IPCS INCHEM Home
    MONOGRAPH FOR UKPID




    LYE




    Nicky Bates

    National Poisons Information Service (London Centre)
    Medical Toxicology Unit
    Guy's & St Thomas' Hospital Trust
    Avonley Road
    London
    SE14 5ER
    UK


    This monograph has been produced by staff of a National Poisons
    Information Service Centre in the United Kingdom.  The work was
    commissioned and funded by the UK Departments of Health, and was
    designed as a source of detailed information for use by poisons
    information centres.

    Peer review group: Directors of the UK National Poisons Information
    Service.


    1  SUBSTANCE/PRODUCT NAME

    1.1  Origin of substance

    Made by reacting calcium hydroxide with sodium carbonate, from sodium
    chloride electrolysis and from sodium metal and water vapour at low
    temperature.

    1.2  Name

    1.2.1  Brand/trade name

    1.2.2  Generic name

    1.2.3  Synonyms

    Sodium hydroxide, lye (sodium hydroxide solution, but may also refer
    to potassium hydroxide solution), caustic soda, sodium hydrate, white
    caustic, caustic flake, natriumhydroxide (German), natriumhydroxyde
    (Dutch).

    1.2.4  Common names/street names

    Not applicable.

    1.3  Chemical group/family

    Alkali.

    1.4  Substance identifier and/or classification by use

    1.5  Reference numbers

         CAS            1310-73-2
         RTECS          WB 4900000
         EINECS         2151855
         UN             1823 (dry solid)

    1.6  Manufacturer

    No data.

    1.7  Supplier/importer/agent/ licence holder

     Bayer Diagnostics, Bayer plc, Evans House, Hamilton Close,
    Basingstoke, Hants, RG21 2YE Tel: 01256 29181
     Diversey Ltd, Weston Favell Centre, Northampton, NN3 8PD Tel: 01604
    405311 Fax: 01604 406809
     Evans Vanodine International plc, Brierley Rd, Walton Summit Centre,
    Preston, PR5 8AH Tel: 01772 322200 Fax: 01772 626000
     Lever Brother s Ltd, 3 St James; Rd, Kingston-upon-Thames, Surrey
    KT1 2 BA Tel: 0181 542 8200

     SC Johnson, Frimley Green, Camberley, Surrey, GU16 5AU Tel: 01276
    852377 Fax: 01276 852688
     SC Johnson Professional, Frimley Green, Camberley, Surrey, GU16 5AU
    Tel: 01276 852706 Fax: 01276 852800

    1.8  Presentation

    1.8.1  Form

    Clinitest(R): white pale blue mottled tablets.

    Auto Dishwash: thin yellow liquid.
    Auto Dishwash Extra: thin yellow liquid.
    Auto Glass Wash: thin yellow liquid.
    Beerline: thin colourless liquid.
    Chlorinated Dishwash: thin pale yellow liquid.
    Crystal: white powder.
    D9 Oven Cleaner: thin brown liquid.
    Diverforce L3: thin yellow/green liquid.
    Diverforce L4: thin straw coloured liquid.
    Diverforce L6: thin straw coloured liquid.
    Divocare MSD: thin red liquid.
    Divocare MSD Super: thin red liquid.
    Divocare OC: brown gel.
    Filter Cleaner: thin yellow liquid.
    Green Top Bottle Washing Detergent: pale yellow powder.
    HMD Plus: thin straw coloured liquid.
    M1: thin red liquid.
    Mr Muscle Drain Opener, sachet: granules.
    Mr Muscle Oven Cleaner: aerosol.
    Mr Muscle Professional Oven Cleaner: aerosol.
    Mr Muscle Sink and Plughole Opener: hazy, straw coloured liquid.
    Oven Cleaner: viscous colourless liquid.
    Pastueriser Cleaner Liquid: thin yellow liquid.
    Pastueriser Powder: white powder.
    SPA Plus: thin straw coloured liquid.
    Sun Liquid: white viscous liquid.
    Sun Lemon Fresh Liquid: white viscous liquid.
    TC Plus: thin yellow/green liquid.
    Unipak G1: white powder.
    Unipak G3: white powder.
    Unipak G5: white powder.
    Pastueriser Cleaner Liquid: thin yellow liquid.
    Vanosan: thin pale yellow liquid.

    1.8.2  Formulation details

    Clinitest(R) tablets (Bayer Diagnostics): sodium hydroxide 232.5mg.

    Auto Dishwash (Evans Vanodine): sodium hydroxide 8.33% (pH 13.6 as
    supplied), 11.3 (working solution), TAR 8.71g NaOH/100g)
    Auto Dishwash Extra (Evans Vanodine): sodium hydroxide 8.83% (pH 13.7
    as supplied), 11.6 (working solution), TAR 10g NaOH/100g)

    Auto Glass Wash (Evans Vanodine): sodium hydroxide 4.75% (pH 13.6 as
    supplied), 11.7 (working solution), TAR 5.58g NaOH/100g)
    Beerline (Evans Vanodine): sodium hydroxide 14.8% (pH 13.7 as
    supplied), 12.6 (working solution), TAR 14.8g NaOH/100g)
    Chlorinated Dishwash (Evans Vanodine): sodium hydroxide 10-20% (pH
    13.65 as supplied), 11.8 (working solution), TAR 10.22g NaOH/100g)
    Crystal (Evans Vanodine): sodium hydroxide 45% (pH 13, TAR 44.1%
    NaOH/100g)
    D9 Oven Cleaner (Diversey): sodium hydroxide 10% (pH 13.9)
    Diverforce L3 (Diversey): sodium hydroxide 10% (pH 13 (as supplied),
    11 (working solution))
    Diverforce L4 (Diversey): sodium hydroxide 14% (pH 13.9 (as supplied),
    11(working solution))
    Diverforce L6 (Diversey): sodium hydroxide 10% (pH 13.9 (as supplied),
    11(working solution))
    Divocare MSD (Diversey): sodium hydroxide 14% (pH 13.9 (as supplied),
    11(working solution))
    Divocare MSD Super (Diversey): sodium hydroxide 14% (pH 13.9 (as
    supplied), 11(working solution))
    Divocare OC (Diversey): sodium hydroxide 6% (pH 13.9)
    Filter Cleaner (Evans Vanodine): sodium hydroxide 8.33% (pH 13.6 as
    supplied), 11.3 (working solution), TAR 8.71g NaOH/100g)
    Green Top Bottle Washing Powder: (Evans Vanodine): sodium hydroxide
    76%, (pH 12.5, TAR 67.85g NaOH/ 100g)
    HMD Plus (Diversey): sodium hydroxide 14% (pH 13 (as supplied),
    11(working solution))
    M1 (Diversey): sodium hydroxide 5.5% (pH 13.9 (as supplied),
    11.5(working solution))
    Mr Muscle Drain Opener, sachet (SC Johnson): sodium hydroxide 52% (pH
    13.5)
    Mr Muscle Oven Cleaner (SC Johnson): sodium hydroxide 1-5% (pH 13.2)
    Mr Muscle Professional Oven Cleaner (SC Johnson Professional): sodium
    hydroxide 1-5% (pH 13.5-14)
    Mr Muscle Sink and Plughole Opener (SC Johnson): sodium hydroxide 1-5%
    (pH 13)
    Oven Cleaner (Evans Vanodine): sodium hydroxide 5-10% (pH 13.8, TAR
    9.67g NaOH/100g)
    Pastueriser Cleaner Liquid (Evans Vanodine): sodium hydroxide 8.83%
    (pH 13.7 (as supplied), 11.6 (working solution), TAR 10g NaOH/100g)
    Pastueriser Powder (Evans Vanodine): sodium hydroxide 40% (pH 12.5,
    TAR 40.65g NaOH/100g)
    SPA Plus (Diversey): sodium hydroxide 10% (pH 13 (as supplied),
    11(working solution))
    Sun liquid (Lever): sodium hydroxide 1-5% (pH 13.6)
    Sun Lemon Fresh Liquid (Lever): sodium hydroxide 1-5% (pH 13.6)
    TC Plus (Diversey): sodium hydroxide 10% (pH 13 (as supplied), 11
    (working solution))
    Unipak G1 (Diversey): sodium hydroxide 57% (pH 12)
    Unipak G3 (Diversey): sodium hydroxide 35% (pH 12)
    Unipak G5 (Diversey): sodium hydroxide 20% (pH 12)
    Vanosan (Evans Vanodine): sodium hydroxide 10-20% (pH 13.65 (as
    supplied), 11.5 (working solution), TAR 10.91 NaOH/100g)

