempted to compare dicobalt

    edetate with 4-DMAP in dogs.  Klimmek et al. (1979a) found better
    survival with 4-DMAP when the antidotes were given 4 min after the
    poisoning but similar antidotal efficacy when they were given 1 min
    afterwards.  Marrs et al. (1985) found that dicobalt edetate
    successfully treated up to about three times the LD50 of cyanide,
    when given at the moment of apnoea, whilst 4-DMAP ensured survival
    at about six time the LD50 under the same conditions.

    5.8.2.3  Interactions with other drugs

         The only interactions that have been studied in animals are
    those with other cyanide antidotes.  The effect, in experimental
    animals, is generally additive.  This is despite the theoretical
    risk that cobalt toxicity might be exacerbated by other antidotes
    that prevent cyanide from binding to cobalt.  Antidotal combinations
    including dicobalt edetate have been studied by Evans (1964) and
    Frankenberg & Sörbo (1975).

        Table 4.  Studies of the efficacy of dicobalt edetate
              compared with other cyanide antidotes
                                                                        

    Comparison of dicobalt         Species     Author
    edetate with
                                                                        

    Sodium nitrite                   Dog       Paulet (1960 a,b)

    Sodium nitrite/thiosulfate       Dog       Paulet (1960 a,b)

    Sodium nitrite                  Mouse      Terzic & Milosevic (1963)

    4-DMAP                           Dog       Klimmek et al. (1979a)

    4-DMAP                           Dog       Marrs et al. (1985)

    Thiosulfate/rhodanese          Rabbit      Atkinson et al. (1974)

    Other cobalt antidotes      Mouse/Rabbit   Evans (1964)
                                                                        
    
    5.8.3  Toxicology

         Extensive toxicology was not carried out on dicobalt edetate
    before its introduction.  Published information is largely confined
    to acute toxicity testing, but it is believed that the
    organ-specific toxicity is similar to that of cobalt salts and that
    it is ameliorated by chelation (Paulet, 1960b).

    5.8.3.1  In vitro studies

         Dicobalt edetate gives negative results in the Ames test
    (Morris, 1987)

    5.8.3.2  Acute toxicity studies

         The acute toxicity of dicobalt edetate has been studied
    exclusively in small laboratory animals (Table 5).  An interesting
    feature is that the acute toxicity is decreased by the presence of
    glucose or of cyanide.  Thus, Paulet (1960a,b, 1961) observed that,
    in the mouse, the intravenous LD50 of dicobalt edetate (50 mg/kg)
    was increased threefold by giving it in hypertonic glucose.  The
    LD50 was increased to 82 mg/kg or 103 mg/kg, respectively, by the
    simultaneous administration of 2.5 or 5 mg NaCN/kg.

    5.8.3.3  Repeated dose toxicity

         Bartelheimer (1962b) gave intraperitoneal injections of cobalt
    edetate daily for 14 days at doses of 30, 40, or 50% of the LD50
    to rats, surviving animals being observed for 4 weeks.  One of six
    and two of six of the high and middle groups, respectively, died,
    whilst none of the low-dose group did so.  The majority of the rats
    in the two higher groups suffered diarrhoea and weight loss,
    although some recovery was seen in the latter half of the exposure
    period.  Autopsy and histological examination of the dead animals
    showed necrosis of the intestinal mucous membrane.  No pathological
    changes were seen in the liver, kidney, or heart.  Late effects were
    not observed in the survivors.

        Table 5.  The acute toxicity of dicobalt edetate

                                                                        

      Species     Route     LD50 (mg/kg)    Reference
                                                                        

       Mouse       iv           50          Paulet (1961)

       Mouse       iv           71          Evans (1964)

       Mouse       ip          213          Paulet (1961)

       Mouse       ip          225          Terzic & Milosevic (1963)

        Rat        iv           43          Tauberger & Klimmer (1963)

        Rat        ip          100          Bartelheimer (1962b)
                                                                        
    
    5.8.3.4  Circulatory effects in dogs

         Klimmek et al. (1979b), during their studies comparing dicobalt
    edetate with 4-dimethylaminophenol (4-DMAP), found that the former
    caused circulatory depression and hyperventilation.  This was
    accompanied by metabolic acidosis.

    5.8.3.5  Other toxicity studies

         No information is available on reproductive toxicology or
    carcinogenicity.  No mammalian cell or in vivo mutagenicity test has
    been published.

    5.9  Volunteer Studies

         No human volunteer studies have been carried out.

    5.10  Clinical Trials

         Clinical trials have not been undertaken.

    5.11  Clinical Studies - Case Reports

    5.11.1  Successful use

         There are a number of case reports of the successful use of
    dicobalt edetate against different types of cyanide poisoning,
    including inhalation of HCN (e.g. Bain & Knowles, 1967; Nagler et
    al., 1978) and sodium cyanide (Hillman et al., 1974), and potassium
    cyanide ingestion (Hoang The Dan et al., 1981; Klaui et al., 1984). 
    The use of dicobalt edetate in poisoning with fused sodium cyanide
    has also been reported (Bourrelier & Paulet, 1971).  The case
    reported by Bain & Knowles involved a patient who was semicomatose
    but who became fully conscious immediately after the first of two
    doses of 10 ml of Kelocyanor: the blood cyanide concentration just
    before the antidote injection was found to be 5.1 mg/l.  In the
    report by Nagler et al. (1978), three workers were poisoned as a
    result of the accidental addition of sodium cyanide to an acid bath. 
    In each case the patients had definite evidence of cyanide
    poisoning, yet recovered.  A blood cyanide level of 5.25 mg/l was
    found in the patient reported by Hoang The Dan et al. (1981), yet
    the subject recovered without sequelae.

    5.11.2  Use in pregnant women and children

         No data exist on the use of dicobalt edetate in pregnant women
    or children.

    5.11.3  Adverse effects

         A number of adverse effects have been reported, generally
    following the inappropriate administration of dicobalt edetate.  It
    has been proposed that their occurrence can be minimized by using
    strict clinical criteria for giving the antidote (Bryson, 1978,1987;
    Tyrer, 1981; Aw & Bishop, 1981; Peden et al., 1986).  Nevertheless,
    it has been suggested that adverse reactions to dicobalt edetate may
    not be confined to instances of inappropriate administration (Dodds
    & McKnight, 1985).  Typical case reports of adverse reactions
    include that of Froneman (1975), who reported a massive urticarial
    reaction affecting the face, eyelids, and lips; it was questioned if
    the subject had, in fact, consumed any cyanide.  Four adverse
    reactions, where the diagnosis of cyanide poisoning was
    questionable, were reported by Tyrer (1981).  Three of the patients
    had a variety of symptoms and signs, including convulsions, oedema
    of the face and neck, urticaria, chest pains, dyspnoea, and
    hypotension.  Hypotension was also reported by Daunderer et al.
    (1974) (see section 5.11.4).  A patient described by McKiernan
    (1980) had a normal blood cyanide level, in spite of severe cyanide
    burns.  He reacted to Kelocyanor by developing facial and laryngeal
    oedema.  A case reported by Yacoub et al. (1974) was complicated by
    the use of other antidotes;  this patient developed urticaria.

    5.11.4  Use in combination with other antidotes

         Reference has been made above (section 5.8.3) to animal studies
    that suggest that the use of dicobalt edetate in combination with
    other antidotes could be beneficial.  Thus, Jeretin (1963) used it
    with sodium nitrite and thiosulfate, and Daunderer et al. (1974)
    successfully resuscitated with 4-DMAP and Kelocyanor a patient who
    had ingested about 10 g potassium cyanide.

    5.12  Summary of Evaluation

    5.12.1  Indications

         The only indication for the use of dicobalt edetate is acute
    cyanide poisoning.  It should not be used except where there is
    definite indication of poisoning.  Reactions are likely to occur if
    the drug is given inappropriately.

    5.12.2  Administration

         It is recommended that one ampoule should be given over one
    min.  A second or third may be given in the case of inadequate
    response.  Because glucose reduces the toxicity of dicobalt edetate
    (Paulet et al., 1960; Paulet, 1961), the recommendation is that the
    injections should be immediately followed by 50 ml dextrose
    (500 g/l).

    5.12.3  Other consequential or supportive therapy

         See section 1.10

    5.12.4  Contraindications

         Dicobalt edetate should not be used in suspected cyanide
    poisoning unaccompanied by signs of poisoning such as impairment or
    loss of consciousness.

    5.12.5  Comparison with other antidotes

         Dicobalt edetate probably crosses the blood-brain barrier,
    while methaemoglobin quite clearly does not.  Moreover, both DMAP
    and nitrite require follow-up with a sulfur donor;  with dicobalt
    edetate the cyanide is excreted with cobalt in the urine.  In these
    two respects dicobalt edetate seems to offer some advantages.  It is
    in clinical use that reservations have arisen, almost entirely
    because of adverse reactions.  Whereas occupational health
    practitioners seem to be particularly impressed with dicobalt
    edetate (Davison, 1969; Bryson, 1987), it is in hospital practice
    that problems are likely to occur.  The probable reason for this is
    that, at cyanide-using facilities, not only is the antidote to hand,
    but also occupational health personnel are familiar with the
    presentation of cyanide poisoning and the hazards of inappropriate
    use of antidotes.  A particular problem of this form of poisoning is
    that insubstantial or suspected cyanide poisoning may produce panic,
    which is mistaken for incipient intoxication.  It is then that the
    antidote is given (Anon, 1977).  Provided it is given only when the
    patient is unconscious many of the problems can be avoided.  While
    in the past it was suggested that in casualty departments Kelocyanor
    should be given to all patients with symptoms (Anon, 1977), this
    policy is no longer advocated.  Were the use of dicobalt edetate to
    be confined to cases of unequivocal poisoning with cyanide, it seems
    likely that adverse reactions would be far less frequent.

    5.13  Model Information Sheet

    5.13.1  Uses

         Dicobalt edetate is a specific antidote for use in acute
    cyanide poisoning.

    5.13.2  Dosage and route

         It should be given in the form of ampoules of Kelocyanor, which
    also contain glucose.  One ampoule (300 mg dicobalt edetate) should
    be given intravenously over 1 min.  If the response is inadequate, a
    second ampoule may be given.  The use of Kelocyanor should be
    followed immediately by 50 ml Dextrose IV infusion (500 g/l).

    5.13.3  Precautions/contraindications

         Kelocyanor should not be administered in the absence of
    cyanide.  Fully conscious patients, especially those exposed to HCN
    by inhalation, are unlikely to require Kelocyanor.

    5.13.4  Adverse effects

         These include facial, laryngeal, and neck oedema, chest pain,
    vomiting, and rashes.

    5.13.5  Use in pregnancy and lactation

         There is no experience of use in this situation.  In
    life-saving situations the recommended dosage should not be
    modified.

    5.13.6  Storage

         This antidote should be stored at a temperature below 25 °C. 
    The shelf-life is 3 years.

    5.14  References

    Anon (1977) Which antidote for cyanide? Lancet, 2: 1167.

    Atkinson A, Rutter DA, & Sargeant K (1974) Enzyme antidote for
    experimental cyanide poisoning. Lancet, 2: 1446.

    Aw TC & Bishop CM (1981) Cyanide poisoning. J Soc Occup Med, 31:
    173-175.

    Bain JTB & Knowles EL (1967) Successful treatment of cyanide
    poisoning. Br Med J, 2: 763.

    Ballantyne B & Marrs TC (1987) Postmortem features and criteria for
    the diagnosis of acute cyanide poisoning. In: Ballantyne B & Marrs
    TC ed. Clinical and experimental toxicology of cyanides. Bristol,
    United Kingdom, John Wright, pp 217-247.

    Bartelheimer EW (1962a) [Analysis of acute cobalt poisoning in
    experimental animals.] Naunyn-Schmiedebergs Arch Exp Pathol
    Pharmakol, 243: 237-253 (in German).

    Bartelheimer EW (1962b) [Differences between the toxic and
    cyanide-antagonistic efficacy of cobalt chelate compounds
    (Co-histidine and CO2 - EDTA.] Naunyn-Schmiedebergs Arch Exp
    Pathol Pharmakol, 243: 254-268 (in German).

    British Pharmacopoeia (1980) London, Her Majesty's Stationery
    Office, vol. 2, p 579.

    Bourrelier J & Paulet G (1971) Intoxication cyanhydrique consécutive
    à des brûlures graves par cyanure de sodium fondu. Presse Méd, 79:
    1013-1014.

    Bryson D (1978) Cyanide poisoning. Lancet, 1: 92.

    Bryson D (1987) Occupational cyanide poisoning. In: Ballantyne B &
    Marrs TC ed. The clinical and experimental toxicology of cyanides.
    Bristol, United Kingdom, John Wright, pp 348-358.

    Calabrese EJ (1983) Principles of animal extrapolation. New York,
    John Wiley & Sons.

    Christian GD (1969) Medicine, trace elements, and atomic absorption
    spectroscopy. Anal Chem, 41: 24A.

    Cotton FA & Wilkinson G (1966) Advanced inorganic chemistry, 2nd ed.
    New York, London, Interscience Publishers, p 869.

    Daunderer M, Theml H, & Weger N (1974) [Treatment of prussic acid
    poisoning with 4-dimethylaminophenol (4-DMAP).] Med Klin, 69:
    1626-1631 (in German).

    Davison V (1969) Cyanide poisoning. Kelocyanor - a new treatment.
    Occup Health, 21: 306-308.

    Dodds C & McKnight C (1985) Cyanide toxicity after immersion and the
    hazards of dicobalt edetate. Br Med J, 291: 785-786.

    European Pharmacopoeia (1980) 2nd ed. Sainte Ruffine, France,
    Maisonneuve SA.

    Evans CL (1964) Cobalt compounds as antidotes for hydrocyanic acid.
    Br J Pharmacol, 23: 455-475.

    Eybl V, Sykora J, & Kocher Z (1959) [The influence of EDTA on the
    effects of cobalt  in vivo.] Naunyn-Schmiedebergs Arch Exp Pathol
    Pharmakol, 236: 199-201 (in German).

    Feldstein MA & Klendshoj WJ (1954) The determination of cyanide in
    biologic fluids by microdiffusion analysis. J Lab Chem Med, 44:
    166-170.

    Frankenberg L & Sörbo B (1975) Effect of cyanide antidotes on the
    metabolic conversion of cyanide to thiocyanate. Arch Toxicol, 33:
    81-89.

    Froneman JPC (1975) A first experience in the use of Kelocyanor.
    Proc Mine Off Assoc (Johannesburg), 54: 43.

    Hagler L & Coppes RI (1982) Inhibition of methemoglobin and
    metmyoglobin reduction by cobalt. Biochem Pharmacol, 31: 1779-1782.

    Hillman B, Bardhan KD, & Bain JTB (1974) The use of dicobalt edetate
    (Kelocyanor) in cyanide poisoning. Postgrad Med J, 50: 171-174.

    Hoang The Dan P, Bretsztajn A, Pourriat JL, Lapandry C, & Cupa M
    (1981) Un cas d'intoxication aiguë au cyanure. Anesth Anal Réanim,
    38: 161-162.

    Jeretin S (1963) [A case of successful recovery of a patient
    suffering from potassium cyanide poisoning.] Zdrav Vestn, 32: 3-4
    (in Slovene).

    Klaui H, Russi E, & Bauman PC (1984) [Cyanide poisoning.] Schweiz
    Med Wochenschr, 114: 983-989 (in German).

    Klimmek R, Fladerer H, & Weger N (1979a) Circulation, respiration
    and blood homeostasis in cyanide-poisoned dogs after treatment with
    4-dimethylaminophenol or cobalt compounds. Arch Toxicol, 43:
    121-133.

    Klimmek R, Fladerer H, Szinicz L, Weger N, & Kiese M (1979b) Effects
    of 4-dimethylaminophenol and CO2 EDTA on circulation, respiration,
    and blood homeostasis in dogs. Arch Toxicol, 42: 75-84.

    Lewis SA, O'Haver TC, & Harnley JM (1985) Determination of metals at
    the microgram-per-liter level in blood serum by simultaneous
    multielement atomic absorption spectrometry with graphite furnace
    atomization. Anal Chem, 57: 2-5.

    McKiernan MJC (1980) Emergency treatment of cyanide poisoning.
    Lancet, 2: 1980.

    Marrs TC, Swanston DW, & Bright JE (1985) 4-Dimethylaminophenol and
    dicobalt edetate (Kelocyanor) in the treatment of experimental
    cyanide poisoning. Hum Toxicol, 4: 591-600.

    Mercker H & Bastian G (1959) [Cobalt compounds for prussic acid
    detoxification.]  Naunyn-Schmiedebergs Arch Exp Pathol Pharmakol,
    236: 449-458 (in German).

    Morris B (1987) Postmortem features and criteria for the diagnosis
    of acute lethal cyanide poisoning In: Ballantyne B & Marrs TC ed.
    Clinical and experimental toxicology of cyanides. Bristol, United
    Kingdom, John Wright, pp 217-247.

    Mulford CE (1966) Solvent extraction techniques for atomic
    absorption spectroscopy. At Absorpt Newsl, 5: 88-90.

    Mushett CW, Kelly KL, Boxer GE, & Rickards JC (1952) Antidotal
    efficacy of vitamin B12a (hydroxo-cobalamin) in experimental
    cyanide poisoning. Proc Soc Exp Biol Med (NY), 81: 234, 237.

    Nagler J, Provoost RA, & Parizel G (1978) Hydrogen cyanide
    poisoning: treatment with cobalt EDTA. J Occup Med, 20: 414-416.

    Paulet G (1957) Valeur des sels organiques du cobalt dans le
    traitement de l'intoxication cyanhydrique. C R Soc Biol (Paris),
    151: 1932-1935.

    Paulet G (1958) Intoxication cyanhydrique et chélates de cobalt. J
    Physiol (Paris), 50: 438-442.

    Paulet G (1960a) L'intoxication cyanhydrique et son traitement.
    Paris, Masson SA.

    Paulet G (1960b) Les chélates de cobalt dans le traitement de
    l'intoxication cyanhydrique. Pathol Biol, 8: 256-266.

    Paulet G (1961) Nouvelles perspectives dans le traitement de
    l'intoxication cyanhydrique. Arch Mal Prof, 22: 102-127.

    Paulet G (1965) Au sujet du traitement de l'intoxication
    cyanhydrique par les chélates de cobalt. Urgence, 11: 611-613.

    Paulet G, Chary R, Bocquet P, & Fouilhoux M (1960) Valeur comparée
    du nitrite de sodium et des chélates de cobalt dans le traitement de
    l'intoxication cyanhydrique chez l'animal (chien-lapin)
    non-anesthésié. Arch Int Pharmacodyn, 127: 104-117.

    Peden NR, Taha A, McSorley PD, Bryden GT, Murdoch IB, & Anderson JM
    (1986) Industrial exposure to cyanide: implications for treatment.
    Br Med J, 293: 538.

    Sharpe AT (1976) The chemistry of cyano complexes of the transition
    metals. New York, London, Academic Press.

    Speijers GJA, Krajnc EI, Berkvens JM, & Van Logten MJ (1982) Acute
    oral toxicity of inorganic cobalt compounds in rats. Food Chem
    Toxicol, 20: 311-314.

    Tauberger G & Klimmer OR (1963) [Animal experimental studies of some
    cobalt compounds after intravenous injection.] Arch Int Pharmacodyn,
    143: 219-239 (in German).

    Terzic M & Milosevic M (1963) Action protectrice de
    l'éthylène-diamine-tétra-acétate-dicobaltique dans l'intoxication
    cyanée. Thérapie, 18: 55-61.

    Thiers RE, Williams JF, & Yoe JH (1955) Separation and determination
    of millimicrogram amounts of cobalt. Anal Chem, 27: 1725-1731.

    Tyrer FH (1981) Treatment of cyanide poisoning. J Soc Occup Med, 31:
    65-66.

    United States National Formulary (1985) 16th ed. Rockville,
    Maryland, US Pharmacopeial Convention Inc., p 1559.

    Yacoub M, Faure J, Morena H, Vincent M, & Gaure H (1974)
    L'intoxication cyanhydrique aiguë. Données actuelles sur le
    métabolisme du cyanure et le traitement par hydroxocobalamine. J Eur
    Toxicol, 7: 22-29.

    6.  AMYL NITRITE

    6.1  Introduction

         Amyl nitrite by inhalation has been used for many years as a
    simple first-aid measure for cyanide poisoning.  It generates
    methaemoglobin and may be administered by lay personnel. 
    Methaemoglobin combines with cyanide to form non-toxic
    cyanmethaemoglobin.  However, amyl nitrite is not a powerful
    methaemoglobin-forming agent in humans and therefore it has
    generally been used in combination with intravenous sodium nitrite. 
    Of 44 patients poisoned with cyanide, and of which 43 recovered
    [Chen et al., 1944 (15 patients); Wolfsie, 1951 (12 patients); 
    Miller & Toops, 1951 (1 patient); and Chen & Rose, 1952 (16
    patients)], only 7 patients were treated with amyl nitrite alone; 2
    were treated with amyl nitrite in combination with sodium
    thiosulfate.

         Recently there has been a resurgence of interest in amyl
    nitrite.  Artificial respiration with amyl nitrite broken into an
    Ambu bag was reported by Vick & Froelich (1985) as being life-saving
    in dogs severely poisoned with cyanide before significant formation
    of methaemoglobin occurred (Vick & Froelich, 1985).  The vasogenic
    effects of amyl nitrite may therefore play a role in its antidotal
    effect in cyanide poisoning.

    6.2  Name and Chemical Formula

         Chemical name:  Amyl nitrite, Amylis nitris, Nitris amylicus

         Molecular formula:  C5H11NO2

         Structural formula:

         Amyl nitrite is a mixture of nitrite esters of 
    2-methylbutan-l-ol and 3-methylbutan-l-ol.

                           ONO                 ONO
                           |                   |
                           CH2                 CH2
    nitrate ester of       |                   |      nitrate ester of
    2-methylbutan-l-ol    HCCH3                CH2    3-methylbutan-l-ol
                           |                   |
                           CH2                HCCH3
                           |                   |
                           CH3                 CH3

         It contains not less than 85% and not more than 103% of
    C5H11NO2 (Martindale, 1989; United States Pharmacopeia, 1985).

         Relative molecular mass:    117.15

         Appearance:                 yellow, transparent fluid

         CAS numbers:                8017-89-8 and 110-46-3

         IUPAC name:                     -

    6.3  Physio-chemical Properties

         Boiling point:              96 °C
         Solubility:                 practically insoluble in water;
                                     miscible with alcohol, chloroform,
                                     ether, and light petroleum
         Boiling point:              no data available
         Acidity:                    no data available
         pK:  no data available
         Stability in light:         protect from light
         Thermal stability:          it is volatile even at low
                                     temperatures and is liable to
                                     decompose with the evolution of
                                     nitrogen, particularly if it has
                                     become acidic.  Flash point 10 °C
                                     (closed-cup test)
         Specific gravity:           0.8970-0.880
         Refractive index:           1.386-1.389 (Nederlandse
                                     Pharmacopee, 1958)
         Explosive properties:       amyl nitrite forms an explosive
                                     mixture with air and oxygen.  It is
                                     very inflammable and must not be
                                     used where it may be ignited.

    6.4  Synthesis

         Amyl nitrite is generally manufactured by the action of sodium
    nitrate and sulfuric acid on the appropriate alcohols.

    6.5  Analytical Methods

    6.5.1  Identification

         a)   To a mixture of 2 drops of amyl nitrite and 2 drops of
              water, add 2 ml of sulfuric acid and dilute with water. 
              The odour of amyl valerate should become apparent.

         b)   To a few drops of amyl nitrite, add a mixture of 1 ml of
              hydrochloric acid (3 mol/l).  A greenish-brown colour
              should be produced (United States Pharmacopeia, 1985).

    6.5.2  Purity

    6.5.2.1  Acidity

         To 0.30 ml in a glass-stoppered cylinder, add a mixture of
    0.60 ml of sodium hydroxide (0.1 mol/l), 10 ml of water, and 1 drop
    of phenolphthalein TS, and invert the cylinder three times.  The red
    tint of the water layer should remain perceptible (United States
    Pharmacopeia, 1985).

    6.5.2.2  Non-volatile residue

         Allow 10 ml to evaporate at room temperature in a weighed
    evaporating dish within a well-ventilated hood, and dry the residue
    at 105 °C for 1 h.  The weight of the residue should not exceed 2 mg
    (0.02%) (United States Pharmacopeia, 1985).

    6.5.2.3  Assay for total nitrites

         Inject an aliquot of amyl nitrite of suitable volume, but not
    more than 2 ml, into a suitable gas chromatograph equipped with a
    thermal conductivity detector.  Under typical conditions, the
    instrument should contain a column (2 m x 3 mm) packed with a methyl
    polysiloxane oil, 25% by weight on suitable calcine diatomite.  The
    column should be maintained at 80 °C, the injection port and
    detector block maintained at 10 °C above the temperature of the
    column, and helium used as a carrier gas at a flow rate of about
    60 ml per min.  The percentage of total nitrites is calculated from
    the area under the curve; a total of not less than 97% should be
    found.

    6.6  Shelf-life

         The following precautions should be taken.

              *    Protect from light.

              *    Insist that the suppliers only provide ampoules that
                   have at least 18 month shelf-life left of the nominal
                   24-month total life.

              *    Take reasonable precautions to store at the lowest
                   possible temperature compatible with immediate access
                   in case of emergency; refrigerated storage is not
                   necessary.

              *    Inspect monthly for any broken tubes  (Beasley et
                   al., 1978).

         As indicated in section 6.3, amyl nitrite forms an explosive 
    mixture with air and oxygen.  It is highly inflammable and must not
    be used where it may be ignited.  Storage for a prolonged period or
    at an excessive temperature presents the following hazards:

              *    breakage of the capsule by the build up of pressure
                   internally and loss of amyl nitrite before use;

              *    explosive distribution of glass splinters when the
                   ampoule is broken for use;

              *    loss of amyl nitrite by chemical decomposition.

    6.7  General Properties

         The principal pharmacological action of amyl nitrite is to
    cause the relaxation of vascular smooth muscle.  Peripheral venous
    resistance is decreased as a result of a selective action on venous
    capacitance vessels with resultant venous pooling of blood and
    decreased venous return to the heart.  The vasodilating effect of
    amyl nitrite on arteriolar resistance is less than that on the
    venous side.  As a result of this combined action, both venous
    filling pressure (preload) and, to a lesser extent, arterial
    impedance (afterload) are reduced (AHFS, 1988).

         It has been postulated that cyanide causes pulmonary and/or
    coronary arteriolar vasoconstriction, which results directly or
    indirectly in pump failure and an observed decrease in cardiac
    output (Sarnoff, 1980a).  This theory is supported by the reported
    observations of Vick & Froelich (1985) that a sharp increase in
    central venous pressure occurs at a time when the arterial blood
    pressure is falling.  Following the administration of amyl nitrite
    by inhalation, central venous pressure decreased rapidly while the
    arterial blood pressure increased.  The apparent life-saving effect
    of this drug in dogs could have been due to a reversal of early
    cyanide-induced vasoconstriction and the restoration of normal
    cardiac function. Subsequently, amyl nitrite causes the formation of
    methaemoglobin, which may have an additional antidotal action.

                 
    a  Personal communication by S.J. Sarnoff, Survival Technology
         Inc., Bethesda, USA.

    6.8  Animal Studies

    6.8.1  Pharmacokinetics

         Amyl nitrite is readily absorbed into the circulation from
    mucous membranes, the rate of absorption being greatest from the
    lungs.  It is rapidly inactivated by hydrolysis to isoamyl alcohol
    and nitrite.  Amyl nitrite is rapidly hydrolysed in the
    gastrointestinal tract and is therefore pharmacologically inactive
    when administered orally (Martindale, 1989).


    6.8.2  Pharmacodynamics

         Amyl nitrite has proved to be effective in experimental animals
    poisoned by cyanide.  Indeed, antagonism between amyl nitrite and
    hydrocyanic acid was mentioned as long ago as l888 by Pedigo.

         Chen et al. (1933) demonstrated that nine to ten inhalations of
    amyl nitrite were effective in dogs against 4 times the lethal dose
    of cyanide (12-24 mg/kg) (sodium nitrite was also given
    subcutaneously).

         Amyl nitrite had a significant therapeutic effect on dogs
    exposed to cyanogen chloride (CNCl) at Ct values (the product of the
    concentration of the gas in mg/m3, and the length of exposure in
    min) ranging from 3300 to 6300 mg.min.m-3.  However, no such
    effect was observed in the groups subjected to higher exposures, at
    Ct values ranging from 6900 to 11800 mg.min.m-3 (Jandorf &
    Bodansky, 1946).

         Amyl nitrite treatment of mice exposed to CNCl resulted in a
    statistically significant increase in the number of survivors
    (Jandorf & Bodansky, 1946).

         Amyl nitrite administered by inhalation increased the median
    lethal subcutaneous dose of sodium cyanide in dogs from 5.36 (+0.28)
    to 24.5 (+1.2) mg/kg (Chen & Rose, 1952).

         Part of the antidotal effect may result from methaemoglobin
    formation.  Any attempt to correlate the therapeutic effect of amyl
    nitrite with the degree of methaemoglobinaemia encounters 
    difficulties.  After exposure to HCN or CNCl, the blood of treated
    animals contains, in addition to haemoglobin and oxyhaemoglobin,
    cyanmethaemoglobin and methaemoglobin.  To date, it has not been
    possible to quantify the amount of cyanmethaemoglobin.  However,
    experiments have been undertaken to estimate the degree of
    methaemoglobinaemia produced by inhalation of amyl nitrite in
    animals not previously exposed to cyanide.

         When ten mice were placed in a chamber containing approximately
    12 mg amyl nitrite per l, the methaemoglobin levels 10 seconds,
    30 seconds, and 1, 2, and 4 min after the beginning of the exposure,
    were 1.0, 1.8, 5.6, 14, and 30%, respectively (Jandorf & Bodansky,
    1946).

         A cone holding one crushed ampoule containing 0.3 ml amyl
    nitrite was fitted over the muzzle of each of four dogs
    anaesthetized with pentobarbital, and kept in place for 3 min. 
    Under these conditions, 16, 18, 20, and 32% methaemoglobinaemia were
    found in the four dogs at the end of amyl nitrite inhalation
    (Jandorf & Bodansky, 1946).

         Bastian & Mercker (1959) recorded a methaemoglobin level of 60%
    in mice placed in an atmosphere of amyl nitrite (0.112% v/v) for
    15 min.  Thereafter the degree of methaemoglobinaemia remained
    constant and it appeared that an equilibrium existed between the
    formation and reduction of methaemoglobin, the level depending on
    the concentration of amyl nitrite in the inspired air.

         In cats, inhalation of amyl nitrite (0.06 and 0.12% v/v)
    resulted in methaemoglobin levels of 30% and 70%,  respectively.  At
    the higher concentration of amyl nitrite, a fall in blood pressure
    from 10.7 to 2.7 kPa (80 to 20 mmHg) was observed (Bastian &
    Mercker, 1959).

         Part of the antidotal effect of amyl nitrite may be vasogenic
    in origin, rather than being due to methaemoglobin formation (Way et
    al., 1984).  Amyl nitrite given after cyanide administration was
    reported by Vick & Froelich (1985) to reverse both cardiovascular
    changes and respiratory paralysis in 24 of 30 dogs.  These changes
    occurred before significant methaemoglobin formation and suggest
    that early death caused by cyanide could be due in part to
    cardiovascular and respiratory failure.

    6.8.3  Toxicology

         Syncope or shock induced by large doses of amyl nitrite are the
    result of a pooling of blood in dilated capacitance vessels.  Reflex
    tachycardia normally occurs, but a vaso-vagal reflex may induce a
    transient bradycardia immediately before collapse (Wilkins et al.,
    1937).

         The degree of methaemoglobinaemia caused by inhalation of amyl
    nitrite is insufficient to affect oxygen transport (see section
    6.8.3).

         Amyl nitrite also causes a substantial increase in
    cerebrospinal fluid pressure (Norcross, 1938), and a fall in pO2
    in brain tissue has been recorded using an oxygen electrode
    (Rosemann et al., 1946).

    6.9  Volunteer Studies

         Inhalation of therapeutic doses of amyl nitrite in man did not
    result in methaemoglobin formation (Mathes & Gross, 1939).

         When six volunteers inhaled 0.1 ml amyl nitrite 10 times for
    20 seconds once per min, the median concentration of methaemoglobin
    was 3.45%, with a maximum of 6% and a minimum of 1.4%.  During and
    shortly after inhalation, a small fall in blood pressure was
    observed, followed by immediate recovery.  The pulse rate rose from
    40 to 120 beats per min.  Repeated use of the same square of gauze
    without application of a new ampoule of amyl nitrite did not have
    any effect (Bastian & Mercker, 1959).

    6.10  Clinical Studies

         No controlled human clinical trials have been undertaken.

    6.11  Clinical Studies - Case Reports

         No case reports are available.

    6.12  Summary of Evaluation

    6.12.1  Indications

         Amyl nitrite is used as a first-aid measure in cases of cyanide
    poisoning where the patients are in deep coma, with dilated
    non-reactive pupils and deteriorating cardio-respiratory function.

    6.12.2  Advised routes and dose

         In adults, administer 0.2-0.4 ml amyl nitrite, via an Ambu bag,
    prior to artificial ventilation.  In children, administer a maximum
    of 0.1 ml, via an Ambu bag, prior to artificial ventilation.

    6.12.3  Other consequential or supportive therapy

         First-aid therapy with amyl nitrite must be followed by
    additional antidotal treatment (see chapter 1).  However, supportive
    therapy is the most important aspect of the treatment of cyanide
    poisoning.  Special attention should be paid to the circulation. 
    The vasodilating action of amyl nitrite may lead to hypotension,
    which should be treated immediately with plasma expanders.

    6.13  Model Information Sheet

    6.13.1  Uses

         Amyl nitrite is used as a first-aid measure in cases of cyanide
    poisoning where patients are in deep coma, with dilated non-reactive
    pupils and deteriorating cardio-respiratory function.

    6.13.2  Dosage and route

         In adults, 0.2-0.4 ml amyl nitrite should be administered via
    an Ambu bag prior to artificial respiration.

         In children, a maximum of 0.l ml should be administered via an
    Ambu bag prior to artificial respiration.

    6.13.3  Precautions/contraindications

         Special attention should be paid to the circulation.  The
    vasodilating action of amyl nitrite may lead to hypotension, which
    should be treated immediately with plasma expanders.

    6.13.4  Storage

              *    Protect from light.

              *    Insist that the suppliers provide only ampoules that
                   have at least 18 months shelf-life left of the
                   nominal 24-months total life.

              *    Take reasonable precautions to store at the lowest
                   possible temperature compatible with immediate access
                   in case of emergency; refrigerated storage is not
                   necessary.

              *    Inspect monthly for any broken tubes.

         Amyl nitrite forms an explosive mixture with air and oxygen. 
    It is highly inflammable and must not be used where it may be
    ignited.

    6.14  References

    American Hospital Formulary Service (AHFS) (1988)  In: McEnvoy KG,
    Litvak K, Welsh OH, Campbell JF, Ziegler K, Douglas PM, Shannon-Lass
    EP, Thomson K, McQuerrie GM, Schmadel LK, Fredrickson MK, Mendham
    NA, & Morisseau AL ed. Drug Information, Bethesda, American Society
    of Hospital Pharmacists.

    Bastian G & Mercker H (1959) [The efficacy of amyl nitrite
    inhalation in the treatment of cyanide poisoning.]
    Naunyn-Schmiedebergs Arch Exp Pathol Pharmakol, 237: 285-295 (in
    German).

    Beasley RWR, Blown RJ, Lunau FW, & Taylor BM (1978) Amyl nitrite and
    all that.  J Soc Occup Med, 28: 142-143.

    Chen KK & Rose CL (1952) Nitrite and thiosulfate therapy in cyanide
    poisoning. J Am Med Assoc, 149: 113-119.

    Chen KK, Rose CL, & Clowes GHA (1933) Amyl nitrite and cyanide
    poisoning. J Am Med Assoc, 100: 1920-1922.

    Chen KK, Rose CL, & Clowes GHA (1944) The modern treatment of
    cyanide poisoning.  J Ind Hyg Toxicol, 37: 344-350.

    Jandorf BJ & Bodansky O (1946) Therapeutic and prophylactic effect
    of methemoglobinemia in inhalation poisoning by hydrogen cyanide and
    cyanogen chloride. J Ind Hyg Toxicol, 28: 124-132.

    Martindale (1989) In: Reynolds JEF ed. The extra pharmacopoeia, 29th
    ed. London, The Pharmaceutical Press, pp 1492-1493.

    Mathes K & Gross F (1939) [The determination of methaemoglobin and
    cyanmethaemoglobin in circulating blood.] Naunyn-Schmiedebergs Arch
    Exp Pathol Pharmakol, 191: 701 (in German).

    Miller MH & Toops TC (1951) Acute cyanide poisoning. Recovery with
    sodium thiosulfate therapy. J Indiana Med Assoc, 44: 1164.

    Nederlandse Pharmacopee (1958) 6th ed. The Hague, Staatsdrukkerij.

    Norcross NC (1938) Intracerebral blood flow. Arch Neurol Psychiatry
    (Chicago), 40: 291.

    Pedigo LG (1888) Antagonism between amyl nitrite and Prussic acid.
    Trans Med Soc Virginia, 19: 124-131.

    Rosemann E, Goodman CW, & McCulloch WS (1946) Rapid changes in
    cerebral oxygen tension induced by altering the oxygenation and
    circulation of the blood. J Neurophysiol, 9: 33.

    United States Pharmacopeia (1985) 21st ed. Rockville, Maryland,
    United States Pharmacopeial Convention, Inc.

    Vick JA & Froelich HL (1985) Studies on cyanide poisoning. Arch Int
    Pharmacodyn, 273: 314-322.

    Way JL, Sylvester D, Morgan RL, Isom GE, & Burrows CB (1984) Recent
    perspectives on the toxidynamic basis of cyanide antagonism. Fundam
    Appl Toxicol, 4: S231-S239.

    Wilkins RW, Haynes FW, & Weiss S (1937) Role of venous system in
    circulatory collapse induced by sodium nitrite. J Clin Invest, 16:
    85-91.

    Wolfsie JH (1951) Treatment of cyanide poisoning in industry. Am Med
    Assoc Arch Ind Hyg, 4: 417-425.

    7.  SODIUM NITRITE

    7.1  Introduction

         Methaemoglobin has a great affinity for cyanide  in vitro (Hug
    & Marenzi, 1933a).  Sodium nitrite is an inducer of methaemoglobin
    (Hug, 1933; Hug & Marenzi, 1933b) and has been in clinical use for
    over 50 years as an antidote for acute cyanide poisoning, most often
    in combination with amyl nitrite and sodium thiosulfate (Viana et
    al., 1934; Chen & Rose, 1952; Hall & Rumack, 1986).

         Potential side-effects of sodium nitrite infusion are
    hypotension and excessive methaemoglobin formation (Viana et al.,
    1934; Berlin, 1970; Feihl et al., 1982; Hall & Rumack, 1986). 
    Patients who are already hypotensive are at special risk, and those
    with certain types of congenital methaemoglobinaemia are
    theoretically more at risk from excessive methaemoglobin induction. 
    Children administered large doses of sodium nitrite have developed
    excessive methaemoglobinaemia (Berlin, 1970).  Victims of smoke
    inhalation may have both elevated carboxyhaemoglobin levels and
    cyanide poisoning (Hart et al., 1985), and may develop further
    hypoxia from methaemoglobin induction (Becker, 1985; Jones et al.,
    1987).

    7.2  Name and Chemical Formula

         Sodium nitrite, otherwise known as diazotizing salts, dusitan
    sodny (Czech), erinitrit, filmerine, natrium nitrit (German),
    nitrite de sodium (French), nitrous acid sodium salt, sodii nitris,
    natrii nitris, natrium nitrosum, sodium nitrite (DOT), NCI-C02084,
    NIOSH No. RA 1225000, CAS 7632-00-0 (Sax, 1984; NIOSH, 1983;
    Windholz et al., 1983; Martindale, 1982), is supplied in the USA by
    Eli Lilly & Co., Indianapolis, Indiana, as a component of the Lilly
    Cyanide Antidote Package [R] (Product Information, 1984).  It is
    sold in Australia by Orapharm under the proprietary name of OAR
    (Martindale, 1989).  The relative molecular mass of sodium nitrite
    is 69.00, and the chemical formula is NaNO2 (Sax, 1984).

    7.3  Physico-chemical Properties

         Sodium nitrite has a melting point of 271 °C (Windholz et al.,
    1983).  It is soluble in either 1.5 parts of cold water or 0.6 parts
    of boiling water (Windholz et al., 1983).  Solutions may be made in
    water (1:1.5) or alcohol 1:160 (Martindale, 1989).  Sodium nitrite
    is slightly soluble in ether (ITI, 1985).  Acids will decompose
    sodium nitrite and cause the evolution of brownish N2O3 fumes
    (Windholz et al., 1983).  Aqueous solutions are alkaline to litmus,
    with a pH between 7 and 9 (Windholz et al., 1983; United States
    Pharmacopeia, 1985).  The specific gravity is 2.17 (ITI, 1985).  A
    10% solution of sodium nitrite has a density of 1.065 (Izmerov,
    1982).  Excipients are apparently not added.

         On drying over silica gel for 4 h, solid sodium nitrite loses
    no more than 0.25% of its weight (United States Pharmacopeia, 1985).

         Injectable sodium nitrite solution is made by diluting solid
    sodium nitrite with "Water for Injection" (United States
    Pharmacopeia, 1985).  The final solution is sterile and contains not
    less than 95% and not more than 105% of the labelled amount of
    sodium nitrite (United States Pharmacopeia, 1985).  Sodium nitrite
    is supplied in the Lilly Cyanide Antidote Package [R] in two 10 ml
    vials of a 3% solution (30 mg/ml; 300 mg per vial) (Product
    Information, 1984).

         Sodium nitrite is stable in light, but slowly oxidizes in air
    to nitrate (Windholz et al., 1983).  It is incompatible with
    acetanilide, antipyrine, caffeine, citrate, chlorates,
    hypophosphites, iodides, mercury salts, morphine, oxidizing agents,
    permanganate, phenazone, sulfites, tannic acid, and vegetable
    astringent decoctions, infusions, or tinctures (Martindale, 1989;
    Windholz et al., 1983).

    7.4  Synthesis

         Commercial sodium nitrite is derived from sodium hydroxide and
    nitrogen oxides by one of three common methods (Izmerov, 1982).

         One method uses a low concentration of nitrogen oxides from gas
    formed during dilute nitric acid production, which is absorbed by a
    sodium hydroxide solution.

         A second method (the "hot method") uses sodium hydroxide
    absorption of high concentrations of nitrogen oxides at atmospheric
    or higher pressure.

         A third method (or "two-staged method") combines features of
    the first two.

         Pharmaceutical sodium nitrite is produced by reduction of
    sodium nitrate, previously heated until fused, with sulfur dioxide,
    lead, or a sulfite (Harvey, 1980).  Alternatively, it can be derived
    from sodium nitrate by nitric oxide absorption (the nitric oxide
    being obtained from ammonia by catalytic oxidation in a sodium
    carbonate solution) (Harvey, 1980).  The sodium nitrite mixture
    obtained is lixiviated with water and filtered, then partially
    evaporated and allowed to crystallize (Surgenor, 1970).

         Injectable sodium nitrite solution is prepared by dissolving
    solid sodium nitrite in "Water for Injection" to a final
    concentration of 3% (30 mg/ml) (Product Information, 1984; United
    States Pharmacopeia, 1985).  It is then sterilized by autoclaving or
    membrane filtration (Martindale, 1989).  The final delivery form

    meets both United States and European Pharmacopoeia requirements for
    pyrogens and sterility.

         The United States Pharmacopeia specifies that no more than
    0.002% of heavy metals may be present (United States Pharmacopeia,
    1985).  "Top grade" sodium nitrite from some processes has a sodium
    nitrite concentration of at least 99%, with at most 1.4% moisture
    content and impurities of not more than 0.8% sodium nitrate, 0.3%
    calcinated residue, and 0.17% sodium chloride (Izmerov, 1982).

    7.5  Analytical Methods

    7.5.1  Quality control

    7.5.1.1  Solid sodium nitrite (United States Pharmacopeia, 1985)

         Sodium nitrite (1g) is dissolved in water, making up to 100 ml. 
    Then 10 ml of this solution is pipetted into a mixture containing
    50.0 ml of 0.1 N potassium permanganate volumetric solution (VS)
    (0.1 mol/l), 100 ml of water, and 5 ml of sulfuric acid, and the
    pipette tip is immersed beneath the surface of this mixture during
    transfer.  The liquid is warmed to 40 °C and allowed to stand for
    5 min.  After the addition of 25.0 ml oxalic acid VS (0.05 mol/l),
    the resultant mixture is heated to 80 °C and titrated with potassium
    permanganate (0.1 mol/l), equivalent to 3.45 mg sodium nitrite.

    7.5.1.2  Sodium nitrite injection (United States Pharmacopeia, 1985)

         The method described in 7.5.1.1 is followed except that, in
    place of the 10 ml sodium nitrite solution, an accurately measured
    volume of sodium nitrite injection containing about 150 mg sodium
    nitrite is used.