    1.8.3  Pack sizes available

    Clinitest(R) tablets: 36 tablet pack.

    Auto Dishwash: 5 L and 25 L bottles.
    Auto Dishwash Extra: 5 L and 25 L bottles.
    Auto Glass Wash: 5 L and 25 L bottles.
    Beerline: 5 L and 25 L bottles.
    Chlorinated Dishwash: 25 L bottles.
    Crystal: 15 kg buckets.
    D9 Oven Cleaner: 2 L bottle.
    Diverforce L3: 20 L drum.
    Diverforce L4: 5 L, 25 L and 200 L drums.
    Diverforce L6: 5 L and 20 L drums.
    Divocare MSD: 20 L and 5L drums.
    Divocare MSD Super: 20 L and 5L drums.
    Divocare OC: 5 L drums.
    Green Top Washing Powder: 25 kg bucket.
    HMD Plus: 20 L drum.
    M1: 2L bottle.
    Mr Muscle Drain Opener, sachet: 60g.
    Mr Muscle Oven Cleaner: 250ml, 300ml, 450ml, 500ml.
    Mr Muscle Professional Oven Cleaner: 500ml.
    Mr Muscle Sink and Plughole Opener: 500ml.
    Oven Cleaner: 5 L bottles.
    Pastueriser Cleaner Liquid: 25 L bottles.
    Pastueriser Powder: 25 kg buckets.
    SPA Plus: 5 L and 20 L drum.
    Sun Liquid: 1L, 3L bottles.
    Sun Fresh Lemon Liquid: 1L, 3L bottles.
    TC Plus: 20 L drum.
    Unipak G1: 4.5 kg direct dispense bottles.
    Unipak G3: 4.5 kg direct dispense bottles.
    Unipak G5: 4.5 kg direct dispense bottles.
    Pastueriser Cleaner Liquid: 25 L bottles.
    Vanosan: 25 L bottles.

    1.8.4  Packaging

    See above.

    1.9  Physico-chemical properties

     Chemical structure
    NaOH, molecular weight 40.01.

     Physical state
    Fused solid with crystalline structure available as pellets, sticks,
    flakes or lumps.

     Colour
    White.

     pH
    Of 0.05% solution about 12, of 0.5% solution about 13 and of a 5%
    solution about 14.

     Solubility in water and organic solvents
    1g of sodium hydroxide dissolves in 0.9ml water, 0.3ml boiling water,
    7.2ml absolute alcohol and 4.2ml methanol. It is also soluble in
    glycerol.

     Important chemical interactions
    Readily absorbs carbon dioxide and water from air. Generates heat when
    dissolving or when mixed with an acid. Corrosive to organic tissue and
    to aluminium metal in the presence of moisture. Reacts violently with
    acetaldehyde, acetic anhydride, acrolein, acrylonitrile, allyl
    alcohol, allyl chloride, chlorohydrin, chloronitrotoluenes,
    chlorosulphonic acid, 1,2-dichloroethylene, ethylene cyanhydrin,
    glyoxal, hydroquinone, maleic anhydride, nitroethane, pentol, oleum,
    phosphorous, tetrahydrofuran, trichloroethylene, water,
    4-chloro-2-2methylphenol, cinnamaldehyde, cyanogen azide, diborane,
    4-methyl-2-nitrophenol, 3-methyl-2-penten-4-yn-1-ol,
    1,2,4,5-tetrachlorobenzene (to form 2,3,7,8-tetrachlorobenzodioxin),
    1,1,1-trichloroethanol, trichloronitromethane, zinc and zirconium.

     Major products of combustion/pyrolysis
    Sodium hydroxide is not combustible but the solid form on mixing with
    moisture or water may produce sufficient heat to ignite combustible
    material. When heated to decomposition sodium oxide (Na2O) is
    produced.

     Boiling point
    1390°C
    5% solution    102°C
    10% solution   105°C
    20% solution   110°C
    30% solution   115°C
    40% solution   125°C
    50% solution   140°C

     Melting point
    318.4°C

     Density
    2.13 g/cm3 at 25°C

     Vapour pressure
    1 mmHg at 739°C

     Reactivity
    Sodium hydroxide reacts vigorously with  1,2,4,5-tetrachlorobenzene 
    to form 2,3,7,8-tetrachlorobenzodioxin.

    Mixtures with  aluminium and  arsenic compounds can form arsine.

    Contact with some metals (such as  aluminium, tin, lead and  zinc) 
    can generate hydrogen.

    1.10  Hazard/risk classification

    Corrosive.

    1.11  Uses

    Sodium hydroxide is used to neutralise acid solutions and make sodium
    salts. It is used in the manufacture of rayon, mercerised cotton,
    soap, paper, aluminium, petroleum products and in metal cleaning,
    electrolytic extraction of zinc, tin plating and oxide coating.

    Sodium hydroxide solutions hydrolyze fats to form soaps, they
    precipitate bases and most metals (as hydroxides) from aqueous
    solutions of their salts.

    Sodium hydroxide is a common constituent of many household and
    industrial cleaners including oven cleaners and beerline cleaners. It
    is present as a stabilising agent in bleach. It may also be found in
    dishwasher detergents. Some paint strippers and drain cleaners contain
    sodium hydroxide. It is used as a pipe line cleaner in dairies, bars
    and public houses.

    Small amounts of sodium hydroxide are produced as a by-product and
    released into the car interior when motor car air bag systems are
    activated. Air bag systems are triggered when a sensor in the bumper
    sends an electrical charge to a gas generator containing sodium azide
    (70 g). A chemical reaction is initiated that produces nitrogen gas
    from the sodium azide.

    Sodium hydroxide is present in Clinitest(R) tablets which are used by
    diabetics as an  in vitro semiquantitive test for glycosuria.

    1.12  Toxicokinetics

    1.12.1  Absorption

    Sodium hydroxide is not absorbed. Alkalis act at the site of contact.

    1.12.2  Distribution

    Not applicable.

    1.12.3  Metabolism

    Not applicable.

    1.12.4  Elimination

    Not applicable.

    1.12.5  Half-life

    Not applicable.

    1.12.6  Special populations

    Not applicable.

    2  SUMMARY

    3  EPIDEMIOLOGY OF POISONING

    Sodium hydroxide burns, whether dermal, ocular or gastrointestinal,
    are potentially very serious and may cause severe complications
    requiring repeated or prolonged hospitalisation and long term
    treatment.

    Ingestion

    Sodium hydroxide is the most commonly ingested strong alkali. The
    majority of cases of ingestion are accidental, particularly in
    children (Christesen, 1994a; Clausen et al, 1995; Gandhi et al, 1989;
    Edmonson, 1987; Crain et al, 1984; Grenga, 1983; Tewfik and Schloss,
    1980; Lane, 1975; Ashcraft and Padula, 1974; Leape et al, 1971) where
    severe effects are relatively rare but potentially devastating. Adults
    may also ingest sodium hydroxide accidentally often mistaking it for
    something else (Meredith et al, 1988; Chen et al, 1988; Thompson,
    1987) or where it has been transferred to a different, poorly labelled
    container. This may also occur in children (Friedman, 1987) Oral burns
    have also occurred in a child after she put an oven cleaner pad in her
    mouth (Vilogi et al, 1985).

    Sodium hydroxide has been used as a means of suicide both from
    drinking the solution (Winek et al, 1995; Christesen, 1994b; Arif and
    Karetzky, 1991; Hendrickx et al, 1990; Rabinowitz et al, 1990;
    Ferguson et al, 1989; Rubin et al, 1989; Estrera et al, 1986; Crain et
    al, 1984; Cello et al, 1980; McCabe et al, 1969; Matenga et al, 1987;
    Oakes et al, 1982; Okonek et al, 1981; Balasegaram, 1975) and from
    ingestion of capsules filled with sodium hydroxide (Carroll et al,
    1994; Nelson et al, 1983; Gill et al, 1986; Oakes et al, 1982; Lowe et
    al, 1979). A number of these cases, particularly those involving
    ingestion of sodium hydroxide solution have been fatal (Rabinowitz et
    al, 1990; Matenga et al, 1987; Estrera et al, 1986; McCabe et al,
    1969).

    In a review of 214 cases of caustic ingestion (including bleach and
    acids as well as alkalis), children aged five years and under,
    accounted for 39% of admissions but only 8% of the burns required
    treatment, whereas adults constituted 48% of admissions with 81% of
    the burns requiring treatment (Hawkins et al, 1980). Of the 20
    strictures reported in this study, 16 were due to ingestion of sodium
    hydroxide (12 following ingestion of sodium hydroxide solution).

     Clinitest(R) tablets

    Ingestion of Clinitest(R) tablets may occur accidentally (O'Connor et
    al, 1986; Schlatter-Lanz, 1985; Burrington, 1975; Payten, 1972;
    Mallory and Schaefer, 1977) particularly by children or the elderly or
    intentionally as a means of suicide (Schlatter-Lanz, 1985; Lowe et al,
    1979). Often adults take the tablets with water which may reduce the
    incidence of severe oesophageal injury in this group, however there is
    also the risk of gastric injury. Death from an oesophago-aortic
    fistula has been reported in an elderly woman who accidentally took
    two tablets with tea (O'Connor et al, 1986).

    Ingestion of Clinitest(R) tablets was a much more common problem in
    the past. A number of cases were reported in the literature in the
    1950s and 1960s (as reviewed by Lacoutre et al, 1986).

    Dermal exposure

    Dermal exposure to sodium hydroxide may occur following accidents in
    the industrial (O'Donoghue et al, 1996; Lee and Opeskin, 1995) or
    domestic setting (Lorette and Wilkinson, 1988; O'Donoghue et al,
    1996). Dermal exposure to oven cleaner pads has also resulted in
    injury (Vilogi et al, 1985).

    Chemicals are a relatively uncommon cause of burns requiring treatment
    in a burns unit, but of these, sodium hydroxide is the most commonly
    implicated chemical. Of 3,251 patients admitted to a regional burns
    unit 100 (3.1%) had sustained chemical burns, which accounted for
    16.5% of industrial burning accidents. Alkaline materials caused 37%
    of the accidents, 26% were due to sodium hydroxide, acids accounted
    for 27% of the burns with hydrofluoric acid involved in half these
    cases (Herbert and Lawrence, 1989). Alkali burns may initially appear
    superficial which can lead to a delay in treatment.

    Ocular burns

    Most alkaline burns to the eye occur at work (Moon and Robertson,
    1983; Pfister and Koshi, 1982), but domestic accidents also account
    for a large proportion of ocular burns, particularly in children (Moon
    and Robertson, 1983). The most severe injuries are seen in men of
    working age (Nelson and Kopietz, 1987). Alkalis have also been used in
    assault (Beare, 1990a). Ocular injury has also been reported from
    accidental instillation of a sodium hydroxide solution into the eye
    instead of a contact lens solution (Mauger, 1988) and from exposure to
    sodium hydroxide in oven cleaner pads (Vilogi et al, 1985).

    Motor car air bags

    Small amounts of sodium hydroxide are produced as a by-product and
    released into the car interior when motor car air bag systems are
    activated. There have been reports of both ocular (White et al, 1995;
    Smally et al, 1992; Swanson-Biearman et al, 1993; Ingraham et al,
    1991) and dermal burns (Swanson-Biearman et al, 1993) following
    activation of air bags during motor car accidents.

    4  MECHANISM OF ACTION/TOXICITY

    4.1  Mechanism

    Alkalis cause liquifactive necrosis with saponification of fats and
    solubilisation of proteins. There is a decrease in the collagen
    content of tissue and saponification of cell membrane lipids and
    cellular death. They are also hygroscopic and will absorb water from
    the tissues. These effects result in adherence and deep penetration
    into the tissues.

    Ingestion

    Alkalis cause the most severe corrosive effects on the oesophagus,
    rather than the stomach as is the case with acids. However, following
    deliberate ingestion of a large quantity of an alkali (as with
    intentional ingestion in adults) both the stomach and small intestine
    may be involved. This is particularly the case with liquid sodium
    hydroxide.

    Oesophageal changes can be divided into 3 stages:
    1) acute necrotic phase in which cell death occurs due to coagulation
    of intracellular protein,
    2) intense inflammatory reaction in viable tissues surrounding the
    necrotic area, thrombosis of vessels occurs,
    3) sloughing of superficial necrotic layer 2-4 days later (Adam and
    Birck, 1982).

    Strictures form due to an intense fibroblastic reaction and
    superficial granulation tissue formation that terminates with
    extensive scar formation and luminal narrowing. The newly formed
    collagen contracts both cirumferentially and longitudinally resulting
    in oesophageal shortening and stricture formation.

     Clinitest(R) tablets

    The sodium hydroxide in these tablets reacts with the saliva and
    liberates heat which can produce a full thickness burn of the
    oesophagus. Clinitest(R) tablets also adhere to the oesophagus either
    because of thermal coagulation or because of the carbon dioxide
    bubbles produced from the reaction of the citric acid and sodium
    hydroxide present (Burrington, 1975). Gastric injury may also occur.