    7.5.1.3  Preparation of volumetric solutions (VS) (United States
             Pharmacopeia, 1985)

         (a)  Potassium permanganate VS (0.1 mol/l)

              KMnO4; relative molecular mass, 158.03; 3.161 grams in
              1000 ml

              About 3.3 g of potassium permanganate is dissolved in
              1000 ml of water and boiled for about 15 min in a suitable
              flask.  The flask is then stoppered and allowed to stand
              for at least 2 days.  It is then filtered through a
              fine-porosity, sintered glass crucible, which may be lined
              with a pledget of glass wool.  The solution is then
              standardized by weighing accurately 200 mg of sodium
              oxalate, previously dried to constant weight at 110 °C. 
              This is dissolved in 250 ml of water.  Sulfuric acid
              (7 ml) is added and the mixture is heated to about 70 °C. 
              The potassium permanganate solution is slowly added from a
              burette with constant stirring until a pale-pink colour
              persisting for about 15 seconds is produced.  At the
              termination of titration, the temperature should be no
              lower than 60 °C.  The molarity is calculated, with each
              6.700 mg of sodium oxalate equivalent to 1 ml of potassium
              permanganate (0.1 mol/l).  The solution should be stored
              in glass-stoppered amber-coloured bottles, and contact
              with organic materials such as rubber, which will reduce
              the potassium permanganate, should be avoided.

         (b)  Oxalic Acid VS (0.05 mol/l)

              H2C2O4 . 2H2O; relative molecular mass 126.07;
              6.303 g in 1000 ml

              6.45 g of oxalic acid is dissolved in water and made up to
              1000 ml.  The solution is standardized by titration
              against potassium permanganate VS (1 mol/l) as described
              above.  The solution should be stored protected from light
              in glass-stoppered bottles.

    7.5.2  Identification (United States Pharmacopeia, 1985)

         A solution of sodium nitrite responds to tests for sodium and
    for nitrite.

    7.5.2.1  Sodium

         After conversion to chloride or nitrate, solutions of sodium
    compounds with five times their volume of cobalt-uranyl acetate test
    solution (TS) yield a precipitate that is golden yellow in colour
    and forms after several min of agitation.  Cobalt-uranyl acetate
    test solution is prepared by dissolving 40 g of uranyl acetate in a
    mixture of 30 g of glacial acetic acid and enough water to make
    500 ml.  A solution is also prepared of 200 g of cobaltous acetate
    in a mixture of 30 g of glacial acetic acid and enough water to make
    500 ml.  Both solutions are prepared with warming, mixed while still
    warm, and cooled to 20 °C.  The temperature is maintained at 20 °C
    for about 2 h, allowing the separation of excess salts, and then the
    resultant solution is filtered through a dry filter.  Sodium

    compounds also impart an intense yellow coloration to a non-luminous
    flame.

    7.5.2.2  Nitrite

         Treatment of nitrite-containing solutions with diluted mineral
    acids or acetic acid (6 mol/l) produces brownish-red fumes, and the
    solution will give a blue coloration on starch-iodide paper.

    7.5.3  Impurities (United States Pharmacopeia, 1985)

    7.5.3.1  Preparation of sodium nitrite to test

         Sodium nitrite (1 g) is dissolved in 6 ml hydrochloric acid
    (3 mol/l) and then evaporated to dryness on a steam bath.  The
    residue is reduced to a coarse powder, which is then heated on the
    steam bath until the hydrochloric acid odour is no longer
    perceptible.  This residue is then dissolved in 23 ml of water, and
    2 ml of acetic acid (1 mol/l) is added.

         The test is for metallic impurities that are coloured by the
    sulfide ion under specified test conditions, in terms of the
    percentage of lead (by weight) in the test substance.  Limits for
    sodium nitrite are 0.002%.

    7.5.3.2  Preparation of special reagents

         Lead nitrate stock solution is prepared by dissolving 159.8 mg
    lead nitrate in 100 ml of water and adding 1 ml of nitric acid. 
    This solution is then diluted to 1000 ml with water and stored in
    glass containers free of lead soluble salts.  Standard lead solution
    is prepared by diluting 10.0 ml of the lead nitrate stock solution
    with water to a volume of 100.0 ml, so that each 1 ml of standard
    lead solution contains the equivalent of 10 µg of lead.  A
    comparison solution prepared with 100 µl of standard lead solution
    per g of sodium nitrate contains the equivalent of 1 part of lead
    per million parts of sodium nitrate.

    7.5.3.3  Preparation of standard

         Pipette 2 ml of standard lead solution (containing 20 mg of
    lead) into a 50 ml colour comparison tube.  Dilute with water to
    25 ml and adjust pH to between 3.0 and 4.0 with acetic acid
    (1 mol/l) or ammonium hydroxide (6 mol/l) using narrow-range pH
    paper as an indicator.  Dilute to 40 ml with water and mix.

    7.5.3.4  Preparation of test

         Pipette 25 ml of the prepared sodium nitrite test material (see
    section 7.5.3.1) into a 50 ml colour comparison tube.  Using
    narrow-range pH indicator paper, adjust pH to between 3.0 and 4.0

    with either acetic acid (1 mol/l) or ammonium hydroxide (6 mol/l). 
    Then dilute to 40 ml with water and mix.

    7.5.3.5  Preparation of monitor

         Place 25 ml of solution prepared as in section 7.5.3.4 above
    into a 50 ml colour comparison tube and add 2.0 ml of the standard
    lead solution.  Using short-range pH indicator paper, adjust pH to
    between 3.0 and 4.0 with either (1 mol/l) acetic acid or ammonium
    hydroxide (6 mol/l).  Then dilute to 40 ml with water and mix.

    7.5.3.6  Preparation of hydrogen sulfide test solution (TS)

         Pass hydrogen sulfide gas into cold water making a saturated
    solution that should possess a strong odour of hydrogen sulfide. 
    Store in a cold, dark area in small, dark amber-coloured bottles
    filled nearly to the top.

    7.5.3.7  Test procedure

         Add 10 ml of freshly prepared hydrogen sulfide TS solution to
    each colour comparison tube.  Allow to stand for 5 min, then view
    downwards over a white surface.  The colour of the solution
    described in section 7.5.3.4 should not be darker than that of the
    solution described in section 7.5.3.3.  The colour intensity of the
    solution described in section 7.5.3.5 should be equal to or greater
    than that of the solution described in section 7.5.3.3.

    7.6  Shelf-life

         The shelf-life of sodium nitrite solution for injection is
    5 years, which might be somewhat reduced in very hot climates though
    stability testing at elevated temperatures has not been undertaken
    (Personal Communication, Eli Lilly & Co., March 1987).  The supplier
    recommends that sodium nitrite be stored at controlled room
    temperatures of 15 - 30 °C (Product Information, 1984).  Experiments
    with stored ampoules showed that sodium nitrite is stable in water
    for more than 21 years (Chen & Rose, 1956).

    7.7  General Properties

    7.7.1  Mode of action

         Sodium nitrite has seldom been utilized alone for the treatment
    of cyanide poisoning.  Rather, it is used in combination with
    intravenous sodium thiosulfate and, often, amyl nitrite by
    inhalation.  These combinations have been shown to be more
    efficacious in experimental poisoning than any of the compounds
    alone, demonstrating true antidotal synergy (Hug, 1933; Chen et al.,
    1933a; Chen et al., 1934; Chen & Rose, 1952; Way et al., 1984).

         Sodium nitrite induces methaemoglobin  in vitro, which binds
    cyanide (Hug & Marenzi, 1933a).  In animal studies with dogs and
    rabbits, sodium nitrite induced methaemoglobinaemia, with levels
    averaging 50% following a dose of 40 mg/kg (Hug & Marenzi, 1933a;
    Hug & Marenzi, 1933b).  Sodium nitrite alone has been shown to be
    efficacious in dogs with experimental cyanide poisoning (Chen et
    al., 1933a; Chen et al., 1934; Chen & Rose, 1952).  However, dog
    blood has been shown to be more sensitive to methaemoglobin
    induction by sodium nitrite  in vitro than human blood (Paulet &
    Menoret, 1954).  Methaemoglobin can be an efficacious cyanide
    antidote, as shown by results in experimental poisoning in rats
    treated with an exogenously prepared stroma-free methaemoglobin
    solution (Ten Eyck et al., 1986).  Human red blood cells, with
    methaemoglobin induced  in vitro, were injected intraperitoneally
    to cyanide-poisoned mice and had antidotal efficacy (Kruszyna et
    al., 1982).  However, animals pretreated with methylene blue did not
    develop significant methaemoglobin levels and had unchanged sodium
    nitrite antidotal efficacy in experimental cyanide poisoning (Holmes
    & Way, 1982).  Other mechanisms, including vasodilatation with
    changes in local capillary blood flow, have been postulated to
    account for part or all of the antidotal action of sodium nitrite
    (Way et al., 1984; Cohen & Guzzardi, 1984).

         Alternative mechanisms of action are suggested by the fact that
    sodium nitrite relaxes maximally contracted vascular smooth muscle
     in vitro (Kruszyna et al., 1985).  The contracted vascular smooth
    muscle relaxation was not reversed in this model by the addition of
    cyanide (Kruszyna et al., 1985).  In mice, methaemoglobin levels
    induced by sodium nitrite were more efficacious in protecting
    against cyanide poisoning than were comparable methaemoglobin levels
    induced by either hydroxylamine or 4-dimethylaminophenol (Kruszyna
    et al., 1982).  Equivalent amounts of methaemoglobin induced
     in vitro by these three agents did not bind significantly
    different amounts of cyanide (Kruszyna et al., 1982), suggesting
    that sodium nitrite may have some other mechanism of action.  The
    methaemoglobinaemia induced in these mice by sodium nitrite peaked
    later and at a lower level than the methaemoglobinaemia induced by
    hydroxylamine or 4-dimethylaminophenol, but was longer lasting
    (Kruszyna et al., 1982).  This may explain differences in efficacy
    noted in this animal study, as cyanide initially bound to
    methaemoglobin induced by the other agents may have been rapidly
    released to exacerbate the poisoning (Kruszyna et al., 1982).

         The oxidation of haemoglobin to methaemoglobin by sodium
    nitrite is favoured when the oxygen partial pressure is high and
    haemoglobin is in the oxyhaemoglobin state (Tomoda & Yoneyama,
    1979).  If methaemoglobin induction accounts for all, or part, of
    the antidotal action of sodium nitrite, this might explain the
    potentiation noted with concomitant oxygen administration (Mansouri,
    1985), although other investigators have not reached this conclusion
    (Way et al., 1966).

         The mechanism of action of sodium nitrite is not fully
    understood at present.  Methaemoglobin induced by sodium nitrite and
    measurable with a co-oximeter, however, does not bind cyanide at the
    time of measurement (cyanmethaemoglobin cannot be measured by these
    instruments).  Patients have survived seriously symptomatic acute
    cyanide poisoning, despite developing  measurable methaemoglobin
    levels of 10% or less (Hall et al., 1987; DiNapoli et al., 1989). 
    Administering futher sodium nitrite to a patient who has already had
    a satisfactory clinical improvement in order to produce a
    "therapeutic methaemoglobin level" of 25%, as has somethimes been
    recommended (Jones et al., 1987), is both unnecessary and
    potentially dangerous (DiNapoli et al., 1989).

    7.7.2  Other relevant properties

         Sodium nitrite is a vasodilating agent and has been used in the
    past for the treatment of angina (Martindale, 1989).  Rapid
    intravenous administration may cause hypotension (Viana et al.,
    1934; Hall & Rumack, 1986).  This effect can be avoided by beginning
    the infusion slowly and increasing the infusion rate cautiously with
    careful blood pressure monitoring.

    7.8  Animal Studies

    7.8.1  Pharmacokinetics

         Sodium nitrite blood levels have not been measured during
    antidote therapy and most pharmacokinetic parameters have not been
    determined.  Most studies have focused on the levels of
    methaemoglobin induced by sodium nitrite.

         In one study into fetal methaemoglobin formation, when pregnant
    rats were administered sodium nitrite orally or by intraperitoneal
    injection, peak maternal sodium nitrite blood levels were usually
    produced within the first 20 to 30 min, while peak methaemoglobin
    levels were observed a few minutes later (Gruener et al., 1973). 
    Peak sodium nitrite blood concentrations ranged from 3.9 mg/l
    (56.52 µmol/l), with an administered dose of 2.5 mg/kg, to 32.5 mg/l
    (471.01 µmol/l) with an administered dose of 30 mg/kg, and
    corresponded to peak maternal methaemoglobin levels of 3.4%
    (2.5 mg/kg dose) to 60.2% (30 mg/kg dose) (Gruener et al., 1973).

         Sodium nitrite produced an equimolar conversion of haemoglobin
    to methaemoglobin  in vitro in canine blood (Hug & Marenzi, 1933a). 
    However, only about 74% of an administered intravenous dose was
    involved in methaemoglobin formation in dogs  in vivo (Hug &
    Marenzi, 1933b).  Dogs administered 40 mg/kg of sodium nitrite
    intravenously developed approximately 50% methaemoglobinaemia (Hug &
    Marenzi, 1933b).  In these experiments using sodium nitrite alone,
    all of the administered dose of potassium cyanide was recovered in
    the urine as thiocyanate over a 4-day period (Hug & Marenzi, 1933b).

         In a dog with acute cyanide poisoning, given sodium nitrite
    (22.5 mg/kg) immediately and an additional 11.25 mg/kg at 29 and
    54 min after the poisoning, peak methaemoglobin levels of 65% at
    30 min and 60% at 4 h were observed (Chen & Rose, 1952).  The
    methaemoglobin concentration decreased rapidly after each peak,
    reaching 20% about 30 min after the first peak, and 4-5 h after the
    second peak (Chen & Rose, 1952).  Methaemoglobin disappeared from
    the circulation 11 h after the first injection of sodium nitrite 
    (Chen & Rose, 1952).

         In mice administered sodium nitrite 75 (mg/kg)
    intraperitoneally, methaemoglobin levels peaked at 30-35% about
    20 min after the injection (Kruszyna et al., 1982).  The levels
    remained in this range until 40 min after sodium nitrite
    administration and then decreased slowly, reaching 25%
    methaemoglobinaemia at 1 h and 10% at 2 h (Kruszyna et al., 1982). 
    These findings contrast with studies undertaken in mice with other
    methaemoglobin-inducing agents (4-dimethylaminophenol and
    hydroxylamine), which produced higher peak methaemoglobin levels
    more quickly (Kruszyna et al., 1982).  Following the peak level, the
    degree of methaemoglobinemia then decreased faster than that induced
    by sodium nitrite (Kruszyna et al., 1982).

          In vitro studies have indicated that canine blood is more
    susceptible to sodium nitrite-induced methaemoglobin conversion than
    human blood (Paulet & Menoret, 1954).  Rodent blood is more
    resistant to methaemoglobin formation than human blood (Kruszyna et
    al., 1982).

    7.8.2  Pharmacodynamics

         Sodium nitrite, at a minimum dose of 5 mg/kg, was shown to be
    an effective antidote for experimental cyanide poisoning in dogs,
    while 200 mg/kg was efficacious in rabbits (Hug, 1932; Hug, 1933). 
    In other experiments, sodium nitrite alone was shown to be an
    effective cyanide antidote in dogs (Chen et al., 1934), though the
    combination of sodium nitrite with a sulfur donor allowed dogs to
    tolerate doses of cyanide greater than those antagonized by the two
    agents separately (Chen et al., 1933a; Chen et al., 1933b; Hug,
    1933; Chen et al., 1934).  This antidotal synergy has been noted in
    numerous studies since the 1930s (e.g., Way et al., 1972).  Sodium
    nitrite has been shown to have protective effects against rat brain
    cytochrome oxidase inhibition  in vivo (Piantadosi et al., 1983).

         Two studies have reported data that are inconsistent with the
    classical antidotal mechanism of action, namely methaemoglobin
    induction.  When comparable levels of methaemoglobinaemia were
    induced in mice by 4-dimethylaminophenol, hydroxylamine, or sodium
    nitrite, cyanide was antagonized better by the administration of
    sodium nitrite (Kruszyna et al., 1982).  Following pre-treatment
    with methylene blue prior to cyanide poisoning, mice had unchanged

    sodium nitrite antidotal efficacy, even though they did not develop
    significant methaemoglobinaemia (Holmes & Way, 1982).

         Sodium nitrite causes relaxation of vascular smooth muscle
     in vitro (Kruszyna et al., 1985).  Sodium nitrite alone did not
    have any effect on either the respiratory or urinary excretion of
    14C-labelled sodium cyanide in mice (Burrows et al., 1982).  Other
    postulated antidotal mechanisms include vasodilatation with improved
    capillary blood flow (Way et al., 1984; Cohen & Guzzardi, 1984).

    7.8.3  Toxicology

         A rather large body of sub-acute and chronic toxicity data
    exists for sodium nitrite (Musil, 1966; Izmerov, 1982).  Effects
    including increased methaemoglobin levels, neuromuscular
    excitability, thyroid dysfunction, changes in conditioned reflexes,
    elevated serum protein and albumin concentrations, possible effects
    on hepatic drug metabolism systems, prolonged prothrombin  time, and
    decreased erythrocyte counts have been noted in chronic and subacute
    animal exposures (Musil, 1966; Izmerov, 1982; Gosselin et al.,
    1984).  In antidote administration, most subjects would receive only
    one dose, or at most a few doses over a short period of time, and
    these chronic and subacute animal studies would not seem to be
    particularly relevant.

         Pre-treatment of rats with sodium nitrite potentiates carbon
    tetrachloride hepatotoxicity (Suarez & Bhonsle, 1978).  A
    synergistic effect on acute hepatic necrosis in mice has been
    produced by oral administration of a combination of sodium nitrite
    and various secondary amines (Asahina et al., 1971).  A second
    abnormal haemoglobin derivative, nitrosohaemoglobin, has been
    produced in rats following intraperitoneal sodium nitrite injection
    (Rein et al., 1968).

         Sodium nitrite is listed as a non-carcinogen, but as giving a
    positive result in the Ames test, by the International Agency for
    Research on Cancer (IARC) (Kuroki et al., 1980).  It can react with
    secondary amines either  in vitro or  in vivo to produce
    potentially carcinogenic nitrosamines (Izmerov, 1982).   In vitro
    studies have shown induction of volatile mutagens in the faeces of
    normal human volunteers following treatment with sodium nitrite
    (MacDonald & Rao, 1982).  Administration of sodium nitrite and
    morpholine together to rats produced hepatic and forestomach tumours
    (Mirvish et al., 1983).  However, a cohort study of workers
    chronically exposed to a combination of sodium nitrite and various
    amines in cutting fluids did not show an increased cancer morbidity
    (Jarvholm et al., 1986).  Epidemiological data suggest a
    relationship between the consumption of large quantities of
    nitrite-nitrate-treated foodstuffs, a diet poor in ascorbic acid,
    and the development of gastric, oral cavity, and oesophageal cancers
    (Weisburger et al., 1983).

         Sodium nitrite administered orally or intraperitoneally to
    pregnant rats produced peak fetal sodium nitrite blood levels
    ranging from trace amounts, with a 2.5 mg/kg maternal dose and a
    corresponding maternal blood level of 3.9 mg/l (56-52 µmol/l), to
    9.4 mg/l (136.23 µmol/l) with a 30 mg/kg maternal dose and a
    corresponding maternal blood level of 32.5 mg/l (471.01 µmol/l)
    (Gruener et al., 1973).  Fetal methaemoglobin levels ranged from
    1.2% (with a maternal sodium nitrite dose of 2.5 mg/kg and a
    corresponding maternal peak methaemoglobin level of 3.4%) to 27.2%
    (with a maternal sodium nitrite dose of 30 mg/kg and a corresponding
    maternal peak methaemoglobin level of 60.2%) (Gruener et al., 1973). 
    Fetal effects other than methaemoglobin induction were not reported.

         Sodium nitrite was shown not to be teratogenic in several
    animal species when used as a cardiovascular agent or as a fungicide
    in combination with iodine (Schardein, 1985).  In combination with
    ethylurea, sodium nitrite forms the potent teratogen ethylnitrosurea
    (Schardein, 1985).  The median effective dose (ED50) for mortality
    in a chicken embryo study was 22 µmol/egg (Korhonen et al., 1983). 
    A maximum of 27% of the embryos were noted to be malformed (Korhonen
    et al., 1983).  A combination of sodium nitrite (50 mg/kg) and
    methyl urea (30 mg/kg), administered orally to rats on the 9th day
    of pregnancy, killed over 40% of the fetuses and produced
    malformations in about 15% of the surviving fetuses (Izmerov, 1982). 
    Administration of ascorbic acid blocked both the embryotoxic and
    teratogenic effects (Izmerov, 1982).  Sodium nitrite doses of
    100 mg/kg were teratogenic in rats (Izmerov, 1982).  Cyanosis was
    observed in 25% of newborn rats at birth following maternal
    administration of sodium nitrite (100 mg/kg) and in 6% of newborn
    rats with a maternal dose of 5 mg/kg (Izmerov, 1982).

         As sodium nitrite is only likely to be administered once, or at
    most a few times, as an antidote, the main concern would appear to
    be induction of fetal methaemoglobinaemia if administered to a
    pregnant patient.  The risk to the fetus from maternal cyanide
    poisoning would seem to override the risk from possible fetal
    methaemoglobin induction.  At clinically relevant doses (300 mg, or
    about 4.3 mg/kg for a 70-kg subject) (Product Information, 1984),
    data from a rat study suggested that the peak fetal methaemoglobin
    level would be in the neighbourhood of 2.7% (Gruener et al., 1973). 
    This amount of fetal methaemoglobinaemia is not likely to be
    clinically significant.  However, based on  in vitro comparisons of
    methaemoglobin reductase activity, the human fetus may have a weaker
    defense mechanism against the development of significant
    methaemoglobinaemia than has the rat fetus (Gruener et al., 1973).

         Some representative animal toxicity values for sodium nitrite
    are: LD50 rat (oral), 85 mg/kg; LDLo hamster (oral), 3 mg/kg;
    LD50 mouse (oral), 175 mg/kg; LD50 rat (intravenous), 65 mg/kg;
    LDLo dog (oral), 330 mg/kg; LDLo dog (intravenous), 15 mg/kg; LDLo
    rabbit (intravenous), 80 mg/kg (Sax, 1984; ITI, 1985).  A

    comparative human intravenous antidotal dose is 300 mg (or about
    4.3 mg/kg for a 70-kg subject) (Product Information, 1984).

    7.9  Volunteer Studies

    7.9.1  Pharmacokinetics

         Sodium nitrite is rapidly absorbed following oral
    administration (Martindale, 1989).  About 60% of the absorbed
    nitrite ion is metabolized in the body, partially to ammonia, while
    the rest of the absorbed dose is excreted unchanged in the urine
    (Martindale, 1982).  The nitrite ion disappears from the circulatory
    system quickly, but little is known about its fate (Baselt, 1982). 
    An intravenous injection of 400 mg in humans produced a peak
    methaemoglobin concentration of 10.1%, while 600 mg produced a peak
    methaemoglobin level of 17.5% (Chen & Rose, 1952).  A later study 
    in normal volunteers showed that a methaemoglobin level of 6% was
    produced by the intravenous injection of 1 mg sodium nitrite per kg
    (about 300 mg, 10 ml of a 3% solution of sodium nitrite) (Weger,
    1968).  In this same study, injecting 12 mg/kg (about 900 mg, 30 ml
    of a 3 % solution of sodium nitrite) resulted in a 30%
    methaemoglobinaemia, but also in clinical shock (Weger, 1968).

    7.9.2  Sodium nitrite poisoning

         Most human toxicity data relates to accidental or suicidal
    ingestion of sodium nitrite from industrial, laboratory, or food
    additive sources.  Fatal poisoning has occurred when sodium nitrite
    has been substituted for table salt and used as seasoning
    (MacQuiston, 1936; Padberg & Martin, 1939) or when food contaminated
    with motor vehicle cooling fluid was consumed (Ten Brink et al.,
    1982).  A case of suicide from the ingestion of sodium nitrite has
    been reported, with a post-mortem methaemoglobin level of 90%
    (Standefer et al., 1979).  Ingestion of sausages cured with sodium
    nitrite produced symptomatic poisoning with syncope, hypotension,
    and methaemoglobinaemia and a decreased arterial oxygen saturation
    (Bakshi et al., 1967).  Two patients who mistakenly ingested sodium
    nitrite instead of table salt developed methaemoglobin levels of 34%
    and 54%, but recovered after being administered methylene blue and
    supplemental oxygen (Aquanno et al., 1981).

         The range of toxicity of sodium nitrite is difficult to
    determine, as the ingested dose is seldom known.  A patient who
    ingested 14.5 g complained of transient darkening of both visual
    fields (Grant, 1986).  A 17-month-old child died with a
    methaemoglobin level greater than 90% after being given 450 mg
    (32 mg/kg) intravenously in the mistaken impression that acute
    cyanide poisoning was present (Berlin, 1970).  A two-month-old
    infant developed severe poisoning but survived after ingesting
    130 mg sodium nitrite (Gosselin et al., 1984).  The mean lethal oral
    dose in adults is probably in the neighbourhood of 1 g if no

    treatment is given (Gosselin et al., 1984), though survival has
    followed the ingestion of 1 g (Baselt, 1982).

         Treatment of acquired methaemoglobinaemia from sodium nitrite
    poisoning in circumstances similar to those described above may
    involve only supplemental oxygenation and observation if the patient
    is asymptomatic and the methaemoglobin level is 20-30% or less (Hall
    et al., 1986a).  With higher methaemoglobin levels or in symptomatic
    patients, intravenous infusion of methylene blue at a usual dose of
    0.1-0.2 ml/kg of a 1% solution (1-2 mg/kg) may be necessary (Hall et
    al., 1986b).  Toluidine blue may be used when methylene blue is not
    available.  Exchange transfusion may be required if severely
    poisoned patients are not responsive to the above measures (Hall et
    al., 1986a).  Experimental animal studies have suggested that
    hyperbaric oxygen is both efficacious (Goldstein & Doull, 1971) and
    not efficacious (Sheehy & Way, 1974) in reducing nitrite-induced
    methaemoglobinaemia.  Hyperbaric oxygen can be used to maintain
    tissue oxygenation while exchange transfusion is prepared (Hall et
    al., 1986a).

         If the therapeutic administration of sodium nitrite in cyanide
    poisoning produces excessive methaemoglobinaemia (Viana et al.,
    1934; Berlin, 1970; Lasch & El Shawa, 1981; Feihl et al., 1982),
    there is some controversy about appropriate treatment.  The
    administration of methylene blue has been recommended (Product
    Information, 1984), and toluidine blue has also been used.  However,
    exchange transfusion rather than methylene or toluidine blue
    administration has been suggested because conversion of
    cyanmethaemoglobin back to normal haemoglobin could theoretically
    release bound cyanide and worsen the poisoning (Rumack, 1987).  This
    controversy is unresolved at present.

    7.10  Clinical Studies

         No controlled human clinical trials have been performed to
    compare the efficacy of various cyanide antidotes in acute human
    poisoning.

    7.11  Clinical Studies - Case Reports

         In the absence of controlled clinical trials, anecdotal reports
    of human poisoning treated with sodium nitrite are all that can be
    evaluated.  To compound the problem, only a single case of acute
    cyanide poisoning treated solely with sodium nitrite has been
    reported (Mota, 1933).  All other reported patients were given more
    than one antidote.  Many patients have received a combination of
    sodium nitrite and sodium thiosulfate, or amyl nitrite and sodium
    nitrite plus sodium thiosulfate (Chen & Rose, 1952; Chen & Rose,
    1956).  In other cases, sodium nitrite has been administered in
    various combinations with sodium thiosulfate, methylene blue,
    dicobalt-EDTA (Kelocyanor [R]), and hydroxocobalamin (Motin et al.,

    1970; Lutier et al., 1971; Yacoub et al., 1974).  The clinical
    efficacy of sodium nitrite used alone in acute cyanide poisoning is
    therefore impossible to separate from its use in an antidotal
    combination, and, judging from various animal studies showing
    antidotal synergism between sodium nitrite and sodium thiosulfate
    (Chen et al., 1934; Chen & Rose, 1952; Way et al., 1972), it should
    probably not be used alone.

         The first reported cases of human acute cyanide poisoning to be
    treated with a combination of sodium nitrite and sodium thiosulfate
    were described by Viana et al. (1934).  Their first patient ingested
    about 5 g of potassium cyanide and became comatose with a weak pulse
    and respiratory distress, but recovered after being given 1500 mg of
    sodium nitrite and 18 g of sodium thiosulfate.  Their second patient
    ingested about 2 g of potassium cyanide and developed coma,
    respiratory distress, and convulsions, but recovered after being
    given 750 mg of sodium nitrite and 12 g of sodium thiosulfate.  No
    cyanide assays were undertaken.  Hypotension and cyanosis developed
    in both cases (Viana et al., 1934).

         The largest case series of acute cyanide poisoning treated with
    the amyl nitrite/sodium nitrite/sodium thiosulfate antidote
    combination was assembled by Chen & Rose (1952, 1956) comprising a
    total of 49 patients.  One patient who was moribund before being
    administered these antidotes did not recover (Chen & Rose, 1952). 
    Only historical evidence of cyanide poisoning was available in these
    cases; no cyanide assays were reported (Chen & Rose, 1952; Chen &
    Rose, 1956).

         Survival following acute cyanide poisoning from potassium and
    sodium cyanide (De Busk & Seidl, 1969; Stewart, 1974; Feihl et al.,
    1982; Peters et al., 1982; Wood, 1982; Litovitz et al., 1983; Wesson
    et al., 1985; Hall et al., 1987), cyanogenic plants (Sayre &
    Kaymakcalan, 1964; Rubino, 1978; Lasch & El Shawa, 1981; Shragg et
    al., 1983), and laetrile (Moss et al., 1981; Beamer et al., 1983;
    Hall et al., 1986b) has been reported.  However, not all patients
    with severe cyanide poisoning given the sodium nitrite/sodium
    thiosulfate antidote combination have survived (Chen & Rose, 1952;
    Braico et al., 1979; Litovitz et al., 1983).  Patients with severe
    acute cyanide poisoning (including coma, convulsions, and metabolic
    acidosis) and blood cyanide levels, measured at various times after
    exposure, ranging from 0.26 to 40 mg/l (10 to 1539 µmol/l) have
    survived following the administration of sodium nitrite and sodium
    thiosulfate in combination with supportive measures (Rubino, 1978;
    Moss et al., 1981; Peters et al., 1982; Feihl et al, 1982; Litovitz
    et al., 1983; Shragg et al., 1983; Wesson, et al., 1985; Hall et
    al., 1987; Hall et al., 1986b).  Patients poisoned with cyanide have
    survived with supportive treatment only (Graham et al., 1977; Vogel
    et al., 1981; Brivet et al., 1983), but the highest reported blood
    cyanide level in a patient who survived without receiving specific
    antidote treatment was 3.8 mg/l (147 µmol/l) (Edwards & Thomas,

    1978).  This suggests that patients with more severe cyanide
    poisoning may have a better chance of recovery with both supportive
    measures and specific antidotes than with supportive measures alone
    (Hall & Rumack, 1986).

         Administration of the sodium nitrite/sodium thiosulfate
    antidote combination in five smoke inhalation victims with carbon
    monoxide and cyanide poisoning and mean blood cyanide levels of
    1.62 mg/l (62 µmol/l) has been reported (Hart et al., 1985).  These
    five patients also received hyperbaric oxygen therapy.  There were
    four survivors and one fatality in this series.

    7.12  Summary of Evaluation

    7.12.1  Indications

         Sodium nitrite is indicated for use in acute cyanide poisoning. 
    It should not be used except where there is definite indication of
    severe poisoning, such as loss of consciousness and deteriorating
    vital functions.  It is usually administered with sodium thiosulfate
    and its administration may be preceded by the use of amyl nitrite.

    7.12.2  Contraindications

         Sodium nitrite should not be administered to asymptomatic
    patients following exposure to cyanide.  It should not be
    administered to patients with smoke inhalation and combined carbon
    monoxide and cyanide poisoning unless hyperbaric oxygen therapy is
    available and such therapy has already been initiated.

         G6DP-deficient individuals are theoretically at great risk from
    nitrite therapy because of the likelihood of severe haemolysis,
    although no such cases have been reported.

    7.12.3  Advised route and dosage

         Sodium nitrite is administered intravenously in an initial
    adult dose of 300 mg (10 ml of a 3% solution) (Chen & Rose, 1952;
    Hall & Rumack, 1986).  While many authors recommend that this dose
    be infused over a 5-min period, hypotension from the vasodilating
    properties of sodium nitrite may occur.  Hypotension may be avoided
    by diluting sodium nitrite with normal saline and infusing the
    medication over a 20-min period with frequent blood pressure
    monitoring (Hall, 1986).  The rate of infusion may be increased if
    no hypotension occurs (Hall, 1986).  Methaemoglobin levels should be
    monitored to avoid excessive methaemoglobin induction (Berlin, 1970)
    (see chapter 10).  The paediatric dose for an average child,
    frequently quoted in the literature and based on  in vitro
    experiments with canine blood, is 0.33 ml of a 3% solution per kg
    body weight (about 10 mg/kg), administered intravenously with the
    precautions noted above (Hall & Rumack, 1986).  A pediatric dose of

    approximately 0.2 ml/kg of a 3% solution has also been recommended
    (Product Information, 1984).  If there is reason to suspect anaemia,
    tables devised to calculate a reduced dose, taking into account the
    relatively lesser amount of haemoglobin present, should be consulted
    (Berlin, 1970).

         The adult dose may be lethal for a child.  With a haemoglobin
    level of 12 g/100 ml, only 10 mg/kg per body weight can be
    administered immediately and 5 mg/kg repeated within 30 min if
    necessary (Berlin, 1970).  However, the calculation of Berlin was
    based on a therapeutic level of methaemoglobinaemia of 15 mg/kg. 
    This may be too high because Weger (1968) observed severe
    circulatory collapse in volunteers after 12 mg/kg was given
    intravenously.  In the same study, the administration of 4 mg/kg
    (10 ml of a 3% solution NaNO2 solution, i.e., 300 mg) induced only
    6% methaemoglobinaemia.  As the adult dose of about 4 mg/kg has been
    shown to be effective clinically, it may be safer to begin treatment
    in children with sodium nitrite at 4 mg/kg (about 0.13 ml of a 3%
    solution per kg body weight) and to administer additional sodium
    nitrite only if excessive methaemoglobinaemia is not present and a
    satisfactory clinical response has not occurred.

         In both adults and children, another sodium nitrite dose of 50%
    the initial amount administered may be repeated 30 min after the
    first dose if there is inadequate clinical improvement (Hall &
    Rumack, 1986).

    7.12.4  Other consequential or supportive therapy

         See section 1.10.

    7.13  Model Information Sheet

    7.13.1  Uses

         Sodium nitrite is indicated for the treatment of acute cyanide
    poisoning.

    7.13.2  Dosage and route

         The adult dose is 300 mg (10 ml of a 3% solution) infused
    intravenously at the fastest rate possible without causing
    hypotension (usually over 5 to 20 min).  The initial paediatric dose
    is 0.13-0.33 ml of a 3% solution per kg body weight (about
    4-10 mg/kg).  It is advisable to begin with lower doses in children
    and increase to the desired effect.  If anaemia is suspected,
    standard tables should be consulted to calculate a reduced
    paediatric sodium nitrite dose.

         In both adults and children, another dose of one-half the
    initial amount administered may be repeated 30 min after the first
    dose if there is inadequate clinical improvement.

    7.13.3  Precautions/contraindications

         Excessive methaemoglobinaemia may occur, especially when doses
    larger than those recommended are administered to children.

         Hypotension may occur following rapid administration of sodium
    nitrite, owing to its vasodilating properties.  Blood pressure
    should be monitored carefully during sodium nitrite administration,
    and the infusion rate slowed if hypotension occurs.

         Patients with smoke inhalation and combined carbon monoxide and
    cyanide poisoning with elevated carboxyhaemoglobin levels should not
    be given sodium nitrite unless treatment in a hyperbaric oxygen
    chamber is available and such treatment has been initiated.

         G6PD-deficient individuals are theoretically at great risk from
    sodium nitrite therapy because of the likelihood of severe
    haemolysis, although no such cases have been reported.

    7.13.4  Adverse effects

         Excessive methaemoglobinaemia may occur, especially with doses
    exceeding those recommended.  Hypotension may occur with rapid
    intravenous infusion.

    7.13.5  Use in pregnancy and lactation

         There are no reported cases of the use of sodium nitrite during
    pregnancy or lactation.  Animal experiments indicate that some
    sodium nitrite crosses the placenta and that fetal
    methaemoglobinaemia may be induced.  The risk to the fetus from
    severe maternal cyanide poisoning would seem to be greater than the
    risk of fetal methaemoglobin induction.  No animal studies have
    addressed the question of sodium nitrite excretion in breast milk or
    its possible effects on the nursing infant.

    7.13.6  Storage

         Sodium nitrite should be stored at a controlled temperature of
    15 to 30 °C (59 to 86 °F).  The shelf-life is 5 years.

    7.14  References

    Aquanno JJ, Chan KM, & Dietzler DN (1981) Accidental poisoning of
    two laboratory technologists with sodium nitrite. Clin Chem, 27:
    1145-1146.

    Asahina S, Friedman MA, Arnold E, Millar GN, Mishkin M, Bishop Y, &
    Epstein SS (1971) Acute synergistic toxicity and hepatic necrosis
    following oral administration of sodium nitrite and secondary amines
    to mice. Cancer Res, 31: 1201-1205.

    Bakshi SP, Fahey JL, & Pierce LE (1967) Sausage cyanosis - Acquired
    methemoglobinemic nitrite poisoning. New Engl J Med, 277: 1072.

    Baselt RC (1982) Cyanide and nitrite. In: Disposition of toxic drugs
    and chemicals in man, 2nd ed. Davis, California, Biomedical
    Publications, pp 209-214, 555-557.

    Beamer WC, Shealy RM, & Prough DS (1983) Acute cyanide poisoning
    from laetrile ingestion. Ann Emergency Med, 12: 449-451.

    Becker CM (1985) The role of cyanide in fires. Vet Hum Toxicol, 27:
    487-490.

    Berlin AM (1970) The treatment of cyanide poisoning in children.
    Pediatrics, 46: 793-796.

    Braico KT, Humbert JR, Terplan KL, & Lehotay JM (1979) Laetrile
    intoxication:  Report of a fatal case. New Engl J Med, 300: 238-240.

    Brivet F, Delfraissy JF, Duche M, Bertrand P, & Dormont J (1983)
    Acute cyanide poisoning: Recovery with non-specific supportive
    therapy. Intensive Care Med, 9: 33-35.

    Burrows GE, Liu DHW, Isom GE, & Way JL (1982) Effect of antagonists
    on the physiologic disposition of sodium cyanide. J Toxicol Environ
    Health, 10: 181-189.

    Chen KK & Rose CL (1952) Nitrite and thiosulfate therapy in cyanide
    poisoning. J Am Med Assoc, 149: 113-119.

    Chen KK & Rose CL (1956) Treatment of acute cyanide poisoning. J Am
    Med Assoc, 162: 1154-1155.

    Chen KK, Rose CL, & Clowes GHA (1933a) Methylene blue, nitrites, and
    sodium thiosulphate against cyanide poisoning. Proc Soc Exp Biol
    Med, 31 : 250-251.

    Chen KK, Rose CL, & Clowes GHA (1933b) Potentiation of antidotal
    action of sodium tetrathionate and sodium nitrite in cyanide
    poisoning. Proc Soc Exp Biol Med, 31: 252-253.

    Chen KK, Rose CL, & Clowes GHA (1934) Comparative values of several
    antidotes in cyanide poisoning. Am J Med Sci, 188: 767-781.

    Cohen MA & Guzzardi LJ (1984) A letter on the treatment of cyanide
    poisoning.  Vet Hum Toxicol, 26: 503-504.

    De Busk RF & Seidl LG (1969) Attempted suicide by cyanide:  A report
    of two cases. California Med, 110: 394-396.

    DiNapoli J, Hall AH, Orake R, & Rumack BH (1989) Cyanide and arsenic
    poisoning by intravenous injection. Ann Emergency Med, 18: 308-311.

    Edwards AC & Thomas ID (1978) Cyanide poisoning. Lancet, 1: 92-93.

    Feihl F, Domenighetti G, & Perret Cl (1982) Intoxication massive au
    cyanure avec évolution favorable. Schweiz Med Wochenschr, 112:
    1280-1282.

    Goldstein GM & Doull J (1971) Treatment of nitrite-induced
    methemoglobinemia with hyperbaric oxygen. Proc Soc Exp Biol Med,
    138: 137-139.

    Gosselin RE, Smith RP, & Hodge HC ed. (1984) Nitrite. In: Clinical
    toxicology of commercial products, 5th ed. Baltimore, Maryland,
    Williams and Wilkins, pp 314-319.

    Graham DL, Lamand D, Theodore J, & Robin ED (1977) Acute cyanide
    poisoning complicated by lactic acidosis and pulmonary edema. Arch
    Intern Med, 137: 1051-1055.

    Grant WM (1986) Sodium nitrite. In: Toxicology of the eye, 3rd ed.
    Springfield, Illinois, Charles C. Thomas, p 840.

    Gruener N, Shuval HI, Behroozi K, Cohen S, & Shechter H (1973)
    Methemoglobinemia induced by transplacental passage of nitrites in
    rats. Bull Environ Contam Toxicol, 9: 44-48.

    Hall AH (1986) Cyanide poisoning: Dealing with unexpected menace.
    Emergency Med, 18: 919-205.

    Hall AH & Rumack BH (1986) Clinical toxicology of cyanide. Ann
    Emergency Med, 15: 1067-1074.

    Hall AH, Kulig KW, & Rumack BH (1986a) Drug- and chemical-induced
    methaemoglobinaemia: Clinical features and management. Med Toxicol,
    1: 253-260.

    Hall AH, Linden CH, Kulig KW, & Rumack BH (1986b) Cyanide poisoning
    from laetrile ingestion: Role of nitrite therapy. Pediatrics, 78(2):
    269-272.

    Hall AH, Doutre W, Ludden T, Kulig K, & Rumack BH (1987)
    Nitrite/thiosulfate treated acute cyanide poisoning: Estimated
    kinetics after antidote. Clin Toxicol, 25: 121-133.

    Hart GB, Strauss MB, Lennon PA, & Whitcraft DD (1985) Treatment of
    smoke inhalation by hyperbaric oxygen. J Emergency Med, 3: 211-215.

    Harvey SC (1980) Sodium nitrite. In: Osol A ed. Remington's
    pharmaceutical science, 16th ed. Easton, Pennsylvania, Mack
    Publishing Co., p 782.

    Holmes RK & Way JL (1982) Mechanism of cyanide antagonism by sodium
    nitrite. Pharmacologist, 24: 182 (Abstract).

    Hug E (1932) L'intoxication par l'acide cyanhydrique. Action
    antidote du bleu de méthylène, du nitrite de sodium et du sulfure de
    sodium. C R Soc Biol (Paris), 111: 88-90.

    Hug E (1933) Action combinée du nitrite de sodium dans le traitement
    de l'intoxication cyanhydrique chez le lapin. C R Soc Biol (Paris),
    114: 87-89.

    Hug E & Marenzi AD (1933a) La fixation de l'acide cyanhydrique par
    les hématies qui contiennent de la méthémoglobine. C R Soc Biol
    (Paris), 114: 84-86.

    Hug E & Marenzi AD (1933b) Mécanisme de l'action antidote du sodium
    vis-à-vis de l'intoxication par l'acide cyanhydrique. C R Soc Biol
    (Paris), 114: 86-87.

    ITI (1985) Sodium nitrite.  In: Toxic and hazardous industrial
    chemicals safety manual. Tokyo, The International Technical
    Information Institute, p 485.

    Izmerov NF ed. (1982) IRPTC Scientific Reviews of Soviet Literature
    on Toxicity and Hazards of Chemicals No. 22:  Nitrites. Moscow,
    Centre of International Projects, GKNT, 32 pp.

    Jarvholm B, Lavenius B, & Sallsten G (1986) Cancer morbidity in
    workers exposed to cutting fluids containing nitrites and amines. Br
    J. Ind Med, 43: 563-565.

    Jones J, McMullen MJ, & Dougherty J (1987) Toxic Smoke Inhalation: 
    cyanide poisoning in fire victims.  Am J Emerg Med, 5: 318-321.

    Korhonen A, Hemminki K, & Vainio H (1983) Toxicity of rubber
    chemicals towards three-day chicken embryos. Scand J Work Environ
    Health, 9: 115-119.

    Kruszyna R, Kruszyna H, & Smith RP (1982) Comparison of
    hydroxylamine, 4-dimethylaminophenol and nitrite protection against
    cyanide poisoning in mice. Arch Toxicol, 49: 191-202.

    Kruszyna H, Kruszyna R, & Smith RP (1985) Cyanide and sulfide
    interact with nitrogenous compounds to influence the relaxation of
    various smooth muscles. Proc Soc Exp Biol Med, 179: 44-49.

    Kuroki T, Abbondandolo A, Drevon C, Huberman E, & Laval F (1980)
    Mutagenesis assays with mammalian cells. In: Long-term and
    short-term screening assays for carcinogens: A critical appraisal.
    Lyons, International Agency for Research on Cancer, pp 107-133 (IARC
    Monographs on the Evaluation of the Carcinogenic Risk of Chemicals
    to Humans, Supplement 2).

    Lasch EE & El Shawa R (1981) Multiple cases of cyanide poisoning by
    apricot kernels in children from Gaza. Pediatrics, 68(1): 5-7.