    Ocular burns

    Alkali burns of the eye are very serious because they cause disruption
    of the protective permeability barriers and rapidly penetrate the
    cornea and anterior chamber. They combine with cell membrane lipids
    which causes disruption of the cells and stromal mucopolysaccharides
    with concomitant tissue softening. Sodium hydroxide can pass freely
    through the cornea and cause damage to all layers of the cornea and to
    the anterior segment structures in severe cases (Wright, 1982).

    In the acute phase the following occurs: sloughing of the corneal
    epithelium, necrosis of the cells of the corneal stroma and
    endothelium, loss of corneal mucoid, oedema of the corneal stroma and
    ciliary processes, ischaemic necrosis and oedema of the conjunctiva
    and limbal region of the sclera and infiltration of inflammatory cells
    into the cornea and iris. Corneal infiltration and degeneration occurs
    1-3 weeks after injury (Grant and Schuman, 1993).

    Alkalis cause rapid loss of corneal mucoprotein. They also bind to
    corneal mucoprotein and collagen and the eye may remain alkaline
    despite prolonged irrigation due to slow dissociation of hydroxyl ions
    from corneal proteins. A large number of animal experiments,
    particularly on rabbits, have been conducted to study alkali injury to
    the eye and it is known from these experiments that in the rabbit eye
    the pH of the aqueous humour can rise to 10-11 within a few minutes or
    even higher in severe cases. The pH slowly falls over a period of
    hours, except in extreme cases (Grant and Schuman, 1993).

    Injury of the endothelium causes failure of the endothelial pump which
    normally keeps the cornea deturgesced and clear. This failure causes
    the corneal stroma to become oedematous and susceptible to
    vascularisation and scarring.

    When the protective sheath around collagen is damaged, collagenases
    produced by polymorphonuclear leucocytes infiltrate the damaged area
    and degrade the corneal collagen causing melting or ulceration of the
    stroma, formation of descemoteceles (herniation of Descemet's
    membrane) and perforation of the cornea. This phase is evident 3-7
    days post-injury.

    The role of injury to blood vessels remains unclear, as with other
    chemical burns to the eye the blood vessels of the conjunctiva and
    episclera are seen to be thrombosed immediately after exposure.

    Both alkali and acid burns cause a transient rise in intraocular
    pressure due to shrinkage of the eye coats. A second phase of raised
    intraocular pressure may occur due to prostaglandin release and a
    still later phase with glaucoma due to obstruction of aqueous outflow
    caused by inflammation or synechiae.

    In severely damaged untreated burns inflammatory destruction and
    ulceration continues to occur for weeks or months resulting in damage
    ranging from dense vascular invasion to opaque scarring and corneal
    perforation.

    During the recovery process, surviving keratocytes begin to form new
    collagen but a decrease in available ascorbate (caused by injury to
    the main source in the ciliary body) is a limiting factor in this
    process. This lack of ascorbate may also render the cornea more
    susceptible to attack by oxygen free radicals.

    4.2  Toxic dose

    The severity of injury will depend on a number of factors including
    the concentration of the agent, the duration of contact and the volume
    ingested. It is greatest where the pH is above 12. However, pH is not
    the only factor which determines the extent to which a substance can
    cause corrosive injury. Alkaline reserve (which is the amount of a
    standard acid solution needed to titrate an alkali to a specified pH -
    usually pH 8, the pH of normal oesophageal mucosa) has been found to
    correlate better than pH with the production of caustic oesophageal
    injury (Hoffman et al, 1989). Although this method is not currently
    used (except internally by some manufacturers to classify products as
    irritant or corrosive) it appears to be a better predictor of injury
    than pH. Alkaline reserve may also be referred to as the titratable
    alkaline reserve (TAR).

    Solid preparations and viscous liquids are also more likely to produce
    severe injury due to prolonged contact. Following ingestion of a small
    amount the injury is usually limited to the oropharyngeal region and
    the oesophagus. The greater the volume the greater the risk of
    duodenal and gastric damage.

    Several studies have been carried out in an attempt to correlate
    clinical effects and injury. Gaudreault et al (1983) found that signs
    and/or symptoms do not adequately predict the presence or severity of
    an oesophageal lesion. Crain et al (1984) found that the presence of
    two or more signs or symptoms (vomiting, drooling, stridor) may be a
    reliable predictor of oesophageal injury. In the study by Nuutinen et
    al (1994) prolonged drooling and dysphagia (12-24 hours) were observed
    to predict oesophageal scar formation with 100% sensitivity. In the
    study of 224 children (aged 0-14 years) by Clausen et al (1994)
    serious complications were due to ingestion of sodium hydroxide or a
    dishwasher product. Children without any signs or symptoms at the
    first examination did not develop stricture or epiglottal oedema.

    The study by Christesen (1995) also found that complications only
    developed in children who had ingested strong alkalis (sodium
    hydroxide, ammonia and dishwasher products). The author found that
    children with respiratory symptoms were at greater risk of developing
    complications and that liquid sodium hydroxide tended to cause more
    complications than the granular form. It was also concluded that
    asymptomatic patients are not at risk of complications and probably do
    not require endoscopy.

    Knopp (1979) reported that in patients with oral burns approximately
    one third had significant oesophageal injury, whereas 2-15% of
    patients with oesophageal injury had no oral burns.

     Clinitest(R) tablets

    Ingestion of 1 tablet is sufficient to cause oesophageal stricture.
    However, ingestion of 47 tablets over 1 month by an adult caused only
    gastritis and eschar formation in the lower two thirds of the stomach
    with full recovery (Mallory and Schaefer, 1977).

    5  FEATURES OF POISONING

    5.1  Acute

    5.1.1  Ingestion

    Ingestion of sodium hydroxide may cause an immediate burning pain in
    the mouth, oesophagus and stomach (retrosternal and epigastric pain),
    with swelling of the lips. This is followed by vomiting, haematemesis,
    increased salivation, ulcerative mucosal burns, dyspnoea, stridor,
    dysphagia and shock. Oesophageal and pharyngeal oedema may occur.

     It should be noted that oesophageal damage may occur in the absence 
     of oral burns (Krenzelok and Clinton, 1979; Kynaston et al, 1989).

     Acute complications
    These include gastrointestinal haemorrhage and perforation of the gut
    (suggested by increasing abdominal pain, persistent vomiting, direct
    and indirect tenderness and a rigid abdominal wall).

    Gastric ulcers on the greater curvature and the antrum have been
    reported following ingestion of capsules filled with sodium hydroxide
    (Gill et al, 1986; Carroll et al, 1994; Oakes et al, 1982; Lowe et al,
    1979).

     Late complications
    Oesophageal stricture and pyloric stenosis may occur as late
    complications. Stricture formation usually begins to develop 14-21
    days after ingestion. Most strictures become manifest within the first
    two months. Strictures may prevent an adequate nutritional intake and
    in severe cases patients may be unable to swallow their own saliva.

    Healing of oesophageal injury depends on the severity of the injury.
    In the study by Di Costanzo et al (1980) first degree burns healed in
    about a week, second degree burns in 20-30 days and third degree burns
    within 90 days.

    Gastric necrosis and stricture may occur, usually in patients who have
    oesophageal injury as well. Gastric injury is more likely to occur
    following ingestion of liquid sodium hydroxide than ingestion of the
    solid. The small intestine may also be involved, duodenal and colonic
    necrosis has also been reported with ingestion of sodium hydroxide
    solution (Guth et al, 1994).