    Litovitz TL, Larkin RF, & Myers RAM (1983) Cyanide poisoning treated
    with hyperbaric oxygen. Am J Emergency Med, 1: 94-101.

    Lutier F, Dusoleil P, & De Montgros J (1971) Action de
    l'hydroxocobalamine à dose massive dans l'intoxication aiguë au
    cyanure. (A propos d'un cas). Arch Mal Prof, 32: 683-686.

    MacDonald IA & Rao BG (1982) Nitrite-induced volatile fecal
    mutagens: Some preliminary studies. Cancer, 49: 1405-1408.

    MacQuiston TAC (1936) Fatal poisoning by sodium nitrite. Lancet, 2:
    1153-1154.

    Mansouri A (1985) Methemoglobinemia. Am J Med Sci, 298: 200-209.

    Martindale (1989) In: Reynolds JEF ed. The extra pharmacopoeia, 29th
    ed. London, The Pharmaceutical Press, p 854.

    Mirvish SS, Salmasi S, Cohen SM, Patil K, & Mahboubi E (1983) Liver
    and forestomach tumours and other forestomach lesions in rats
    treated with morpholine and sodium nitrite, with and without sodium
    ascorbate. J Natl Cancer Inst, 71: 81-85.

    Moss M, Khalil N, & Gray J (1981) Deliberate self-poisoning with
    laetrile. Can Med Assoc J, 125: 1126-1128.

    Mota MM (1933) [On a case of intoxication by potassium cyanide
    treated with sodium nitrite with survival.] Rev Med Rosario, 23:
    670-674 (in Spanish).

    Motin J, Bouletreau P, & Rouzioux JM (1970) Intoxication
    cyanhydrique grave traitée avec succès par l'hydroxocobalamine. J
    Med Strasbourg, 1: 717-722.

    Musil J (1966) [The effects of chronic sodium nitrite intoxication
    in rats.] Acta Biol Med Ger, 16: 388-394 (in German).

    Padberg LR & Martin T (1939) Three fatal cases of sodium nitrite
    poisoning.  J Am Med Assoc, 113: 1733.

    Paulet G & Menoret P (1954) Etude comparée de la sensibilité du sang
    de l'homme (frais ou conservé) et du sang de chien à divers
    méthémoglobinisants. C R Soc Biol (Paris), 148: 1014-1017.

    Peters CG, Mundy JVB, & Rayner PR (1982) Acute cyanide poisoning: 
    The treatment of a suicide attempt. Anaesthesia, 37: 582-586.

    Piantadosi CA, Sylvia AK, & Jobsis FF (1983) Cyanide-induced
    cytochrome a,a3 oxidation-reduction responses in rat brain  in vivo.
    J. Clin Invest, 72: 1224-1233.

    Product Information (1984) Product information for the cyanide
    antidote package[R]. Indianapolis, Indiana, Eli Lilly & Co.

    Rein H, Ristau O, Kupfer E, & Jung F (1968) [Nitrosohaemoglobin from
    poisoning with sodium nitrite.] Folia Haematol, 90: 176-181 (in
    German).

    Rubino JM (1978) Cyanide toxicity: Report of a case. Connecticut
    Poison Inf Bull, 3: 1-6.

    Rumack BH ed. (1987) Cyanide poisoning management. In: 
    Poisindex[R]: Poison information system. Denver, Colorado,
    Micromedex 52: 1-F4.

    Sax NI (1984) Sodium nitrite. In: Dangerous properties of industrial
    materials, 6th ed. New York, Van Nostrand Reinhold Co., pp
    2442-2443.

    Sayre JW & Kaymakcalan S (1964) Cyanide poisoning from apricot seeds
    among children in central Turkey. New Engl J Med, 270: 1113-1115.

    Schardein JL (1985) Chemically induced birth defects. New York,
    Marcel Dekker, pp 74, 759, 810-824.

    Sheehy MH & Way JL (1974) Nitrite intoxication: Protection with
    methylene blue and oxygen. Toxicol Appl Pharmacol, 30: 221-226.

    Shragg TA, Albertson TE, & Fischer CJ (1983) Cyanide poisoning after
    bitter almond ingestion. West J. Med, 136: 65-69.

    Standefer JC, Jones AM, Street E, & Inserra R (1979) Death
    associated with nitrite ingestion: Report of a case. J Forensic Sci,
    24: 678-771.

    Stewart R (1974) Cyanide poisoning. Clin Toxicol, 7: 561-564.

    Suarez KA & Bhonsle P (1978) Enhanced hepatotoxicity of carbon
    tetrachloride following sodium nitrite pre-treatment. Arch Int
    Pharmacodyn, 234: 329-334.

    Surgenor DM (1970) Sodium nitrite USP. In: Osol E ed. Remington's
    pharmaceutical sciences, 14th ed. Easton, Pennsylvania, Mack
    Publishing Co., pp 846-847.

    Ten Brink WAG, Wiezer JHA, Luijpen AFMG, Van Heijst ANP, Pikaar SA,
    & Seldenrijk R (1982) Nitrate poisoning caused by food contaminated
    with cooling fluid. J Toxicol Clin Toxicol, 19: 139-147.

    Ten Eyck RP, Schaerdel AD, & Ottinger WE (1986) Comparison of
    nitrite treatment and stroma-free methemoglobin solution as
    antidotes for cyanide poisoning in a rat model. Clin Toxicol, 23:
    477-487.

    Tomoda A & Yoneyama Y (1979) On the relationship of hemoglobin
    oxidation with the conformation of hemoglobin. Experientia (Basel),
    35: 15-16.

    United States Pharmacopeia (1985) Sodium nitrite and sodium nitrite
    injection. In: United States Pharmacopoeia, 21st ed. Rockville,
    Maryland, United States Pharmacopeial Convention, Inc., pp 974,
    1186-1187, 1189-1190, 1206.

    US NIOSH (1983) Registry of toxic effects of chemical substances
    1981-1982.  Atlanta, Georgia, National Institute of Occupational
    Safety and Health, p 857.

    Viana C, Cagnoli H, & Cedan J (1934) Action du nitrite de sodium
    dans l'intoxication au cyanure. C R Soc Biol (Paris), 115:
    1649-1651.

    Vogel SN, Sultan TR, & Ten Eyck RP (1981) Cyanide poisoning. Clin
    Toxicol, 18: 367-383.

    Way JL, Gibbon SL, & Sheehy MH (1966) Effect of oxygen on cyanide
    intoxication. I. Prophylactic protection. J Pharmacol Exp Ther, 153:
    381-385.

    Way JL, End E, Sheehy MH, De Miranda P, Feitknecht UF, Bachand R,
    Gibbon SC, & Burrows GE (1972) Effects of oxygen on cyanide
    intoxication. IV. Hyperbaric oxygen. Toxicol Appl Pharmacol, 22:
    415-421.

    Way JL, Sylvester D, Morgan RL, Isom GE, Burrows GE, Tamulinas CB, &
    Way JL (1984) Recent perspectives on the toxicodynamic basis of
    cyanid antagonism. Fundam Appl Toxicol, 4: S231-S239.

    Weger N (1968) [Aminophenol as a cyanide antidote.] Arch Toxikol,
    24: 49-50 (in German).

    Weisburger JH, Horn CL, & Barnes WS (1983) Possible genotoxic
    carcinogens in foods in relation to cancer causation. Semin Oncol,
    10: 330-341.

    Wesson DE, Foley R, Sabatini S, Wharton J, Kapusnik J, & Kurtzman NA
    (1985) Treatment of acute cyanide intoxication with hemodialysis. Am
    J Nephrol, 5: 121-126.

    Windholz M, Budavari S, & Blumetti RF, ed. (1983) Sodium nitrite.
    In: The Merck index, 10th ed. Rahway, New Jersey, Merck & Co., Inc.,
    p 8480.

    Wood GC (1982) Acute cyanide intoxication: Diagnosis and management.
    Clin Toxicol Consultant, 4: 140-149.

    Yacoub M, Faure J, Morena H, Vincent M, & Gaure H (1974)
    L'intoxication cyanhydrique aiguë. Données actuelles sur le
    métabolisme du cyanure et le traitement par hydroxocobalamine. J Eur
    Toxicol, 7: 22-29.

    8.  4-DIMETHYLAMINOPHENOL (4-DMAP)

    8.1  Introduction

         Cyanide has a special affinity for ferric ions, which are found
    in cytochrome oxidase, the terminal oxidative respiratory enzyme in
    mitochondria.  Blood contains a substantial quantity of ferrous ions
    within haemoglobin and it is possible to convert these readily to
    ferric ions, for example, by the use of nitrites as suggested by
    Chen et al. (1933).  A disadvantage of the use of nitrites is the
    fact that the concentration of methaemoglobin rises only slowly
    after intravenous administration.  In six experiments on volunteers,
    using the dose of sodium nitrite suggested by Chen (4 mg/kg), a
    methaemoglobin level of 6% resulted after 40 min (Kiese & Weger,
    1969).  This was considered to be too low a concentration of
    methaemoglobin after too long a time.  Experiments with several
    aminophenols showed that the use of 4-dimethylaminophenol (4-DMAP)
    resulted in rapid and controlled formation of methaemoglobin.  An
    intravenous dose of 3.25 mg/kg resulted in a methaemoglobin
    concentration of 30% in 10 min, a methaemoglobin level of 15% being
    reached within one min.

    8.2  Name and Chemical Formula

         Chemical name:              Dimethyl( para)aminophenol
                                     hydrochloride

         Formula:

    FIGURE 06

         Total formula:              C8H11 ON.HCl
         Relative molecular mass:    173.5
         Appearance:                 snow-white crystals
         CAS number:                 619-60-3
         Manufacturer:               Dr Franz Koehler Chemie GmbH,
                                     Alsbach, Germany
         Commercial name:            4-dimethylaminophenol (4-DMAP)

    8.3  Physico-chemical Properties

         Raw material:               snow-white colourless crystals
         Melting point:              145 °C + 1 °C
         Solubility:                 very soluble in water.  The
                                     solution is oxidized by contact
                                     with air and changes from being
                                     colourless to black-brown
         Optical properties:
         IR-spectrum:                peaks at (cm-1)
            833:                     l,4-disubstituted aromatics
            1240:                    phenols
            1275:                    amine
            1500:                    aromatics
            2700:                    C-H aromatics
            3200:                    N-H aromatics
         UV data:                    maximum at 271 nm; extinction
                                     coefficient 1255 l/mol
         Acidity:                    no data available

                            pK (-N(CH3)2) = 6.15
                               pK (-OH) = 10.1

         Stability in light:         no data available
         Thermal stability:          no data available
         Refractive index
         and specific gravity:       unknown
         Weight on drying:           no data available
         Excipients and
         pharmaceutical aids:        unknown for the final product
                                     although it may contain an
                                     antioxidant such as sodium
                                     pyrosulfate

    8.4  Synthesis

         Routes of synthesis:        no information available
         Manufacturing processes:    no information available

    8.5  Analytical Methods

    8.5.1  Identity

         To 5 ml of liquid from the ampoule, add one drop 30% hydrogen
    peroxide.  The solution will turn a deep violet colour within 5 min. 
    To 2.5 ml of liquid from the ampoule, add 0.5 ml 10% ferrichloride. 
    A pale-brown colour will appear immediately.

    8.5.2  Quantification

         AgNO3 titration:  2.5 ml liquid from the ampoule is diluted
    in 40 ml water and, after the addition of 3 ml nitric acid
    (2 mol/l), potentiometrically titrated with silver nitrate
    (0.1 mol/l).  1 ml AgNO3 (0.1 mol/l) equiv. 17.365 mg
    dimethylaminophenol hydrochloride.

         HClO4 titration: 2.5 ml liquid from the ampoule is added to
    40 ml of a 3% solution of mercury acetate in glacial acetic acid and
    then potentiometrically titrated with perchloric acid (0.1 mol/l). 
    1 ml perchloric acid (0.1 mol/l) equiv. 17.365 mg
    dimethylaminophenol hydrochloride.

    8.5.3  Purity

         Unknown

    8.5.4  Methods for analysis of 4-DMAP in biological samples

         Unknown.

    8.6  Shelf-life

         Since, in contact with air, 4-DMAP is readily oxidized and the
    solution changes colour to black-brown, 4-DMAP must be stored in
    opaque containers.  An open ampoule cannot be kept.  Coloured and
    turbid solutions cannot be used.  Correctly stored 4-DMAP can be
    used after storage for up to 3 years.  The influence of tropical
    conditions is unknown.

    8.7  General Properties

         The use of 4-DMAP results in rapid formation of methaemoglobin. 
    An intravenous dose of 3.25 mg/kg resulted in a methaemoglobin
    concentration of 30% within 10 min.  Because there is a much larger
    source of haemoglobin than there is of cytochrome oxidase,
    cytochrome oxidase blocked by cyanide can be reactivated and its
    role as an essential catalyst for tissue utilization of oxygen
    restored.

    8.8  Animal Studies

    8.8.1  In Vitro Studies

         In studies involving the addition of various aminophenols such
    as 2-aminophenol, 2-amino-4-chlorophenol, 2-amino-4,6-
    dichlorophenol, 2-amino-5-ethoxyphenol, 2-dimethylaminophenol,
    4-aminophenol, 4-methylaminophenol, and 4-dimethylaminophenol to the
    blood of mice, cats, dogs, cattle, and man, it has been shown that
    4-dimethylaminophenol and 4-methylaminophenol produce consistent
    amounts of methaemoglobin most rapidly.  These studies also showed

    that a concentration of 30-40% methaemoglobin was attained at widely
    varying rates in different species (Kiese & Weger, 1969).

    8.8.1.1  Metabolism of 4-DMAP in the liver

         In a study by Eyer & Kampffmeyer (1978), nearly all the 4-DMAP
    was conjugated in livers perfused with 4-DMAP (0.01 mmol/l)
    (0.4 nmol/mg protein per min).

    8.8.1.2  Red cell metabolism of 4-DMAP

         4-DMAP rapidly catalyses methaemoglobin formation.  The
    reaction mechanism has been studied using purified human
    haemoglobin.  4-DMAP transfers electrons catalytically from
    ferrohaemoglobin to oxygen (Eyer et al., 1974; Eckert & Eyer, 1983)
    and is thereby oxidized to the phenoxyl-radical and  N,N-dimethyl-
    quinoneimine (Eyer & Lengfelder, 1984).  Both species are reduced to
    ferrohaemoglobin with the formation of methaemoglobin.

         Catalytic methaemoglobin formation is terminated by rapid
    binding of oxidized 4-DMAP to reactive SH groups in haemoglobin and
    thioether formation with reduced glutathione.  Formation of
    glutathione-S-conjugates with 4-DMAP has been shown to occur  in
     vitro in red cells (Eyer & Kiese, 1976), and  in vivo in dogs
    (Eyer & Gaber, 1978) and man (Jansco et al., 198l, Klimmek et al.,
    1983).  Kinetic studies of thioether formation by DMAP in the
    presence of purified dog haemoglobin and glutathione in a reducing
    system showed that  S,S-(2-dimethylamino-5-hydroxy-1,3-phenylene)-
    bis-gluthathione (di(GS)-DMAP) is formed initially.  This thioether
    is highly autoxidizable and the addition of another thiol gives
    stable  S,S,S(2-dimethylamino)-5-hydroxy-1,3,4-phenylene)-
    tri-glutathione (tris-(GS)-DMAP).  Despite its chemical stability,
    it can be actively transported across the red blood cell membrane
    (Eckert & Eyer, 1986).

         4-DMAP is excreted in the urine mainly as the tris-cysteinyl
    derivative, tris(Cys)-DMAP.  This compound is presumed to be broken
    down in the glutamyl cycle of the kidney (Meister, 1973), giving
    rise to cysteinyl glycine and cysteinyl derivatives.

    8.8.1.3  Toxic effects of 4-DMAP on erythrocytes

         Studies have been undertaken to test the effect of 4-DMAP on
    the osmotic resistance of human erythrocytes.  The percentage
    haemolysis did not differ from that of controls, even for an
    incubation period of less than 20 h, at a 4-DMAP concentration of
    1 mmol/l.  However, the degree of haemolysis after 20 h was 14.5%
    (±6.9%), compared with a control value of 2.7% (±0.4%), when
    measured in 0.7% NaCl solution (Klimmek et al., 1983).  Since the
    haemolysis that occurred in the tests for osmotic resistance was
    maximal after the same period as in  in vivo studies, one may

    assume a common lytic mechanism under  in vitro and  in vivo
    conditions (Klimmek et al., 1983).

    8.8.1.4  Toxic effects of 4-DMAP on isolated rat kidney tubules

         Various toxic effects were observed when 4-DMAP was added to
    suspensions of isolated rat kidney tubules, apparently as the result
    of the irreversible binding of 4-DMAP to SH-containing compounds,
    e.g., multiple enzyme inhibition, inhibition of gluconeogenesis,
    decrease in glutathione, and membrane damage (Snizicz et al., 1979). 
    The results suggest that irreversible impairment of the integrity of
    the cell membrane causes irreversible damage to tubular epithelial
    cells, the resultant increase in permeability being followed by a
    decrease in nucleotide content without marked effect on
    gluconeogenesis or mitochondrial membrane integrity.

    8.8.1.5  Oxygen saturation and methaemoglobin formation

         From reports by Kiese (1974) and Eyer et al. (1979), it is
    evident that quite small variations of pO2 can cause large
    differences in the rate of methaemoglobin formation.  At high
    percentage oxyhaemoglobin levels, where the R-configuration of
    haemoglobin prevails over the T-configuration, the rate of oxidation
    is lower.  At a 4-DMAP concentration of 0.032 mmol/l, increasing the
    percentage saturation of haemoglobin of humans more than halved the
    rate of methaemoglobin production (Marrs et al., 1982).

    8.8.2  Pharmacokinetics

         No data are available.

    8.8.3  Pharmacodynamics

         4-DMAP (3.25 mg/kg), given intravenously to dogs one min after
    poisoning with a lethal dose of KCN (4 mg/kg), resulted in the
    survival of all dogs.  The maximum venous methaemoglobin content was
    reached within 5-10 min and was 38.8% (±1.7%) of the total
    haemoglobin.  The injection of the same 4-DMAP dose intramuscularly
    led to a maximum methaemoglobin level of 41.6% (±1.3%) after 30 min. 
    A methaemoglobin concentration of 35% was achieved within 5 min of
    an intravenous dose of 3.25 mg/kg, 15 min of an intramuscular dose
    of 3.25 mg/kg, and 30 min of an oral dose of 15 mg/kg (Klimmek et
    al., 1983).

         Bright & Marrs (1982) found that there was considerable
    variation in the methaemoglobin levels produced in different dogs
    using the same oral dose of 4-DMAP.

         In experiments using different aminophenols, it was shown that
    4-DMAP given intravenously rapidly produced consistent amounts of
    methaemoglobin in the blood of different animals species.  In mice

    and rabbits, a rapid fall in the concentration of methaemoglobin was
    observed after 2 min and after 20 min the concentration in these
    species was less than 10%.  In dogs and mice, however, the decrease
    in methaemoglobin concentration was substantially less; after
    20 min, the level was still above 30%.  In contrast to 4-DMAP,
    sodium nitrite caused increased methaemoglobin levels which lasted
    longer than an hour in mice and rabbits (Kiese & Weger, 1969).

         There was a marked increase in arterial pO2 after the
    administration of 4-DMAP to dogs poisoned with cyanide, compared
    with non-poisoned dogs.  Presumably, part of the oxygen released
    during the formation of methaemoglobin under hypoxic conditions can
    be utilized by tissues (Klimmek et al., 1979a,b).  In the same
    experiments, an increase in respiratory volume and mean arterial
    blood pressure, was observed together with a lessening of the
    lactate-pyruvate ratio.

    8.8.4  Toxicology

         LD50 intravenous mouse:    50-70 mg/kg (Kiese & Weger, 1965;
                                    Marrs et al., 1984)

         LD50 oral mouse:           946 mg/kg (Marrs et al., 1984)

         LD50 intravenous rat:      57 mg/kg (Kiese et al., 1975)

         LD50 oral rat:             689-780 mg/kg (Marrs et al., 1984)

         LD50 intravenous dog:      26 mg/kg (Weger, 1975)

         In all five studies, the cause of death was inadequate oxygen
    transport by the blood to the tissues because 85% of the haemoglobin
    had been converted to methaemoglobin.

    8.8.4.1  Nephrotoxicity

         Some of the aminophenols, such as 4-aminophenol, produce kidney
    lesions after intravenous administrations (Green et al., 1969;
    Calder et al., 1971; Hinsberg & Treupel, 1984).  The intravenous
    injection of 100 mg 4-DMAP/kg, a dose nearly double the LD50 and
    which oxidizes 60% of the haemoglobin to methaemoglobin, caused
    renal damage in rats.  Moderate to severe necrosis of the convoluted
    tubules, with either few or no inflammatory cells, was found 24 h
    after injection.  The glomeruli were not affected and there was no
    papillary damage.  However, no tubular necrosis was detected with a
    dose of 30 mg/kg.  A high or low sodium diet did not noticeably
    affect the nephrotoxicity of 4-DMAP (Kiese et al., l975).

         A single oral dose of up to 25 mg 4-DMAP/kg did not result in
    any macroscopic or histological abnormality in the gastrointestinal
    tract, liver, or kidneys of rats killed up to 24 h after dosing.  In
    animals that had been kept for 7 days after dosing, no histological
    abnormalities were found other than engorgement of the liver and
    kidneys (Marrs et al., 1982).

    8.8.4.2  Mutagenicity

         A clear relationship between dose and mutation frequency was
    shown by a mammalian cell assay employing V79 (Chinese hamster)
    cells (Lee & Webber, 1983), despite a negative Ames test with and
    without metabolic activation.

    8.9  Volunteer Studies

         In experiments by Weger (1968) on seven volunteers, an
    intravenous dose of  N,N-dimethyl- p-aminophenol (3.25 mg/kg)
    resulted in a 15% methaemoglobin concentration within 1 min.  After
    10 min, the concentration was 30%.  Other aminophenols tested gave
    less favourable results.   N-methyl- p-aminophenol had much the
    same effect but only at an iv dose of 20 mg/kg.   O-aminophenol
    (30 mg/kg) administered intravenously resulted in 15%
    methaemoglobinaemia after 12 min and 30% after 60 min.  Comparison
    was also made with sodium nitrite. When a dose of 4 mg/kg was given
    intravenously to seven volunteers, the maximal methaemoglobin level
    was 6%.  This dose of sodium nitrite was that recommended by Chen et
    al. (l933) and Chen & Rose (l952) and suggests that the antidotal
    effect may be due to a mechanism other than methaemoglobin
    formation.

         In experiments by Klimmek et al. (1983) on volunteers a
    methaemoglobin concentration of about 30% was reached within 5 min
    after an intravenous dose of 3.25 mg 4-DMAP/kg (3 volunteers),
    50 min after an intramuscular dose of 3.5 mg 4-DMAP/kg
    (6 volunteers), and 30 min after an oral dose of 900 mg 4-DMAP
    (5 volunteers).

    8.9.1  Metabolism of 4-DMAP in the liver

         The metabolic conversion of 4-DMAP to glucuronide, sulfate, and
    thioether derivatives is likely to follow linear Michaelis-Menten
    kinetics.  The pattern of metabolites was studied in urine collected
    over 24 hours from subjects treated with intravenous 4- DMAP
    (3.25 mg/kg).  An average of 68% of the dose was excreted as
    metabolites in urine:  41% as glucuronide, 12% as sulfate, and 15%
    as thioethers (Klimmek et al., 1983).

         4-DMAP rapidly auto-oxidizes at pH values above 7 and is
    transformed to a variety of degradation products (Eyer et al.,
    1974).  Thus partial degradation of 4-DMAP in the gastrointestinal
    tract may account for the reduced total metabolite recovery
    following oral administration compared with that following
    intravenous administration.

    8.9.2  Metabolism of 4-DMAP in erythrocytes

         Rapid thioether formation following the intravenous
    administration of 4-DMAP has been shown to occur in erythrocytes
    where 4-DMAP is oxidized by oxyhaemoglobin to the corresponding
    quinoneimine which yields adducts with red cell glutathione (Eyer &
    Kiese, 1976).  One of these, tris-(GS)-DMAP penetrates the red cell
    membrane slowly and is excreted in the urine mainly as a
    tris-cysteinyl derivative, tris-(Cys)-DMAP.

         The 4-DMAP thioethers excreted in human urine (15.3% (±1.8%) of
    the dose) are 1ikely to originate in the erythrocytes;  this
    underlines the importance of erythrocytes in 4-DMAP metabolism in
    man.

    8.9.3  Adverse effects

         After 6-7 days, a phlebitis was observed in the antecubital
    vein where 4-DMAP was infused (Klimmek et al., 1983).  Following an
    intramuscular injection of 4-DMAP, a slight pressure was felt after
    5-10 min at the site of injection, slowly growing in intensity and
    finally resulting in severe pain.  Approximately 10 h after the
    injection, shivering, sweating, and fever occurred.  An oral 4-DMAP
    dose of 300 mg was well tolerated, whereas 600 mg slowed mental
    activity and caused tiredness, and 900 mg gave an effect of
    dizziness, with headache and buzzing in the ears (Klimmek et al.,
    1983).

         After an intravenous injection of 4-DMAP (3.25 mg/kg), the
    total bilirubin concentration increased by 140%, that of conjugated
    bilirubin by 180% and that of iron by 200%.  Within 24 h of an
    intramuscular injection of 4-DMAP (3.5 mg/kg), the total bilirubin
    increased by 270% and then declined rapidly, while the bilirubin
    concentration rose by 120% and that of iron by 50% (Klimmek et al.,
    1983).

         After an oral dose of 900 mg 4-DMAP, total and conjugated
    bilirubin levels increased by 170% and 80%, respectively, and the
    iron concentration rose by 60%.  The changes in bilirubin and iron
    concentrations were the result of reduced erythrocyte survival
    caused by 4-DMAP, which took about one day to develop.  Since the
    haemolysis that occurred in the tests for osmotic resistance was
    maximal after the same period as in  in vivo studies, one may

    assume a common lytic mechanism both under  in vitro and  in vivo
    conditions (Klimmek et al., 1983).

    8.10  Clinical Studies

         No data are available.

    8.11  Clinical Studies - Case Reports

         Table 6 summarizes clinical observations made in 19 cases of
    cyanide poisoning treated by 4-DMAP.  In most patients, there was a
    long time interval between the occurrence of poisoning and the
    administration of 4-DMAP.  Patients with impaired vital function at
    the time of treatment recovered.  In six cases an overdose of 4-DMAP
    was given, resulting in excessive methaemoglobinaemia, and in three
    cases haemolysis was observed on the second day after administration
    of the antidote.  Even the recommended dose of 3.25 mg 4-DMAP/kg has
    resulted in a methaemoglobin concentration of about 70% (van Dijk et
    al., 1987).

    8.12  Summary of Evaluation

    8.12.1  Indications

         4-DMAP should only be given to patients poisoned with cyanide
    who are in a deep coma and who have dilated non-reactive pupils and
    deteriorating cardio-respiratory function.

    8.12.2  Recommended routes and dosage

         An intravenous dose of 3.25 mg/kg body weight should be given
    immediately.  The intramuscular injection of 4-DMAP to patients in
    shock due to cyanide poisoning cannot be recommended because
    absorption is unpredictable.

    8.12.3  Other consequential or supportive therapy

         The administration of 4-DMAP should always be followed by that
    of sodium thiosulfate (150-200 mg/kg intravenously) (see chapter 3). 
    Artificial respiration should be given if necessary.


        Table 6.  Clinical Observations in cyanide-poisoned patients treated by 4-DMAPa

    Estimated   Interval before  Blood cyanide   Interval before     4-DMAP dose,     Additional  Adverse   Outcome    Reference
     cyanide       first-aid     concentration  blood measurement  interval before    antidotes   effects
    ingested      treatmentb        (mg/l)            usede        administrationf

                                                                                                                                              

     10 g oral      15 min             ?                            250 mg, 45 min    CO2-EDTA              recovery   Daunderer et al. (1974)

         ?          < 5 min          5-16.9           5 min          250 mg, 5 min                          CNS        Jacobs (1984)
                                                                                                            damage

         ?          1.5-2 h            c              5.5 h          250 mg, > 5h                           recovery   Von Clarmann (1987)

         ?          20 min            NR               NR            250 mg, 3.5h                           recovery   Von Clarmann (1987)

         ?           < 1h             1.9            > 24 h          250 mg, < 1 h    CO2-EDTA    met.Hb    death      Von Clarmann (1987)
                                                                                       900 mg     69-77&

         ?           < 1 h            ND              27 h            2 x 250 mg,     CO2-EDTA              death      Von Clarmann (1987)
                                                                     1 h and 27 h      7.5 mg

    50-150 mg?       < 1 h            1.7            20 min         250 mg, 20 min                          recovery   Von Clarmann (1987)

         ?           <1 h             9.2          measurement        ?, < 20 min                 met.Hb    death      Von Clarmann (1987)
                                                   at autopsy                                       77%

       10 g           2h               d               NR            250 mg, > 2h                           recovery   Von Clarmann (1987)

         ?          20 min             d               NR           750 mg, 20 min                          death      Von Clarmann (1987)

      320 mg         8 min           0.24             > 1 h           2 x 250 mg,                           recovery   Von Clarmann (1987)
                                                                    10 min and 1 h
                                                                                                                                              

    Table 6.  Contd.

    Estimated   Interval before  Blood cyanide   Interval before     4-DMAP dose,     Additional  Adverse   Outcome    Reference
     cyanide       first-aid     concentration  blood measurement  interval before    antidotes   effects
    ingested      treatmentb        (mg/l)            usede        administrationf

                                                                                                                                              

         ?          10 min            70             30 min           2 x 250 mg,                           death      Von Clarmann (1987)
                                                                   10 min and 30 min

         ?          20 min            24              1.5 h          250 mg, 1.5j;                          death      Von Clarmann (1987)

         ?             ?              37              < 1 h             1250 mg                 haemolysis  recovery   Von Clarmann (1987)

        1 g           1 h              6              > 1 h           500 mg, 1 h                 met. Hb   recovery   Von Clarmann (1987)

       25 g            ?             10.5             2.5 h          250 mg, > 3 h    CO2-EDTA              recovery   Von Clarmann (1987)
                                                                                       300 mg

         ?          15 min            7.5              1 h            250 mg, 4 h       NaNO2     met.Hb    death      Van Dijk et al. (1987)

         ?          15 min            NR               NR           1000 mg, 30 min   hydroxo-    met.Hb    recovery   Van Heijst et al. (1987)
                                                                                      cobalamin  haemolysis
                                                                                                                                              

    a met.Hb = methaemoblobin; ? = unknown; NR = not recorded; ND = not detectable.
    b Interval between cyanide exposure and beginning of first-aid treatment.
    c Cyanide was present in the expired air but no measurement of blood concentration was made.
    d Positive in qualitative tests.
    e Interval between cyanide exposure and measurement of blood cyanide concentration.
    f Interval between cyanide exposure and commencement of 4-DMAP treatment.
    

    8.12.4  Areas of use where there is insufficient information to make
            recommendations

         There is insufficient evidence regarding the efficacy of the
    prophylactic use of oral 4-DMAP, e.g., with rescue personnel.

    8.13  Model Information Sheet

    8.13.1  Uses

         4-DMAP is indicated for the treatment of severe cyanide
    poisoning in patients who are in deep coma and who have dilated
    non-reactive pupils and deteriorating cardio-respiratory function.

    8.13.2  Dosage and route

         An intravenous dose of 3.25 mg/kg body weight should be given.

    8.13.3  Precautions/contraindications

         There are differences in individual susceptibility, which may
    result in an unacceptably high level of methaemoglobin after normal
    therapeutic doses (van Dijk et al., 1987).  Excess
    methaemoglobinaemia may be corrected with either methylene or
    toluidine blue, but this will result in the release of cyanide;
    exchange transfusion is an alternative approach if this is
    practicable.  Determination of methaemoglobin will not allow the
    amount of haemoglobin available for oxygen transport to be
    calculated, because the cyanmethaemoglobin concentration will be
    unknown and, at the present time, cannot be analysed.

         Treatment with 4-DMAP is contraindicated in patients with G6PD
    deficiency.

    8.13.4  Adverse effects

         Haemolysis is observed following an overdose.  Mild headache,
    dizziness, hyperventilation, cyanosis, and green/brown
    discolouration of the urine have also been observed.  Haemolysis may
    occur even in people with normal erythrocytes given an appropriate
    therapeutic dose.

    8.13.5  Use in pregnancy and lactation

         No information is available.

    8.13.6  Storage

         4-DMAP must be stored in opaque containers.  The maximum
    storage time is 3 years.

    8.14  References

    Bright JE & Marrs TC (1982) A comparison of the
    methemoglobin-inducing activity of 4-DMAP and p-aminopropriophenone.
    Toxicol Lett, 13(1-2): 81-86.

    Calder IC, Funder CC, Green CR, Ham K, & Tange JD (1971) Comparative
    nephrotoxicity of aspirin and phenacetin derivatives. Br Med J, 4:
    518-521.

    Chen KK & Rose CL (1952) Nitrite and thiosulfate therapy in cyanide
    poisoning. J Am Med Assoc, 149: 113-119.

    Chen KK, Rose CL, & Clowes GHA (1933) Amyl nitrite and cyanide
    poisoning. J Am Med Assoc, 100: 1920-1922.

    Daunderer H, Theml H, & Weger N (1974) [Treatment of prussic acid
    poisoning with 4-dimethylaminophenol (4-DMAP).] Med Klin, 69:
    1626-1631 (in German).

    Eckert KG & Eyer P (1983) Differences in the reaction of isomeric
    ortho- and para-aminophenols with hemoglobin. Biochem Pharmacol, 32:
    1019-1027.

    Eckert KG & Eyer P (1986) Formation and transport of xenobiotic
    glutathion-S-conjugates in red cells. Biochem Pharmacol, 35:
    325-329.

    Eyer P & Gaber H (1978) Biotransformation of 4-dimethylaminophenol
    in the dog. Biochem Pharmacol, 27: 2215-2221.

    Eyer P & Kampffmeyer H (1978) Biotransformation of
    4-dimethylaminophenol in the isolated perfused rat liver. Biochem
    Pharmacol, 27: 2223-2228.

    Eyer P & Kiese M (1976) Biotransformation of 4-dimethylaminophenol;
    reaction with glutathion and some properties of the reaction
    products. Chem-Biol Interact, 14: 165-178.

    Eyer P & Lengfelder E (1984) Radical formation during autooxidation
    of 4-dimethylaminophenol and some properties of the reaction
    product. Biochem Pharmacol, 33: 1005-1013.

    Eyer P, Kiese M, Lipowsky G, & Weger N (1974) Reactions of
    4-dimethylaminophenol with hemoglobin, and autooxidation of
    4-dimethylaminophenol. Chem-Biol Interact, 8: 41-59.

    Eyer P, Herle H, Kiese M, & Kluin G (1979) Kinetics of
    ferrihaemoglobin formation by some reducing agents and the role of
    hydrogen peroxide. Mol Pharmacol, 11: 326-334.

    Green CR, Ham KN, & Tange JD (1969) Kidney lesions induced in rats
    by p-aminophenol. Br Med J, 1: 162-164.

    Hinsberg O & Treupel G (1984) [The physiological effects of
    p-aminophenol and some of its derivatives.] Naunyn-Schmiedebergs
    Arch Exp Pathol Pharmakol, 33:  216-250 (in German).

    Jacobs K (1984) [Report on experience with the administration of
    4-DMAP in severe prussic acid poisoning.  Consequences for medical
    practice.] Zentralbl Arbeitsmed, 34: 274-277 (in German).

    Jansco L, Szinicz L, & Eyer P (1981) Biotransformation of
    4-dimethylaminophenol in man. Arch Toxicol, 47: 39-45.

    Kiese M (1974) Methaemoglobinaemia, a comprehensive treatise.
    Cleveland, Ohio, CRC Press, Inc.

    Kiese M & Weger N (1965) The treatment of experimental cyanide
    poisoning by hemoglobin formation. Arch Toxicol, 21: 89-100.

    Kiese M & Weger N (1969) Formation of ferrihemoglobin with
    aminophenols in the human for the treatment of cyanide poisoning.
    Eur J Pharmacol, 7: 97-105.

    Kiese M, Szinick L, Thiel N, & Weger N (1975) Ferrihemoglobin and
    kidney lesions in rats produced by 4-aminophenol or
    4-dimethylaminophenol. Arch Toxicol, 34: 337-340.

    Klimmek R, Fladerer H, & Weger N (1979a) Circulation, respiration
    and blood homeostasis in cyanide poisoned dogs after treatments with
    4-aminophenol or cobalt compounds. Arch Toxicol, 42: 121-133.

    Klimmek R, Fladerer H, Szinicz L, Weger N, & Kriese M (1979b)
    Effects of 4-dimethylaminophenol and CO2 EDTA on circulation,
    respiration, and blood homeostasis in dogs. Arch Toxicol, 42: 75-84.

    Klimmek R, Krettek C, Szinicz L, Eyer R, & Weger N (1983) Effects of
    biotransformation of 4-dimethylaminophenol in man and dogs. Arch
    Toxicol, 53: 275-288.

    Lee CG & Webber TD (1983) The mutagenicity of a cyanide antidote,
    dimethylaminophenol, in Chinese hamster cells. Toxicol Lett, 16:
    275-288.

    Marrs TC, Bright JE, & Swanson DW (1982) The effects of prior
    treatment with 4-dimethylaminophenol on animals experimentally
    poisoned with hydrogen cyanide. Arch Toxicol, 51: 247-253.

    Marrs TC, Scawin J, & Swanson DW (1984) The acute intravenous and
    oral toxicity in mice, rats and guinea pigs of 4-DMAP and its
    effects on haematological variables. Toxicology, 31: 165-173.

    Meister A (1973) On the enzymology of amino acid transport. Science,
    180: 33-39.

    Szizicz L, Weger N, Schneiderhan W, & Kiese M (1979) Nephrotoxicity
    of aminophenols: Effects of the 4-dimethylaminophenol on isolated
    rat kidney tubules. Arch Toxicol, 42: 63-73.

    Van Dijk A, Glerum JH, Van Heijst CNP, & Douze JMC (1987) Clinical
    evaluation of the cyanide antagonist 4-DMAP in a lethal cyanide
    poisoning case. Vet Hum Toxicol, 29(suppl 2): 38-39.

    Van Heijst ANP, Douze JMC, Van Kesteren RG, Bergen JEAM, & Van Dijk
    A (1987) Therapeutic problems in cyanide poisoning. Clin Toxicol,
    25: 383-398.

    Von Clarmann M (1987) 4-dimethylaminophenol. Communication at the
    joint meeting of the EAPCC, CEC, IPCS and the National Poison
    Control Centre in the Netherlands, Utrecht.

    Weger N (1968) [Aminophenols as antidotes to prussic acid.] Arch
    Toxikol, 24: 49-50 (in German).

    Weger N (1975) [Cyanide poisoning and therapy.] Wehrmed Monatschr,
    19(1): 6-11 (in German with English summary).

    9.  Methylene Blue and Toluidine Blue

    9.1  Methylene Blue

    9.1.1  Introduction

         Methylene blue is a redox dye that has been used in clinical
    medicine for approximately 100 years.  Its present uses are based on
    its tissue-staining properties and its oxidative-reductive capacity
    (Bodansky & Gutman, 1947; Blass & Fung, 1976; Metz et al., 1976). 
    Administered locally it has been used for the detection of fistulae
    and for the recognition of ruptured amniotic membranes in obstetrics
    (Sparhr & Salisbury, 1980; Martindale, 1989; Windholz, 1983).  It
    may also be administered systemically to assist the detection of
    endocrine tissue, e.g., on parathyroid glands or pancreatic adenoma
    during an operation (Blass & Fung, 1976; Whitman et al., 1979;
    Martindale, 1989).

         Methylene blue has long been recognized as an effective
    antidote for methaemoglobinaemia in man and in domestic animals
    (Etteldorf, 1951; Beutler & Baluda, 1963; Burrows et al., 1977;
    Gosselin et al., 1984).  Injected systemically in low doses, its
    reducing action is utilized in the treatment of toxic
    methaemoglobinaemia.

         Methylene blue, however, is relatively ineffective against
    toxic methaemoglobinaemia in individuals with glucose-6-phosphate
    dehydrogenase deficiency and in chlorate poisoning (Bodansky &
    Gutman, 1947; Metz et al., 1976; Gosselin et al., 1984).

    9.1.2  Name and chemical formula of antidote

         3,7-bis(dimethylamino)-phenazothionium chloride trihydrate,
              tetramethylthionine chloride trihydrate

         Synonyms:  Methylioninii chloridum, Methylenum caerulum, Azul
              de Metileno, Swiss blue, Blu di metilene, Schultz No.
              1038, CI Classic Blue 9.

         Colour index no: 52015

    FIGURE 07

         Molecular formula:  C16H18C1N3S.3H2O (Martindale, 1989;
                             Windholz, 1983)

         Relative molecular mass:  373.9

         CAS number: 7220-79-3 (trihydrate) and 67-73-4 (anhydrous)

              Commercial methylene blue is the double chloride of
              tetramethylthionine and zinc and is not suitable for
              medicinal use (Windholz, 1983).

    9.1.3  Physico-chemical properties

         Methylene blue is a dark green, almost odourless, hygroscopic
    crystalline powder with a bronze-like lustre.  It loses 8-18% of its
    weight on drying.  One gram dissolves in about 25 ml of water, in
    about 65 ml alcohol, and in 450 ml of chloroform.  Methylene blue is
    insoluble in ether (Martindale, 1989;  Windholz, 1983).

         A 1% solution in water has a pH of 3 to 4.5.  Solutions in
    water are intensely blue-coloured and incompatible with caustic
    alkalis, oxidising and reducing substances and iodides (Martindale,
    1989).

         Solutions are sterilised by autoclaving or by filtration and
    should be stored in airtight containers (Windholz, 1983).

         Data on melting and boiling points and on stability in light
    are not available.

         Excipients:  water for injection.

    9.1.4  Synthesis

         First prepared by Caro in 1876, methylene blue is produced by
    the so-called "thiosulfate process" in which a mixture of
    dimethyl- p-phenylenediamine and dimethylaniline is oxidized,
    usually with potassium dichromate, in the presence of sodium
    thiosulfate and zinc chloride (Windholz, 1983).

    9.1.5  Analytical methods

         No information on analytical methods for methylene blue is
    available.

    9.1.6  Shelf-life

         No data about specific conditions of temperature and humidity
    are available, and the shelf-life is unknown.  Methylene blue should
    be stored in airtight containers.

    9.1.7  General properties

         Methylene blue functions as an intermediate in the transfer of
    electrons from pyridine nucleotides to a suitable electron acceptor,
    thereby stimulating the hexose monophosphate shunt (HMPS) pathway in
    a variety of cell systems (Bodansky & Gutman, 1947; Beutler &
    Baluda, 1963; Smith & Olson, 1973; Metz et al., 1976).

         In the erythrocyte, methylene blue is reduced to leucomethylene
    blue primarily by NADPH-dependent diaphorase (dihydrolipoamide
    dehydrogenase).  This diaphorase is reduced via oxidation of NADPH,
    which in turn stimulates the HMPS, and leucomethylene blue transfers
    electrons to an acceptor such as methaemoglobin (Bodansky & Gutman,
    1947; Metz et al., 1976).  This series of reactions can be used
    clinically for the reduction of ferric to ferrous haem iron in
    patients with acquired methaemoglobinaemia.  The effectiveness of
    methylene blue in reducing methaemoglobin is well established
    (Etteldorf, 1951; Beutler & Baluda, 1963; Smith & Olson, 1973;
    Harrison, 1977; Gosselin et al., 1984).

    9.1.8  Animal studies

         Several animal studies confirm that the administration of
    methylene blue protects against death caused by
    methaemoglobin-generating agents (Blass & Fung, 1976; Burrows et
    al., 1977; Hrushesky et al., 1985).  Burrows et al. (1977)
    administered sodium nitrite (50 mg/kg) intervenously to four ewes. 
    This resulted in the formation of 70-80% methaemoglobinaemia (a
    lethal level) within 45 min.  The mortality associated with this
    dose of sodium nitrite was successfully antagonized by the
    administration of intravenous methylene blue (2.2 mg/kg) 30 min
    later.

         Pharmacological studies have been carried out in dogs and rats
    (Blass & Fung, 1976).  When intravenous boluses as large as 15 mg/kg
    were given, the resultant data could be explained on the basis of a
    one-compartment model with binding of methylene blue to tissues.

         Few data on mutagenicity or teratogenicity testing are
    available.

    9.1.9  Volunteer studies

         No data on volunteer studies are available.

    9.1.10  Clinical studies

         The clinical symptoms and signs associated with
    methaemoglobinaemia depend on the percentage of haemoglobin oxidized
    to the methaemoglobin form (Bodansky & Gutman, 1947).  They consist
    of greyish-blue cyanosis, without signs of cardiac or pulmonary

    distress, which becomes apparent at about 15% methaemoglobinaemia. 
    Hypoxaemia may be accompanied by dyspnoea, dizziness, headache,
    weakness, lethargy, and CNS depression, the severity of which will
    increase with increasing concentrations of methaemoglobin.

         An intense chocolate-brown-coloured blood and a central
    cyanosis that does not respond to the administration of 100% oxygen
    suggest methaemoglobinaemia (Smith  & Olson, 1973; Gosselin et al.,
    1984; Hall et al., 1986).  Methaemoglobin levels may be directly
    measured using a spectophotometric method (Hall et al., 1986).