    Oesophago-aortic fistulae and rupture of the aorta are rare
    complications of corrosive ingestion (Ottoson, 1981).
    Tracheo-oesophageal fistulae and less commonly, broncho-oesophageal
    fistulae, have been reported following ingestion of alkalis (McCabe et
    al, 1969; Sarfati et al, 1987; Allen et al, 1970). They have also been
    reported to form as a complication of oesophageal dilatation (Mutaf et
    al, 1995). The main symptoms of a tracheo-oesophageal fistula are
    persistent pneumonia, choking and cyanosis while feeding.

    Severe corrosive injury to the stomach may result in a small scarred
    immobile stomach and in such cases small, frequent intakes of food may
    be necessary to prevent dumping syndrome. Achlorhydria with reduced or
    absent intrinsic factor may also occur. (Dumping syndrome is a complex
    reaction thought to be secondary to excessively rapid emptying of
    gastric contents into the jejunum. Clinical effects include nausea,
    sweating, palpitations, syncope and diarrhoea).

    Alkalis are known to increase the risk of oesophageal cancer, which
    can occur years after the initial injury (Isolauri and Markkula, 1989;
    Appelqvist and Salmo, 1980; Kinnman et al, 1968). The incidence of
    carcinoma following oesophageal injury from sodium hydroxide is 0.8-4%
    - see section 9.1.

     Clinitest(R) tablets

    These are easily mistaken for oral medication and can cause severe
    oesophageal injury if ingested. They can become lodged in the
    oesophagus and cause stricture formation. Strictures are usually short
    and located at about the level of carina (Burrington, 1975). Death
    from an oesophago-aortic fistula has been reported (O'Connor et al,
    1986).

    5.1.2  Inhalation

    Pulmonary oedema may occur following inhalation of vaporised caustics.

    5.1.3  Dermal

    On the skin alkalis can cause severe burns. The skin is discoloured
    brown or black which may make initial assessment of the injury
    difficult. There may be recurring skin breakdown over a long period
    (O'Donoghue et al, 1996).

    It should be noted that after exposure to low concentration alkalis
    the affected area may remain painless for several hours (Lorette and
    Wilkinson, 1988).

    5.1.4  Ocular

    Sodium hydroxide is responsible for some of the most severe, blinding
    injuries to the eye. There is intense pain with blepharospasm. Visual
    acuity is decreased due to injury to the corneal epithelium and
    corneal oedema.

    In mild cases there may be sloughing of the corneal and conjunctival
    epithelium. In more severe cases there may be conjunctival swelling
    (chemosis) and necrosis with hazing or opacity of the cornea. There is
    a rise in intraocular pressure initially as the sclera of the eye
    shrinks and then due to a release of prostaglandins.

    Severe burns are characterised by ischaemic necrosis. This is due to
    reduced or absent blood flow as a result of thrombosed blood vessels
    which leaves the conjunctiva and sclera white. If the eye has been
    penetrated fibrin may be present in the anterior chamber with
    infiltration of inflammatory cells. The pupil may be dilated and
    unreactive due to damage to the sphincter and dilator muscles. Retinal
    damage may also occur in severe ocular burns.

    Assessment of the injury and a prognosis may be difficult until 48-72
    hours after the exposure (Pfister and Koshi, 1982).

    There may be involvement of the lids with erythema, oedema or
    blistering or in severe cases ischaemic necrosis.

    Late complications of severe ocular burns include: persistent oedema,
    glaucoma, vascularisation of cornea, fibrous tissue scarring of
    cornea, growth of vessels throughout cornea, ulceration, perforation,
    permanent opacity, staphyloma, phthisis bulbi (shrinkage and wastage
    of eyeball), cataract and symblepharon (adhesion between tarsal
    conjunctiva and bulbar conjunctiva).

    5.1.5  Other routes

    No data.

    5.2  Chronic toxicity

    5.2.1  Ingestion

     Clinitest(R) tablets

    A 71 year old woman developed gastritis and eschar formation in the
    lower two thirds of the stomach after accidentally ingesting 47
    Clinitest(R) tablets over a period of 1 month. Her oesophagus was
    normal. She made a full recovery (Mallory and Schaefer, 1977).

    5.2.2  Inhalation

    Obstructive airway disease has been reported following chronic
    occupational exposure to sodium hydroxide mist. The patient developed
    cough, dyspnoea and tachypnoea after a 20 year exposure to sodium
    hydroxide. The solution was used to clean jam containers which were
    boiled in it for two hours. He had a barrel chest with limited
    movements and diffuse expiratory wheezing. A chest X-ray showed severe
    pulmonary hyperinflation (Rubin et al, 1992).

    5.2.3  Dermal

    No data.

    5.2.4  Ocular

    No data.

    5.2.5  Other routes

    No data.

    5.3  Systematic description of clinical effects

    5.3.1  Cardiovascular

    Tachycardia and hypotension may occur in patient with severe corrosive
    damage to the gastrointestinal tract resulting in haemorrhage or
    perforation.

    Oesophago-aortic fistulae and rupture of the aorta are rare
    complications of corrosive ingestion (Ottoson, 1981), they has also
    been reported following ingestion of Clinitest(R) tablets (O'Connor et
    al, 1984). Pericarditis may occur (McCabe et al, 1969; Matenga et al,
    1987).

    5.3.2  Respiration

     Acute
    Dyspnoea and stridor may occur and in severe cases there may be upper
    airway obstruction. Recurrent atalectasis and pneumonia developed in
    two adults who suffered bronchial burns after ingestion of sodium
    hydroxide (Meredith et al, 1988)

    Mediastinitis may occur (McCabe et al, 1969; Matenga et al, 1987;
    Allen et al, 1970).

    Tracheo-oesophageal fistulae and less commonly, broncho-oesophageal
    fistulae, have been reported following ingestion of alkalies (McCabe
    et al, 1969; Sarfati et al, 1987; Balasegaram, 1975). They have also
    been reported to form as a complication of oesophageal dilatation
    (Mutaf et al, 1995). The main symptoms of a tracheo-oesophageal
    fistula are persistent pneumonia, choking and cyanosis while feeding.

    Pulmonary oedema may occur following inhalation of vaporised caustics.

     Chronic
    Obstructive airway disease has been reported following chronic
    occupational exposure to sodium hydroxide mist. The patient developed
    cough, dyspnoea and tachypnoea after a 20 year exposure to sodium
    hydroxide. The solution was used to clean jam containers which were
    boiled in it for two hours. He had a barrel chest with limited
    movements and diffuse expiratory wheezing. A chest X-ray showed severe
    pulmonary hyperinflation (Rubin et al, 1992).

    5.3.3  Neurological

    Neurological effects are not expected with sodium hydroxide exposure,
    except the psychiatric complications of patients requiring long-term
    therapy and the effects of inability to speak or swallow.

    5.3.4  Gastrointestinal

    Ingestion of sodium hydroxide may cause an immediate burning pain in
    the mouth, oesophagus and stomach (retrosternal and epigastric pain),
    with swelling of the lips. This is followed by vomiting, haematemesis,
    increased salivation, ulcerative mucosal burns and dysphagia.
    Oesophageal and pharyngeal oedema may occur.