         Many authors agree that in asymptomatic patients, with
    methaemoglobin levels of 30% or less, methylene blue administration
    is not necessary (Bodansky & Gutman, 1947; Gosselin et al., 1984;
    Hall et al., 1986).

    9.1.11  Clinical studies - case reports

         Ng et al. (1982) described a case of methaemoglobinaemia and
    haemolysis resulting from the ingestion of paraquat by a 32-year-old
    patient. He was administered methylene blue (1 mg/kg) intravenously,
    with dramatic results: cyanosis was reversed and the patient became
    alert and cooperative.  The methaemoglobin level was 19.7% before
    methylene blue was given and 1.6% after.

         A single case of dapsone-induced methaemoglobinaemia was
    treated by continuous methylene blue infusion (Berlin et al., 1985). 
    The patient recovered completely.

    9.1.12  Summary of evaluation

    9.1.12.1  Indications

         Methylene blue is the agent of choice for accelerating the
    reduction of methaemoglobin, at toxic levels, induced by aniline and
    aminophenols, nitrites and nitrates, nitrobenzene, antimalarial
    agents (chloroquine, primaquine), dapsone, local anaesthetics
    (lidocaine, prilocaine, benzocaine), phenacetin, phenazopyridin, and
    sulfamidines.  It should be used with special caution in the
    correction of excess methaemoglobinaemia induced by the treatment of
    cyanide poisoning with nitrites.

    9.1.12.2  Advised route and dosage

         Methylene blue should be administered as an intravenous dose of
    1-2 mg/kg body weight (0.1 to 0.2 ml/kg of a 1% solution), slowly
    over 5 to 10 min.  An improvement should be noted within 30-60 min
    of administration.  If cyanosis has not disappeared within 1 h, a
    second dose should be given (Bodansky & Gutman, 1947; Etteldorf,
    1951; Smith & Olson, 1973; Harrison, 1977;  Hall et al., 1986).

    9.1.12.3  Precautions and contraindications

         Methylene blue should not be given by subcutaneous or
    intrathecal injection because it causes necrotic abscess formation
    (Whitman et al., 1979; Windholz, 1983; Gosselin et al., 1984).  It
    should be administered with caution in patients with severe renal
    impairment (Windholz, 1983; Martindale, 1989; Gosselin et al.,
    1984).

         In patients with glucose-6-phosphate dehydrogenase deficiency
    (who do not generate NADPH), methylene blue is likely to be
    ineffective and has been reported to cause haemolytic anaemia
    (Beutler & Baluda, 1963; Rosen et al., 1971; Whitman, et al., 1979;
    Martindale, 1989).

         Methylene blue should not be used in chlorate-induced
    methaemoglobinaemia because it may enhance the toxicity of the
    chlorate salts (Metz et al., 1976; Martindale, 1989).

    9.1.12.4  Adverse effects

         Nausea, abdominal and chest pain, headache, dizziness, mental
    confusion and profuse sweating may occur following large intravenous
    doses of methylene blue (Naidler et al., 1945; Martindale, 1989;
    Windholz, 1983).

         Whitman et al. (1979) described a 28-year-old patient who
    received methylene blue (5 mg/kg) for the identification of an
    insulin-secreting pancreatic adenoma.  During surgery, the patient
    developed methaemoglobin levels of up to 7.1%.

         Sparhr & Salisbury (1980) reported a case where an
    intra-amniotic injection of 10 mg of methylene blue was given in a
    pregnant woman into the vitelline sac of one of two twins.  On the
    second day, the new-born baby developed a methaemoglobin level of
    20%.  Physical examination revealed a bluish-green skin colour and
    respiratory distress and Heinz bodies were seen in the peripheral
    blood.  Other reports have stressed the neonatal morbidity related
    to the use of the dye in obstetrics (Vincer et al., 1987).

         Blass & Fung (1976) reported the case of a 4-year-old boy who
    was given 1 g methylene blue intravenously during surgery.  He
    developed hypotension, tachycardia, and deep cyanosis and remained
    intensely blue for several days.

         Methylene blue may impart a blue-green colour to urine and
    faeces (Martindale, 1989; Windholz, 1983).

    9.1.12.5  Other consequential or supportive theory

         See chapter 1.

    9.1.13  Model information sheet

    9.1.13.1  Uses

         Methylene blue is used for the treatment of drug-induced, and
    some forms of idiopathic, methaemoglobinaemia.  It should be used
    with great caution in the correction of nitrite-induced
    methaemoglobinaemia arising from urgent treatment of cyanide
    poisoning.

         It is also used as a dye for the identification of fistulae and
    glandular tissues.

    9.1.13.2  Dosage and route of administration

         A dose of 1-2 mg/kg (0.1-0.2 ml/kg of a 1% solution) should be
    administered intravenously over 5-10 min.

    9.1.13.3  Precautions and contraindications

         Methylene blue may be administered when methaemoglobin levels
    resulting from nitrite administration are more than 30-40%. It has
    no value in the treatment of cyanide poisoning.

         Regeneration of haemoglobin from methaemoglobin will release
    cyanide back into the circulation.

    9.1.13.4  Adverse effects

         Large intravenous doses of methylene blue produce nausea,
    abdominal and chest-pain, headache, dizziness, mental confusion, and
    profuse sweating.  Intravascular haemolysis, which may be
    life-threatening, can also occur.

    9.1.13.5  Use in pregnancy/lactation

         Hyperbilirubinaemia and haemolysis may be seen in the neonate.

    9.1.13.6  Storage

              Methylene blue should be stored in airtight containers.

    9.1.14  References

    Berlin G, Brodin B, & Milden JO (1985) Acute dapsone intoxication: a
    case treated with continuous infusion of methylene blue, forced
    diuresis and plasma exchange. Clin Toxicol, 22: 537-548.

    Beutler B & Baluda M (1963) Methemoglobin reduction: studies on the
    interaction between cell populations and of the role of methylene
    blue. Blood, 22: 323.

    Blass N & Fung D (1976) Dyed but not dead - Methylene blue overdose.
    Anesthesiology, 45: 458-459.

    Bodansky O & Gutman H (1947) Treatment of methemoglobinemia. J
    Pharmacol Exp Ther, 89: 45-46.

    Burrows GM, Williams A, Davis A, & Way JI (1977) Antagonism of
    nitrite toxicity by methylene blue or toluidine chloride. Proc West
    Pharmacol Soc, 20: 439-443.

    Etteldorf JN (1951) Methylene blue in treatment of methemoglobinemia
    in premature infants caused by marking ink: 8 cases reported. J
    Pediatr, 38: 24-27.

    Gosselin RE, Smith HP, & Hodge HC ed. (1984) Nitrite. In: Clinical
    toxicology of commercial products, 5th ed. Baltimore, Maryland,
    Williams & Wilkins Co., pp III/314-III/319.

    Hall AH, Konig MW, & Rumack BH (1986) Drug and chemical-induced
    methemoglobinemia. Clinical features and treatment. Med Toxicol, 1:
    253-260.

    Harrison MR (1977) Toxic methemoglobinemia. A case of acute
    nitrobenzene and aniline poisoning. Anaesthesia, 32: 270-272.

    Hrushesky WJ, Olshefski R, Wood O, Meshnick S, & Eaton JW (1985)
    Modifying intracellular redox balance: an approach to improving
    therapeutic index. Lancet, 1: 565-567.

    Martindale (1989) In: Reynolds JEF ed. The extra pharmacopoeia, 29th
    ed. London, The Pharmaceutical Press, pp 843-844.

    Metz, E, Balcerzak S, & Sacome A (1976) Mechanism of methylene blue
    stimulation of the hexose-monophosphate shunt in erythrocytes. J
    Clin Invest, 58(4): 797-802.

    Naidler JE, Green H, & Rosenbaum A (1945) Intravenous injection of
    methylene blue in man with reference to its toxic symptoms and
    effect on the electrocardiogram. Am J. Med Sci, 188: 15-21.

    Ng LL, Naik RB, & Polak A (1982) Paraquat ingestion with
    methemoglobinemia treated with methylene blue. Br J Med, 284:
    1445-1446.

    Rosen PS, Johnson G, McGehee SG, & Beutler E (1971) Failure of
    methylene blue treatment in toxic methemoglobinemia. Ann Intern Med,
    75: 83-86.

    Smith R & Olson M (1973) Drug-induced methemoglobinemia. Semin
    Hematol, 10: 253-260.

    Sparhr RC & Salisbury DJ (1980) Intra-amniotic injection of
    methylene blue leading to methemoglobinemia in one of twins. Int J
    Gynaecol Obstet, 17: 477-478.

    Vincer MJ, Allen AC, Evans JR, Nwaesei C, & Stinson DA (1987)
    Methylene-blue-induced hemolytic anemia in a neonate. Can Med Assoc
    J, 136: 503-4.

    Whitman JG, Taylor AR, & White JM (1979) Potential hazard of
    methylene blue. Anaesthesia, 34: 181-182.

    Windholz M ed. (1983) The Merck index: An encyclopedia of chemicals,
    drugs and biologicals, 10th ed. Rahway, New Jersey, Merck and Co.,
    Inc. pp 333-334.

    9.2  Toluidine Blue

    9.2.1  Introduction

         Toluidine blue is one of a group of dyes that can be used to
    treat methaemoglobinaemia.  One such situation is where a
    methaemoglobin-producing cyanide antidote has produced dangerously
    high methaemoglobin levels.  Additionally, toluidine blue has a
    number of uses such as  in vivo staining, which are not relevant to
    the present monograph (Chobanian et al., 1987).  It has also been
    suggested as an antagonist to heparin (Deichmann & Gerarde, 1969).

    9.2.2  Name and chemical formula of antidote

         The Chemical Abstracts name of toluidine blue is
    phenothiazin-5-ium-3-amino-7-(dimethylamino)-2-methyl chloride.  The
    Chemical Abstracts registration number is 92-31-9.  Synonyms include
    tolonium chloride and toluidinblau.  The compound is listed in the
    Colour Index as Basic Blue 17, chemical constitution number 52040
    (Society of Dyers and Colourists, 1979), the supplier of the dye
    being given as BASF, Ludwigshafen and Rhein, Germany.

    9.2.3  Physico-chemical properties

         The only supplier of the pharmaceutical preparation is Dr Franz
    Kohler, Chemie GmbH, Neue Bergstrasse 3-7, Postfach 17, D-6146
    Alsbach-Hahlein 1,  Germany.  It is supplied in packs containing 5
    or 25 ampoules.

         The 10-ml ampoules contain toluidine blue at a concentration of
    40 mg/ml, i.e. 0.4 g/ampoule.  No further information is available
    on the constitution of the material.

         The relative molecular mass is 305.85.

         Toluidine blue has the following structure:

    FIGURE 08

    9.2.4  Synthesis

         No data are available.

    9.2.5  Analysis

    9.2.5.1  Analysis of methaemoglobin

         Methaemoglobin may be estimated by the spectrophotometric
    method of Evelyn & Malloy (1938) or using the IL 282 CO-oximeter
    (Instrumentation Laboratories in UK, Ltd., Warrington, Cheshire,
    United Kingdom.)  It is very important to note that neither method
    gives meaningful results in the presence of cyanide.

         Toluidine blue may interfere with the estimation of
    methaemoglobin by some methods (Smith, 1971) (see chapter 10).

    9.2.6  Stability

         No information on the stability of toluidine blue is available.

    9.2.7  General properties

         The mode of action of toluidine blue is similar to that of
    methylene blue; it catalyses the transfer of electrons from the
    pentose-phosphate pathway to methaemoglobin via NADPH-methaemoglobin
    reductase (Kiese & Waller, 1951;  Sass et al., 1969; Kiese et al.,
    1972).  It is likely that in individuals with glucose-6-
    dehydrogenase deficiency or NADPH-methaemoglobin-reductase
    deficiency the dyes would not be efficacious in reducing
    methaemoglobin (Brewer & Tarlov, 1961;  Sass et al., 1967). 
    Furthermore, certainly  in vitro, and possibly  in vivo, toluidine
    blue will produce methaemoglobin from previously non-oxidized blood
    (Kiese, 1945).

    9.2.8  Animal studies

    9.2.8.1  Pharmacokinetics

         No data are available.

    9.2.8.2  Pharmacodynamics

         A number of studies have shown toluidine blue to be superior to
    methylene blue in reducing methaemoglobin (Kiese, 1945;  Friehoff &
    Lobermann, 1952, 1953;  Kiese et al., 1972).  Kiese et al. (1972)
    found that methaemoglobin produced by the intravenous injection of
    dogs with 4-dimethylaminophenol was reduced 2-3 times more rapidly
    by toluidine blue than by methylene blue.  Burrows (1979) found that
    toluidine blue given intravenously to sheep at a dose of 1.1 mg/kg
    reduced methaemoglobin produced by sodium nitrite, although the rate
    of reduction could be increased by the use of higher doses.

    9.2.8.3  Toxicology

         The intravenous LD50s values in mice, rats, and rabbits are,
    respectively, 27.56, 28.93, and 13.44 mg/kg.  Toxic signs observed
    include increased respiration rate, convulsions, partial heart
    block, and death due to respiratory and cardiac failure.  Weekly
    intravenous injections into rabbits at a dose one tenth the LD50
    produced no change in haematological parameters, and  histological
    examination of a number of organs after 90 days revealed no
    abnormality (Stolarsky & Haley, 1951).

         Some early work on  Drosophila melanogaster suggested that
    toluidine blue might possess mutagenic properties (Landa et al.,
    1965).  Toluidine blue was reported by Au & Hsu (1979) to produce
    extensive chromosome damage.

    9.2.9  Volunteer studies

         Kiese et al. (1972) reported that, in human volunteers,
    toluidine blue reduced 4-dimethylaminophenol-induced methaemoglobin
    about twice as rapidly as did methylene blue.  No adverse effects
    were reported with toluidine blue.

    9.2.10  Clinical studies

         No data are available.

    9.2.11  Clinical studies - case reports

         Few case reports exist in the literature and some of these
    relate to the use of toluidine blue as a vital stain.  A notable
    case report of its use in toxic methaemoglobinaemia is that of

    Büttner et al. (1967).  Side-effects reported include nausea,
    vomiting and leucocytosis (Deichmann & Gerarde, 1969).

    9.2.12  Summary of evaluations

         Toluidine blue has been recommended in a number of studies as
    being preferable to methylene blue on the grounds of superior
    efficacy (Daunderer, 1979, 1980; Marrs & Ballantyne, 1987). 
    Compared to methylene blue, there is little clinical experience and
    this is confirmed by the paucity of case reports concerning the use
    of toluidine blue.  Furthermore, there is a lack of information
    concerning the nature of the material and its storage
    characteristics.

    9.2.13  Model information sheet

    9.2.13.1  Indications

         Toluidine blue is indicated in methaemoglobinaemia, including
    toxic methaemoglobinaemia; mild cases of methaemoglobinaemia,
    though, would not normally require specific therapy.  Toluidine blue
    is not indicated in cyanosis due to any other cause.  It would not
    be effective in sulfhaemoglobinaemia and, on theoretical grounds,
    would not be expected to be effecitve in the treatment of
    methaemoglobinaemia in individuals with glucose-6-phosphate
    dehydrogenase deficiency.

    9.2.13.2  Side effects

         Owing to the blue colour of the material, cyanosis may occur,
    as may staining of the urine.  Vomiting may occur following an
    overdose.

    9.2.13.3  Advised route and dose

         The compound should be given intravenously at a dose of
    2-4 mg/kg.  This may be repeated after 30 min if the clinical
    response has been inadequate.  It is important to give the material
    intravenously.

    9.2.13.4  Use in pregnancy and children

         No information is available and caution is advised in
    administering toluidine blue to a pregnant woman because of
    mutagenicity findings.  Nevertheless, in life-threatening
    methaemoglobinaemia, it would seem inadvisable to withhold
    treatment.

    9.2.13.5  Storage

         No information is available at present.

    9.2.14  References

    Au W & Hsu TC (1979) Studies on the clastogenic effects of biologic
    stains and dyes. Environ Mutagen, 1: 27-35.

    Brewer GJ & Tarlov AR (1961) Studies on the mechanism of
    primaquine-type hemolysis: the effect of methylene blue. Clin Res,
    9: 65.

    Burrows GE (1979) Methylene blue or tolonium chloride antagonism of
    sodium nitrite induced methemoglobinemia. J Vet Clin Med, 2: 81-86.

    Büttner H, Fortwich F, & Hansen HW (1967) [Treatment of a severe
    case of nitrogensol poisoning with toluidine blue.] Verh Dtsch Ges
    Inn Med, 72:  665-666 (in German).

    Chobanian SJ, Cattau EL, Winters C, Johnson DA, Van Ness MM,
    Miramadi A, Morwitz SL, & Colcher H (1987)  In vivo staining with
    toluidine blue as an adjunct to the endoscopic detection of
    Barrett's esophagus. Gastrointest Endosc, 33: 99-101.

    Daunderer M (1979) [Clinical experience with the antidote 4-DMAP, a
    methaemoglobin component for the treatment of poisonings with
    prussic acid and its salts, sulfuric acid, hydrazoic acid and its
    salts.] University of Munich, Federal Republic of Germany
    (Dissertation for appointment as Reader) (in German).

    Daunderer M (1980) [Antidote therapy: toluidine blue in
    methaemoglobinaemia.]  Fortschr Med, 98: 462-464 (in German).

    Deichmann WB & Gerarde HW (1969) Toxicology of drugs and chemicals.
    New York, London, Academic Press, p 597.

    Evelyn RA & Malloy HT (1938) Microdetermination of oxyhemoglobin,
    methemoglobin and sulfhemoglobin in a single sample of blood. J.
    Biol Chem, 126: 655-662.

    Friehoff FJ & Lobermann KH (1952) [Toluidine blue in the treatment
    of toxic haemoglobinaemias.] Ther Ggw, 91: 446 (in German).

    Friehoff FJ & Lobermann KH (1953) [Dye therapy of toxic
    haemoglobinaemias, bearing in mind the pathophysiology of such
    poisonings.] Arzneimittelforschung, 3: 616 (in German).

    Kiese M (1945) [Reduction of haemoglobin. IV.] Naunyn-Schmiedebergs
    Arch Exp Pathol Pharmakol, 204:  288-312 (in German).

    Kiese M & Waller HD (1951) [Reduction of haemoglobin and oxygen for
    reversibly reducible dyes in red cells.] Naunyn-Schmiedebergs Arch
    Pharmakol, 213: 44 (in German).

    Kiese M, Lorcher W, Weger N, & Zierer A (1972) Comparative studies
    on the effects of toluidine blue and methylene blue on the reduction
    of ferrihemoglobin in man and dog. Eur J Clin Pharmacol, 4: 115-118.

    Landa Z, Klouda P, & Pleskotova D (1965) In: Veleminsky J, Gichner
    T, & Ridesova A ed. Induction of mutations and the mutation process.
    Proceeding of a Symposium, Prague, 26-28 September, 1963. Prague,
    Czechoslovak Academy of Sciences Publishing House, pp 115-122.

    Marrs TC & Ballantyne B (1987) Clinical and experimental toxicology
    of cyanides: an overview. In: Ballantyne B & Marrs TC ed. Clinical
    and experimental toxicology of cyanides. Bristol, United Kingdom,
    John Wright, pp 473-495.

    Marrs TC, Bright JE, & Inns RH (1989) Methaemoglobin production and
    reduction by methylene blue and the interaction of methylene blue
    with sodium nitrite  in vivo. Human Toxicol, 8: 359-364.

    Sass MD, Caruso CJ, & Farhangi M (1967) TPNH-methemoglobin reductase
    deficiency:  a new red-cell enzyme defect. J Lab Clin Med, 70:
    760-767.

    Sass MD, Caruso CJ, & Axelrod DR (1969) Mechanism of the TPNH-linked
    reduction of methemoglobin by methylene blue. Clin Chem Acta, 24:
    77.

    Smith RP (1971) Spectrophotometric determination of methemoglobin
    and oxyhemoglobin in the presence of methylene blue. Clin Toxicol,
    4: 273.

    Society of Dyers and Colourists (1979) The colour index. Bradford,
    United Kingdom, The Society of Dyers and Colourists.

    Stolarsky F & Haley TJ (1951) Acute and chronic toxicity of
    toluidine blue and neutral red. Fed Proc, 10: 337.

    10.  Analytical Methods for Cyanide Alone and in Combination with
         Cyanide Antidotes in Blood

         It is important to be sure of the diagnosis in every case of
    poisoning, and this is particularly true of cyanide intoxication
    where treatment with antidotes is often potentially hazardous.  Fast
    qualitative and quantitative methods for analysing cyanide in blood
    are therefore necessary.  The methods detailed below are approved
    techniques for the detection and/or quantification of cyanide before
    the administration of cyanide antidotes.

         Measurement of whole blood free cyanide concentrations after
    the administration of cyanide antidotes is not yet possible.

    10.1  Qualitative Methods

    10.1.1  Detection in blood with a detector tube (Bedside Test)

         The time required, assuming that instrumentation and chemicals
    are readily available, is 2-3 min.

    10.1.1.1  Principle

         Hydrocyanic acid is liberated from blood by acidification and
    is then passed through a detector tube using a gas detector pump. 
    If hydrocyanic acid is present, the reactive zone of the tube
    changes its colour from yellow to red as a result of the following
    reaction:

                                 HgCl2

                  (1)   HCN       -->        HC1

                  (2)   HCl + Methyl red   -->  red reaction product

         A semiquantitative determination of the HCN concentration can
    be made by means of a scale on the tube (M. von Clarmann, personal
    communication, 1988).

    10.1.1.2  Materials

         Gas detector pump (Dräger a) (Fig. 5)
         Detector tube "Hydrocyanic acid 2/a" (Dräger a) (Fig. 6)
         Test tube height 6 cm, width 2 cm (Fig. 6)
         Stopper with two drill-holes (Fig. 5)
         10% sulfuric acid
         specimen of venous blood, with or without an anticoagulant

                 

    a   Drägerwerk AG, 2400 Lübeck Germany

    FIGURE 09

    FIGURE 10

    10.1.1.3  Procedure

         Read carefully the operating instructions for the gas detector
    pump and the detector tube.

         Insert the detector tube into the free drill-hole of the
    stopper and put the gas detector pump on the detector tube.

         Place 1 ml blood and 1 ml 10% sulfuric acid into the test tube
    and close it immediately with the stopper prepared as described
    above.  Shake the test tube cautiously, avoid splashing by all
    means.

         Compress the pump balk while closing the free end of the glass
    tube with a finger.  Then let the glass tube open while the balk of
    the pump extends.  Repeat the pumping 15 times.

    Attention!

         Take care not to suck some of the acid contents of the test
    tube into the detector tube.  If this happens, a false positive
    result will be the consequence.

         Judge the colour of the indicating layer of the detector tube
    immediately after the pumping.  If hydrocyanic acid is present, the
    indicating layer turns becomes red.  A very rough semiquantitative
    determination of the HCN concentration can be made by means of the
    scale on the tube.  The detection limit in this test is 1 mg
    CN-/l.