     It should be noted that oesophageal damage may occur in the absence 
     of oral burns (Krenzelok and Clinton, 1979; Kynaston et al, 1989).

     Acute complications
    These include gastrointestinal haemorrhage and perforation of the gut
    (suggested by increasing abdominal pain, persistent vomiting, direct
    and indirect tenderness and a rigid abdominal wall).

    Gastric ulcers on the greater curvature and the antrum have been
    reported following ingestion of capsules filled with sodium hydroxide
    (Gill et al, 1986; Carroll et al, 1994; Oakes et al, 1982; Lowe et al,
    1979).

     Late complications
    Oesophageal stricture and pyloric stenosis may occur as late
    complications. Stricture formation usually begins to develop 14-21
    days after ingestion. Most strictures become manifest within the first
    two months. Strictures may prevent an adequate nutritional intake and
    in severe cases patients may be unable to swallow their own saliva.

    Healing of oesophageal injury depends on the severity of the injury.
    In the study by Di Costanzo et al (1980) first degree burns healed in
    about a week, second degree burns in 20-30 days and third degree burns
    within 90 days.

    Gastric necrosis and stricture may occur, usually in patients who have
    oesophageal injury as well. Gastric injury is more likely to occur
    following ingestion of liquid sodium hydroxide than ingestion of the
    solid. The small intestine may also be involved, dudodenal and colonic
    necrosis has also been reported with ingestion sodium hydroxide
    solution (Guth et al, 1994).

    Oesophago-aortic fistulae and rupture of the aorta are rare
    complications of corrosive ingestion (Ottoson, 1981).
    Tracheo-oesophageal fistulae and less commonly, broncho-oesophageal
    fistulae, have been reported following ingestion of alkalies (McCabe
    et al, 1969; Sarfati et al, 1987; Allen et al, 1970). They have also
    been reported to form as a complication of oesophageal dilatation
    (Mutaf et al, 1995). The main symptoms of a tracheo-oesophageal
    fistula are persistent pneumonia, choking and cyanosis while feeding.

    Severe corrosive injury to the stomach may result in small scarred
    immobile stomach and in such cases small, frequent intakes of food may
    be necessary to prevent dumping syndrome (Meredith et al, 1988.
    Achlorhydria with reduced or absent intrinsic factor may also occur. 

    (Dumping syndrome is a complex reaction thought to be secondary to
    excessively rapid emptying of gastric contents into the jejunum.
    Clinical effects include nausea, sweating, palpitations, syncope and
    diarrhoea).

    Alkalis are known to increase the risk of oesophageal cancer, which
    can occur years after the initial injury (Isolauri and Markkula, 1989;
    Appelqvist and Salmo, 1980; Ti, 1983; Hopkins and Postlethwait, 1981;
    Benirschke, 1981; Kinnman et al, 1968). The incidence of carcinoma
    following oesophageal injury from sodium hydroxide is 0.8-4% - see
    section 9.1.

     Clinitest(R) tablets
    These are easily mistaken for oral medication and can cause severe
    oesophageal injury if ingested. They can become lodged in the
    oesophagus and cause stricture formation. Strictures are usually short
    and located at about the level of carina (Burrington, 1975), they
    generally take. Gastric ulcerations have also been reported (Oakes et
    al, 1982). A 71 year old woman developed gastritis and eschar
    formation in the lower two thirds of the stomach after accidentally
    ingesting 47 Clinitest(R) tablets over a period of 1 month. Her
    oesophagus was normal. She made a full recovery (Mallory and Schaefer,
    1977). Death from an oesophago-aortic fistula has been reported
    (O'Connor et al, 1986).

    5.3.5  Hepatic

    Jaundice with an elevated alkaline phosphatase concentration has been
    reported in a fatal case of sodium hydroxide ingestion. The authors
    postulated that biliary obstruction due to oedema, following burns of
    the ampulla as a possible explanation for this finding (Matenga et al,
    1987).

    5.3.6  Urinary

    No data.

    5.3.7  Endocrine and reproductive system

    No data.

    5.3.8  Dermatological

    Sodium hydroxide causes deep penetrating burns and necrosis. The skin
    is discoloured brown or black which may make initial assessment of the
    injury difficult. There may be recurring skin breakdown over a long
    period (O'Donoghue et al, 1996).

    Dermal alkali injuries differ from those of other burns for a number
    of reasons:

    *    The injury may be painless and not be immediately evident. This
         initial lack of pain may lead to a delay in treatment.

    *    Injury can progress over several hours.

    *    Also initial assessment of the burn depth may be difficult
         because the skin may be discoloured brown or black within a short
         period of time.

    5.3.9  Eye, ears, nose and throat

     Ocular
    Sodium hydroxide is responsible for some of the most severe, blinding
    injuries to the eye. There is intense pain with blepharospasm. Visual
    acuity is decreased due to injury to the corneal epithelium and
    corneal oedema.

    In mild cases there may be sloughing of the corneal and conjunctival
    epithelium. In more severe cases there may be conjunctival swelling
    (chemosis) and necrosis with hazing or opacity of the cornea. There is
    a rise in intraocular pressure initially as the sclera of the eye
    shrinks and then with a release of prostaglandins.

    Severe burns are characterised by ischaemic necrosis. This is due to
    reduced or absent blood flow as a result of thrombosed blood vessels
    which leaves the conjunctiva and sclera white. If the eye has been
    penetrated fibrin may be present in the anterior chamber with
    infiltration of inflammatory cells. The pupil may be dilated and
    unreactive due to damage to the sphincter and dilator muscles. Retinal
    damage may also occur in severe ocular burns.

    Assessment of the injury and a prognosis may be difficult until 48-72
    hours after the exposure (Pfister and Koshi, 1982).

    There may be involvement of the lids with erythema, oedema or
    blistering or in severe cases ischaemic necrosis.

    Late complications of severe ocular burns include: persistent oedema,
    glaucoma, vascularisation of cornea, fibrous tissue scarring of
    cornea, growth of vessels throughout cornea, ulceration, perforation,
    permanent opacity, staphyloma, phthisis bulbi (shrinkage and wastage
    of eyeball), cataract and symblepharon (adhesion between tarsal
    conjunctiva and bulbar conjunctiva).

    5.3.10  Haematological

    Neutropaenia and thrombocytopenia have been reported in a fatal case
    of sodium hydroxide ingestion (Matenga et al, 1987).

    5.3.11  Immunological

    No data.

    5.3.12  Metabolic

    5.3.12.1  Acid-base disturbances

    Lactic acidosis may occur in severe cases due to tissue damage and
    shock (Okonek et al, 1981).

    5.3.12.2  Fluid and electrolyte disturbances

    None.

    5.3.12.3  Other

    None.

    5.3.13  Allergic reactions

    No data.

    5.3.14  Other clinical effects

    None.

    5.4  At risk groups

    5.4.1  Elderly

    Since the elderly are not involved in occupations involving sodium
    hydroxide they are at less risk of exposure. However they may still be
    exposed in the domestic setting or from motor car air bags. Accidental
    ingestion of Clinitest(R) tablets is also a problem in this age group.

    5.4.2  Pregnancy

    Not applicable.

    5.4.3  Children

    Children are at particular risk of exposure to sodium hydroxide in the
    home where they may sustain dermal, ocular or gastrointestinal injury.
    Ocular injury has been reported in a child following activation of an
    motor car air bag (Ingram et al, 1991). Children are also at risk from
    accidental ingestion of Clinitest(R) tablets.

    5.4.4  Enzyme deficiencies

    Not applicable.

    5.4.5  Enzyme induced

    Not applicable.

    5.4.6  Occupations

    There is a risk of exposure to sodium hydroxide in a number of
    occupations including chemical workers, those involved in the
    manufacture of sodium hydroxide and cleaners using sodium hydroxide
    containing products.

    5.4.7  Others

    Dermal (Swanson-Biearman et al, 1993) and ocular (White et al, 1995;
    Smally et al, 1992; Swanson-Biearman et al, 1993) burns have been
    reported from exposure to sodium hydroxide following activation of
    motor car air bags.

    6  MANAGEMENT

    6.1  Decontamination

    Ingestion

    Gastric lavage and emesis are contraindicated because of the risks of
    further injury on re-exposure of the oesophagus. Nasogastric
    aspiration of the stomach contents is probably less effective for
    ingestion of alkali than for acids, since alkalis tend to damage the
    oesophagus rather than the stomach. Oral fluids should be given (Homan
    et al, 1994) unless there is evidence of severe injury. Neutralising
    chemicals should never be given because heat is produced during
    neutralisation and this could exacerbate any injury (Rumack and
    Burrington, 1977).

     Ingestion of Clinitest(R) tablets
    Orange juice or milk should be given to drink as soon as possible
    after ingestion of a Clinitest(R) tablet as this reduces the rate of
    release of heat of reaction. Water should be given if milk is not
    available. Even late administration of diluents may be beneficial
    because of the non-uniform break up of tablets (Lacoutre et al, 1986).

    Ocular

    Copious and immediate irrigation of exposed eyes is essential. Water
    (preferably sterile) or normal saline may be used, although other
    solutions have been employed in an emergency including tap water,
    Hartmans solution and universal buffer solution. Particulate matter
    should be removed with cotton wool buds or forceps. The pH of the
    cornea and irrigating fluid from the eye should be monitored with
    universal indicator paper. Irrigation should be continued until the pH
    of the eye is normal and remains so for 2 hours.

    Pain and blepharospasm may make irrigation difficult and the use of
    anaesthetic drops (e.g. amethocaine, lignocaine) may be needed to
    facilitate thorough irrigation. A lid speculum may be used if
    required. It is essential that the whole eye is irrigated including 

    under the upper and lower lids. The use of an irrigating contact lens
    or an anterior chamber tap (paracentesis) have also been recommended
    (Nelson and Kopietz, 1987; Pfister and Koshi, 1982).

    Dermal

    All contaminated clothing must be removed. Copious irrigation of the
    skin should be commenced immediately. If an extensive area has been
    exposed whole body irrigation should be undertaken preferably using a
    high-flow shower unit. The burnt areas should then be covered with wet
    compresses during transfer to hospital (Herbert and Lawrence, 1989).
    See supportice care for further details.

    6.2  Supportive care

    Ingestion

     Asymptomatic/mildly symptomatic patients
     It should be noted that oesophageal damage may occur in the absence 
     of oral burns (Krenzelok and Clinton, 1979; Kynaston et al, 1989).

    These patients must be admitted for observation until the extent of
    the injury (if any) can be determined. Oral fluids may be given if the
    patient is willing to drink. Oral feeding should be maintained if the
    patient is able to tolerate it, otherwise tube feeding or parenteral
    nutrition should be provided. On discharge all patients must be
    advised on the possibility of late onset sequelae and advised to
    return if necessary.

     Severely affected patients
    Treatment is supportive. In severely affected patients aggressive
    intervention is essential (Meredith et al, 1988; Hendrickx et al,
    1990; Andreoni et al, 1995). Urgent assessment of the airway and
    endoscopic evaluation is required (see section 6.6 Investigations). A
    supraglottic-epiglottic burn with erythema and oedema is usually an
    sign that further oedema will occur which will lead to airway
    obstruction and is an indication for early intubation or tracheostomy
    (Meredith et al, 1988). See section 6.7 Management controversies.

    Give plasma expanders/intravenous fluids for shock and check and
    correct the acid/base balance. Analgesia will almost certainly be
    needed. Intubation and ventilation may be necessary for patients with
    respiratory distress. Appropriate antibiotic therapy should be given
    if necessary. Parenteral feeding will be necessary.

     Late complications
    Strictures that prevent adequate nutritional intake and do not respond
    to dilatation require oesophagectomy and colonic interposition.
    Oesophageal strictures which result in a lumen >10mm do not impede
    normal life and should not require intervention (Sarfati et al, 1987).
    Surgical intervention may also be required for gastrointestinal
    perforation or haemorrhage.

    There may be loss of speech and inability to swallow as a result of
    severe corrosive injury. Speech and swallowing rehabilitation is a
    complex subject and is discussed by Shikowitz et al (1996) with
    description of the surgical techniques used and the tools used to
    determine the success of the reconstruction.

    Severe cases of sodium hydroxide ingestion may result in long-term
    hospitalisation and repeated surgical procedures. Psychological
    support is recommended.

    Ocular

    After irrigation further treatment is aimed at preventing optic nerve
    damage from raised intraocular pressure and to protect the cornea from
    ulceration, perforation and infection (Nelson and Kopietz, 1987).
    Raised intraocular pressure can be managed with topical beta-blockers
    (e.g timolol).

    Broad spectrum antibiotic drops (e.g. gentamicin, chloramphenicol) are
    recommended in cases of epithelial damage. Delayed epithelial healing
    is common after severe alkali burns and the use of 'bandage' soft
    contact lenses have been recommended to speed up the process (Nelson
    and Kopietz, 1987; Pfister and Koshi, 1982). Mydriatic drops, such as
    atropine, may alleviate pain but a short acting mydriatic may be
    better, particularly in the elderly who are at risk of glaucoma from
    mydriasis (Beare, 1990b).

    In the long-term care of chemical injuries to the eye the main problem
    is dry eyes due to damage to the Meibomian glands, presence of mucin-
    secreting globlet cells on the eye surface or scarring of the tear
    ducts. The use of tear film substitutes (e.g. hypromellose drops) may
    be beneficial in such cases. Chronic glaucoma and recurrent epithelial
    breakdown are other common problems and make patients with alkali
    damaged corneas poor candidates for corneal grafting.

    Corneal grafting should not be attempted until 12-18 months after the
    injury once the inflammatory response has settled. The failure rate is
    high, about 50-70% (Pfister and Koshi, 1982) because of the reasons
    stated above. In cases where grafting fails a keratoprosthesis is a
    last resort but this has a high complication and failure rate (Pfister
    and Koshi, 1982).

    Plastic surgery may be required for correction of other problems
    including cicatricial lid deformities, trichiasis (ingrowing
    eyelashes) and symblepharon.

    See also section 6.7 Management Controversies.

    Dermal

    On arrival at hospital analgesia, oxygen therapy and intravenous fluid
    replacement should be commenced if the clinical condition of the
    patient warrants it. Assessment of the depth and extent of injury is
    not reliable at an early stage.