    10.1.1.4  Specificity

         The analysis is based on the reaction of hydrocyanic acid with
    mercury salts:  the hydrochloric acid thereby produced is 
    determined by methyl red.  Specificity is ensured by the
    precleansing layer.  Acidic gases (HCl or SO2) have no influence
    on the HCN reading, even if present in considerable excess.  This
    also applies to hydrogen sulfide.

         Sodium azide gives a positive reaction.

    10.1.2  Spot test

         The time required, assuming that instrumentation and reagents
    are readily available, is 2-3 min.

    10.1.2.1  Principle

         Hydrocyanic acid (HCN) is liberated from biological fluids by
    acidification.  The evolved HCN is passed through a filter paper
    impregnated with an alkaline solution of palladium dimethylglyoxime
    in which the palladium is a constituent of an inner-sphere complex
    anion.  Cyanide ions lead to a demasking of dimethylglyoxime, which
    reacts with nickel (II) to produce red nickel-dimethylglyoxime
    (Jakobs, 1984).

    10.1.2.2  Equipment

              Gas washing flask shown in Fig. 7 (measurements of the
              flask:  height 15 cm; width 3 cm).
              Water bath
              Centrifuge
              Disposable syringes containing anticoagulant (e.g., K-EDTA
              Monovetten R, Sarstedt Numbrecht, Germany)
              Pipettes

    10.1.2.3  Chemicals

              Ethanol
              Palladium(II) chloride
              Potassium hydroxide
              Ammonium chloride
              Nickel(II) chloride. 6H2O
              Concentrated sulfuric acid (96-98%)
              Dimethylglyoxime (2,3-butanedione dioxime)
              Antifoaming agent (e.g., silicon antifoaming emulsion

    FIGURE 11

    10.1.2.4.  Reagents

         50% Sulfuric acid
         Hydrochloric acid (0.1 mol/l)
         Potassium hydroxide (3 mol/l)

         Alkali palladium dimethylglyoxime solution:

         (a)  Dissolve 50 mg palladium(II)chloride in 2.5 ml
              hydrochloric acid (0.1 mol/l) by heating it in a water
              bath for a little while.

         (b)  Dissolve 100 mg dimethylglyoxime in 10 ml ethanol.

         (c)  Add the palladium(II)chloride solution drop by drop to the
              ethanolic solution of dimethylglyoxime.

         (d)  Centrifuge at 4000 g.

         (e)  Wash the residue with 10 ml distilled water and
              centrifuge.  Repeat this procedure three times.

         (f)  Dissolve the residue in 10 ml KOH (3mol/l) by heating it
              in water bath for a little while.

         (g)  Allow to cool, centrifuge at 4000 g if necessary.

         Nickel chloride solution (0.25 mol/l) saturated with ammonium
         chloride.a

         Dissolve 4 g ammonium chloride and 0.6 g NiCl2.6H2O in 10 ml
         water.

         All reagents are stable at 4 °C for at least one year.

    10.1.2.5  Specimen collection

         Blood is collected with disposable syringes which contain EDTA
    as anticoagulant.  If the syringes are not disposable, EDTA disodium
    salt is added to a final concentration of 1 mg/ml blood and mixed
    well.  Other anticoagulants can be used as well.

    10.1.2.6  Procedure

         Filter paper is impregnated with the alkaline solution of
    palladium dimethylglyoxime or a drop of this solution is placed on
    filter paper.

                 

    a   The presence of ammonium salts prevents the precipitation of
        green Ni(OH)2

         Into a gas washing flask (Fig. 7), 5 ml blood, 3-5 drops of the
    antifoaming agent, and 5 drops of 50% sulfuric acid are placed.

         The moist reagent paper is held on to the free opening of the
    glass tube.  Air is then blown through the flask, using the rubber
    bag for 2 min.  The part of the paper that was exposed to the stream
    of gas is then spotted with the nickel ammonium chloride solution. 
    If HCN is present, a pink to red stain of nickel dimethylglyoxime
    appears at once, the depth of the colour varying with the cyanide
    content of the gas.  The detection limit in this test is 1 mg
    CN-/l.

    10.1.2.7  Specificity

         Bright yellow inner-sphere complex palladium dimethylglyoxime
    (I) readily dissolves in caustic alkali to give a yellow solution of
    alkaline palladium dimethylglyoxime (II) in which the palladium is a
    constituent of an inner-sphere complex anion.  The inner-sphere
    complex bound dimethylglyoxime is masked in solutions of II, i.e. no
    red Ni-dimethylglyoxime precipitate appears when Ni2+ ions are
    added to the solution.  Likewise, the palladium is masked against
    practically all reagents that are normally characteristic for Pd2+
    ions.  Cyanide ions present an exception.  If added in excess to the
    yellow solutions of II, they cause immediate discharge of the
    colour, and addition of nickel salt solutions (containing NH4Cl)
    brings down red Ni-dimethylglyoxime.

         Since the Ni-dimethylglyoxime reaction is highly sensitive,
    cyanide ions can be detected through their demasking effect. (Feigl
    & Feigl, 1949; Feigl et al., 1966).  The reaction is specific to
    HCN.

    10.2  Quantitative Methods

    10.2.1  Gas chromatographic head space technique

         This procedure, described by Eben & Lewalter (1988), has been
    slightly modified from that developed by McAuley & Reive (1983). 
    The gas chromatographic determination of hydrogen cyanide may be
    carried out simply and rapidly with the head-space technique
    described here.  The use of a thermionic detector that is specific
    for nitrogen makes it possible to obtain a sensitivity 10 times
    greater than with a flame ionization detector.

         The method described is highly accurate because practically no
    sample treatment is required.  This is confirmed by the precision
    data presented here.  Another advantage over alternative analytical
    techniques is that gas chromatographic head space analysis is an
    established part of the repertoire of most medical toxicological
    laboratories.

    10.2.1.1  Principle

         Hydrogen cyanide gas is liberated by adding acid to
    cyanide-containing blood and is then determined by means of gas
    chromatography using head-space analysis.  Fractionation may be
    carried out on a packed capillary column.  The cyanide is determined
    using a thermionic nitrogen detector.

         Calibration standards, which are mixed with whole blood, are
    used for quantification.  To evaluate the data, a calibration curve
    is used in which peak height or peak area is plotted as a function
    of the cyanide concentration in blood.

    10.2.1.2   Equipment

         *    Gas chromatograph with thermionic nitrogen detector, and
              chart recorder or integrator, if necessary with a
              capillary injector

         *    Packed glass column: length, 2 m; inner diameter, 2.2 mm

         *    Column packing:  Porapak Q 100-120 mesh

         *    Alternatively, a Duran (borosilicate) glass or quartz
              capillary (length, 60 m; inner diameter, 0.3 mm) may be
              used

         *    Stationary phase:  SE 30 (Chrompack)a, chemically bonded,
              film thickness 0.3 µm

         *    Gas-tight syringes (50 or 500 µl) for gas chromatography

         *    Bottles (5 ml) with serum caps fitted with PTFE-coated
              septa, together with tools for sealing and opening

         *    Automatic pipettes, variable between 20 and 200 µl

         *    Syringe (100 µl) for gas chromatography

         *    Vortex mixer

         *    Centrifuge

         *    Transfer pipettes (1, 2, 4, and 10 ml)

         *    Volumetric flasks (10, 20, 50, and 100 ml)

         *    Disposable syringes containing anticoagulant (e.g., K-EDTA
              MonovettenR, Sarstedt, Nümbrecht, Germany)
                 

    a   Chompack, P.O. Box 8033, 4330

    10.2.1.3  Chemicals

         *    Potassium cyanide, pa

         *    Sodium cyanide, pa

         *    Glacial acetic acid, pa

         *    Helium (99.999% purity) as carrier gas

         *    Hydrogen (99.90% purity)

         *    Synthetic air (80% purified nitrogen, 20% oxygen)

    10.2.1.4  Solutions

         *    Physiological saline (9 g sodium chloride/l)

         *    Pooled stabilized whole blood or stabilized bovine blood
              for the calibration standards

    10.2.1.5  Calibration standards

         (a)  Starting solution

              12.6 mg potassium cyanide (equivalent to 5.3 mg cyanide)
              is weighed out exactly and transferred to a 50 ml
              volumetric flask, which is then filled to the mark with
              whole blood (106 mg cyanide/l blood).

         (b)  Stock solution

              Starting solution (2 ml) is pipetted into a 20 ml
              volumetric flask, which is then filled to the mark with
              whole blood (10.6 mg cyanide/l blood).  From this
              solution, calibration standards within the concentration
              range 0.2 - 4.2 mg cyanide/l blood are prepared by
              dilution with whole blood as shown in the following table:

                                                                     

      Volume of          Final volume of       Cyanide concentration
    stock solution    calibration standard    of calibration standard
         ml                    ml                      mg/l
                                                                     

          4                    10                      4.24
          2                    10                      2.12
          1                    10                      1.06
          1                    20                      0.53
          1                    50                      0.212
                                                                     
    
         The blood volume required for preparing the calibration
    standards may be reduced to about 10 ml if the standards are
    prepared directly in the bottles used for sample treatment.  In this
    case the procedure is as follows:

    Stock solution A

         Exactly 24.0 mg potassium cyanide (equivalent to 10 mg cyanide)
    is transferred to a 100 ml volumetric flask, which is then filled to
    the mark with physiological saline (100 mg cyanide/l).

    Stock solution B

         10 ml stock solution A is pipetted into a 100 ml volumetric
    flask, which is then filled to the mark with physiological saline
    (10 mg cyanide/l).

         The calibration standards are prepared directly in the bottles
    used for the sample treatment.  Blood (1 ml) is pipetted into each
    bottle and after adding the cyanide solution, the bottle is sealed
    immediately.  The difference in the volumes of the individual
    standards does not reduce the reliability of the calibration for the
    head-space analysis used here.

                                                                       

     Volume of stock     Volume of stock       Cyanide concentration
       solution A          solution B         of calibration standard
           µl                  µl                      mg/l
                                                                       

            -                  20                       0.2
            -                  40                       0.4
            -                  50                       0.5
           10                   -                       1.0
           20                   -                       2.0
           30                   -                       3.0
           50                   -                       5.0
                                                                       
    
    10.2.1.6  Specimen collection and sample preparation

         Blood is collected in disposable syringes containing an
    anticoagulant.  Immediately after collection, 1 ml of whole blood is
    pipetted into a 5-ml bottle, which is sealed at once with a cap and
    PTFE-coated septum.

         Using a 100-µl syringe, 50 µl glacial acetic acid is injected
    through the septum into the bottle to release the hydrogen cyanide. 
    The contents of the bottle are  shaken for 30 seconds on a vortex
    mixer and then centrifuged for 3 min at 3000 rpm.  The bottle is
    then incubated for 30 min at room temperature.

         The blood samples containing cyanide should be processed within
    30 min of collection, as losses can otherwise occur.

    10.2.1.7  Operational parameters for gas chromatography

         (a)   Packed column

              Column:             Material:        Glass
                                  Length:          2 m
                                  Inner diameter:  2.2 mm

              Stationary phase:   Porapak Q 100-120 mesh

              Detector:           Thermoionic nitrogen detector

              Temperatures:       Column:          120 °C
                                  Injector:        220 °C
                                  Detector:        300 °C

              Carrier gas:        Helium:          30 ml/min

              Detector gases:     Hydrogen:        3 ml/min
                                  Synthetic air:   60 ml/min

              Injected head-space 50 µl

              Under these conditions, hydrogen cyanide has a retention
              time of 1.5 min.

         (b)   Capillary column

              Capillary column:   Material:        Quartz or Duran glass
                                  Length:          50 m
                                  Inner diameter:  0.3 mm

              Stationary phase:   SE 30 (chrompack) chemically
                                  bonded, film thickness 0.3 mm

              Detector:           Thermionic specific detector (TSD)

              Temperatures:       Column:          80 °C
                                  Injector:        200 °C
                                  Detector:        300 °C
                                  Split:           1:4

              Carrier gas:        Helium, 1.68 MPa; make up gas,
                                  35 ml/min

              Detector gases:     Hydrogen:        1.45 MPa
                                                   (3.5-4.0 ml/min)
                                  Synthetic air:   160 ml/min

              Injected head-space 400 µl

              Under these conditions, the retention time for hydrogen
              cyanide is 3.1 min.

    10.2.1.8  Analytical determination

         50 µl (packed column) or 400 µl (capillary column) of the head
    space above the prepared blood sample is injected into the gas
    chromatograph under the conditions given in section 10.2.1.7.

    10.2.1.9  Calibration

         The calibration standards containing cyanide in blood are
    prepared as described in section 10.2.1.5 and analysed.  The
    calibration curve is obtained by plotting the peak heights or areas

    for the standards against the blood cyanide concentrations used (see
    Figure 8). The linearity of the calibration curve has been tested up
    to 6 mg/l.

    FIGURE 12

    10.2.1.10  Calculation of the analytical result

         From the peak height or area for the hydrogen cyanide in the
    sample, the corresponding cyanide concentration in whole blood
    (mg/l) is read off the calibration curve (see Figure 8).

    10.2.1.11  Reliability of the method

         The within-series precision and between-day precision were
    determined using a packed or a capillary column and blood samples
    with a defined cyanide content that were prepared as described in
    section 10.2.1.5.  The individual values are given in Tables 7 and
    8.

    Table 7.  Within-series precision for the gas chromatographic
              determination of cyanide in blood using a packed column
              (from Eben & Lewalter, 1988)
                                                        

             n       Cyanide in blood      Sw
                           mg/l             %
                                                        

            10             0.54            7.5

            10             2.20            4.6
                                                        

         n  =  number of analysis performed
         Sw =  relative standard deviation derived from replicate
               analyses of the same specimen
                                                                       

    Table 8.  Within-series precision and between-day precision for the
              gas chromatographic determination of cyanide in blood
              using a capillary column (from Eben & Lewalter, 1988)
                                                                       

         Precision           n      Cyanide in blood       s
                                         (mg/l)            %
                                                                       

       Within-series        17            0.19            5.4
                            17            0.39            0.9
                            17            2.11            0.7

        Between-day         17            0.19           14.5
                            17            0.39            5.5
                            17            2.11            6.7
                                                                       

             n = number of analyses performed
             s = relative standard deviation
                                                                       

         Recovery experiments were carried out using spiked blood
    samples.  This procedure is the same as that used to prepare the
    calibration curve and is thus of limited value.  The results serve
    only as an internal laboratory check.  Recovery rates of r = 84-107%
    were obtained for cyanide concentrations between 0.19 and 2.11 mg/l
    using the capillary column.

         Losses during sample treatment of between 25 and 40% were
    revealed by comparison with a calibration curve established using
    treated aqueous cyanide solution and the capillary column.

    10.2.1.12  Detection limit

         The concentration of cyanide that can be determined by this
    method, using a packed gas chromatography column, is 0.07 mg/l.  If
    a capillary column is used, the detection limit is raised to 0.1 mg
    cyanide/l blood, because with control blood a peak appears in the
    position for hydrogen cyanide.

    10.2.1.13  Specificity

         The separation of the cyanide from interfering blood
    constituents and its detection with the thermionic nitrogen detector
    makes this procedure highly specific for cyanide. With the packed
    column no interference was observed with ethanol, diethyl ether,
    methanol, acetone, ethyl acetate, n-hexane, diisopropyl ether or
    2-propanol (1 mg per analytical sample).

         Blood or plasma samples from people who had not been exposed to
    hydrogen cyanide or other cyanides yielded no peak in the position
    for hydrogen cyanide when a packed column was used.  With the
    capillary column described here a peak was also observed for blood
    samples from people not occupationally exposed to cyanides.  The
    peak, however, indicated a cyanide concentration below the detection
    limit of 0.1 mg/l.

    10.2.2  Microdiffusion technique

         Microdiffusion analysis is admirably suited to the detection
    and estimation of small quantities of volatile substances like HCN. 
    Conway (1950) pioneered the application of this method of analysis
    in the determination of such substances as ammonia, alcohol, carbon
    dioxide, volatile amines, and carbon monoxide.  The Conway diffusion
    dish has become a standard laboratory item.

         Feldstein & Klendshoj (1954) adapted the technique to the
    estimation of cyanide in biological materials.

    10.2.2.1  Principle

         Hydrocyanic acid is set free from the sample (blood, serum,
    urine, stomach contents) by the addition of sulfuric acid and is
    then absorbed into sodium hydroxide.  The cyanide concentration is
    measured colorimetrically in an aliquot of the alkaline solution
    (Asmus & Garschagen, 1953).

    10.2.2.2  Equipment

         *    Spectrophotometer (Vis)

         *    Conway microdiffusion cells (Fig. 9)a

         *    Volumetric flask 50 ml

         *    Test tubes

         *    Pipettes

    FIGURE 13

                 

    a   Culture dishes, Lips-Conway type, Code no. 911313009, Glaswerke
        Wertheim, Germany

    10.2.2.3  Chemicals

         *    Potassium cyanide

         *    Hydrochloric acid (concentrated; 35-37%)

         *    Sulfuric acid (concentrated; 96-98%)

         *    Sodium hydroxide

         *    Sodium phosphate, monobasic (NaH2PO4.2H2O)

         *    Chloramine T ( N-chloro- p-toluenesulfonamide sodium salt
              trihydrate)

         *    Barbituric acid (should be stored in a dissicator)

         *    Pyridine (should be stored in the dark)

         *    Silicone grease or other sealing agent

    10.2.2.4  Solvents and reagents

         *    Sodium phosphate (1 mol/l)

         *    Sulfuric acid (10%)

         *    Sodium hydroxide (0.1 mol/l)

         *    Chloramine T (0.25%) (must be prepared fresh daily)

          Pyridine-barbituric acid reagent

         Barbituric acid (3 g), 15 ml pyridine, and 3 ml concentrated
    hydrochloric acid are placed in a 50-ml volumetric flask.  The
    solution is mixed to dissolve the reagents, made up to 50 ml with
    water, and filtered.  This solution must be prepared fresh each time
    it is used.

    10.2.2.5  Calibration standards

         Standard solutions of sodium cyanide containing 0.1 to 2.0 µg
    of cyanide per ml of sodium hydroxide (0.1 mol/l) are prepared.

    10.2.2.6  Specimen

         Venous blood containing heparin or EDTA as anticoagulants,
    serum, plasma, urine, or stomach content (or tissue slurry) may
    serve as specimens.  The samples should be analysed as soon as
    possible.

    10.2.2.7  Procedure

          (a) Microdiffusion

              A sample (2-4 ml) of the specimen is placed in the outer
              compartment of the microdiffusion cell, and 3.3 ml of the
              sodium hydroxide solution (0.1 mol/l) is pipetted into the
              centre well of the unit.  The ground-glass cover of the
              unit is smeared with silicone grease or other sealing
              agent and is placed on the unit so that a small portion of
              the outer compartment remains uncovered.  Then 3-4  drops
              of 10% sulfuric acid are pipetted into the outer
              compartment, and the lid is quickly moved into place to
              cover the entire cell with an airtight seal.  The unit is
              then gently tilted and rotated to mix the fluids in the
              outer compartment and allowed to stand for 3 h at 20-25 °C
              for diffusion to be completed.  At the end of that time
              the lid is removed and an aliquot of the liquid in the
              centre well is taken for analysis.

          (b) Quantitative determination

              1.0 ml of the absorbing solution in the centre well is
              placed into each of three test tubes graduated at 10 ml. 
              A blank consisting of 1.0 ml of sodium hydroxide (0.1
              mol/l) in a test tube is prepared.  A 1-ml sample of a
              calibration solution is used as a positive control.

              To the blank, the control sample, and the unknown, 2.0 ml
              of sodium phosphate solution (1 mol/l) and 1.0 ml of
              chloramine T solution are added.  After mixing and
              allowing to stand for 2-3 min., 3.0 ml of the
              pyridine-barbituric acid reagent is added.  The contents
              of the tubes are again mixed and allowed to stand for
              10 min.  The presence of a red colour indicates the
              presence of cyanide.  The optical density is determined in
              a spectrophotometer at 580 nm with the blank set at zero
              density.

          (c) Calibration curve and calculation

              With aliquots of the standard solutions of sodium cyanide
              containing 0.1 to 2.0 µg of cyanide/ml of sodium hydroxide
              (1 mol/l), the colour is developed as described above.  A
              calibration curve relating optical density and
              concentration is obtained and the cyanide content of the
              material being analysed is read from the curve.

    10.2.2.8  Reliability of the method

         For cyanide concentrations from 0.04 µg/ml to 0.38 µg/ml, the
    within-run precision was ± 2.75% for blood and ± 5.56% for urine.

         The recovery of cyanide added to biological samples is 96-102%.

    10.2.2.9  Detection limit

         The detection limit depends on the amount of the analysed
    sample and of the NaOH (0.1 mol/l) for the absorption of the
    released hydrocyanic acid.  This limit can be lowered by reducing
    the volume of sodium hydroxide in the centre well of the Conway unit
    to 1.0 ml.

         With 2 ml blood and 3.3 ml NaOH (0.1 mol/l), the detection
    limit is 0.2 µg CN_ /ml blood.

    10.2.2.10  Specificity

         The separation of the cyanide from interfering blood
    constituents by microdiffusion makes this procedure very specific. 
    Hydrogen sulfide and sulfur dioxide may interfere.

    10.3  References

    Asmus E & Garschagen H (1953) [The use of barbituric acid for the
    photometric determination of cyanide and thiocyanate.] Z Anal Chem,
    138: 414-422 (in German).

    Conway EJ (1950) Microdiffusion analysis and volumetric error. New
    York, Van Nostrand Reinhold Co.

    Eben A & Lewalter J (1988) Cyanide determination in blood. In:
    Angerer J & Schaller KH ed. Analysis of hazardous substances in
    biological materials.  Volume II: Methods for biological monitoring.
    Weinheim, VCH-Verlags-gesellschaft.

    Feigl F & Feigl HE (1949) On the reactivity of inner-complex-bonded
    Palladium - a new specific test for cyanides in alkaline solutions.
    Anal Chim Acta, 3: 300-309.

    Feigl F, Anger V, & Oesper RE (1966) Spot tests in organic analysis,
    7th ed. Amsterdam, Oxford, New York, Elsevier Science Publishers, pp
    547-548.

    Feldstein L & Klendshoj NC (1954) The determination of volatile
    substances by microdiffusion analysis. J Forensic Sci, 2: 39-58.

    Jakobs K (1984) [Report on experience with the administration of
    4-DMAP in severe prussic acid poisoning.  Consequences for medical
    practice.] Zentralbl Arbeitsmed, 34: 274-277 (in German).

    McAuley F & Reive DS (1983) Rapid quantitation of cyanide in blood
    by gas chromatography. J Anal Toxicol, 7: 213-215.
    See Also:
       Toxicological Abbreviations