    The most important therapy for dermal alkali injuries is prolonged
    copious irrigation. This effectively cleanses the wound of unreacted
    chemical, dilutes the chemical already in contact with tissue and
    restores tissue water lost to the hygroscopic effect of alkalis
    (Leonard et al, 1982). The earlier the irrigation is begun the greater
    the benefit. Irrigation should, therefore, be started as soon as
    possible and continued, in hospital, until the pH of the skin is no
    longer alkaline. In the 273 cases reported by Bomberg et al (1965)
    irrigation was continued for between 6 hours and 6 days (mean 24
    hours). In this study, irrigation was shown to reduce the length of
    stay in hospital, the time to grafting and the number of cases
    requiring surgery.

    Irrigation of the injury site at the scene of the accident has also
    been shown to decrease the incidence of full thickness skin loss and
    to significantly reduce hospital stay and morbidity (Leonard et al,
    1982).

    Testing the pH of the skin immediately after irrigation may be
    misleading. It is recommended that 15 minutes elapse before this is
    undertaken to allow residual alkali to diffuse up from the deeper
    regions of the dermis (Herbert and Lawrence, 1989; O'Donoghue et al,
    1996).

    Spontaneous separation of the eschar (the slough produced by a burn)
    following a chemical burn takes about 30 days, which is much longer
    than with thermal burns. Waiting for this to occur prolongs treatment
    and O'Donoghue et al (1996) recommend irrigation followed by early
    excision of the injury site followed by immediate skin grafting for
    the best results. (Debridement is the removal of foreign material or
    dead skin from or adjacent to a traumatic or infected lesion until
    healthy tissue is exposed).

    Referral to a burns unit is recommended.

    6.3  Monitoring

    Ingestion/dermal exposure

    In severely affected patients it will be necessary to monitor the
    airway and respiratory status and blood gases. The patient should also
    be monitored for signs of shock.

    Ocular exposure

    Intraocular pressure should be monitored.

    6.4  Antidotes

    None relevant.

    6.5  Elimination techniques

    None relevant.

    6.6  Investigations

    Ingestion

    Abdominal and chest X-rays need to be taken to check for perforation.

     Endoscopic examination
    If facilities allow gastro-oesophagoscopy should be undertaken to
    assess the extent and severity of the injury. This procedure can be
    safely undertaken within 9-96 hours of the ingestion. It should be
    avoided 5-15 days post-ingestion because this is the time period over
    which the tissues slough and there is increased risk of perforation.
    Endoscopy is contraindicated in patients with third degree burns of
    the hypopharynx, burns involving the larynx or those with respiratory
    distress. Where there is complete obliteration of the oesophageal
    lumen the procedure should be terminated and repeated 48 hours later
    (Ali Zargar et al, 1991).

    Traditionally the endoscopist terminates the procedure at the first
    deep, penetrating and/or circumferential burn because of the risk of
    perforation. However with a flexible endoscopy the stomach and small
    intestine (panendoscopy) can be examined regardless of the presence of
    second or non-perforating third degree burns to the oesophagus (Ford,
    1991). Ali Zargar et al (1991) recommend examination of the stomach
    and the first part of the duodenum.

    Pharyngolaryngoscopy and bronchoscopy are also recommended in severely
    affected patients. Laparotomy evaluation of the gastrointestinal tract
    beyond the oesophagus may also be indicated if there are second or
    third degree circumferential burns of the oesophagus (Meredith et al,
    1988).

    Oesophageal burns are generally classified by the following
    description (Ford, 1991):
    *    first degree - erythema and oedema.
    *    second degree - erythema, blistering, superficial ulceration,
         fibrinous exudate.
    *    third degree - erythema, deep ulceration, friability, eschar
         formation, perforation.

    First degree burns do not cause strictures, second degree burns result
    in strictures in 15-30% of cases but third degree injury results in
    stricture formation in over 90% of cases (Howell, 1986).

    Follow up appointments should be made on discharge because of the risk
    of delayed damage and the patient should be advised to return in the
    meantime if there is any evidence of oesophageal stricture (suggested
    by vomiting, weight loss, bloating and anorexia). Patients who develop
    strictures will require monitoring for life for the development of
    malignant disease of the oesophagus.

    Ocular

    Ophthalmological referral is essential for alkali burns to the eye.
    Visual acuity and intraocular pressure should be checked if possible.
    The eye stained with fluorescein since this will reveal the full
    extent of conjunctival and corneal epithelial loss (Beare, 1990b).

    6.7  Management controversies

    Ingestion

     Management of severe corrosive injury
    Although it is agreed that the aim of initial management is the
    prevention of strictures, there is no consensus on the management of
    severe corrosive injury to the oesophagus and stomach. A number of
    very useful reviews have been published on the subject (Shaffer et al,
    1994; Andreoni et al, 1995; Ford, 1991; Moore, 1986; Postlethwait,
    1983; Klein-Schwartz and Oderda, 1983; Knopp, 1979). There is no
    question of the value of endoscopic investigation to determine the
    extent and severity of the injury although there is some debate on the
    relative usefulness of different types of endoscope (rigid and
    flexible). Some authors recommend early aggressive treatment (Andreoni
    et al, 1995; Hendrickx et al, 1990; Meredith et al, 1988; Gossot et
    al, 1987; Thompson, 1987; Estrera et al, 1986; Kirsh et al, 1978;
    Allen et al, 1970) while others recommend more conservative therapy
    (Ribet et al, 1990), particularly in children (Gandhi et al, 1989;
    Rappert et al, 1993; Gundogdu et al, 1992).

    Oesophagogastrectomy or gastrectomy (partial or total) has been
    recommended for large areas of necrosis to prevent perforation and
    subsequent peritonitis or mediastinitis. In practice this will only be
    necessary in a small number of patients. In the experience of Gossot
    et al (1987), which includes ingestion of acids as well as alkalies, a
    necrotic stomach always leads to perforation. There is also the risk
    of fistula formation which can occur between the oesophagus and the
    trachea, bronchus or aorta. Gossot et al (1987) found a high mortality
    rate with oesophagectomy performed transthoracically (76%) and
    recommend the removal of a necrotic oesophagus by a transhiatal blunt
    dissection and stripping technique. In the small number of patients on
    which this was performed the mortality rate was 38%. The authors
    suggest that the avoidance of thoracotomy may prevent the occurrence
    of tracheobronchial lesions.

    Meredith et al (1988) recommend gastrostomy with placement of a string
    to guide future dilatations for patients not undergoing gastrectomy or
    oesophagogastrectomy. Early implacement of a intraluminal stent has
    also been recommended to prevent the formation of strictures (Estrera
    et al, 1986; Wijburg et al, 1989).

    Gandhi et al (1989) recommend the use of string-guided oesophageal
    dilatation with endoscopically guided steroid injection as a safe and
    reliable method for the treatment of oesophageal strictures in 

    children. In these cases there was no need for oesophageal replacement
    or resection in most patients. In the study by Gundogdu et al (1992)
    it was found that conservative treatment of strictures was more
    successful in children under 8 years of age.

     The use of steroids and antibiotics in the management of corrosive 
     oesophageal injury
    Steroids have an anti-inflammatory effect and decrease fibroblastic
    activity and scar tissue formation. Animal data has demonstrated that
    strictures formed in subjects given steroids have been less well
    structured with fewer inflammatory changes and less fibrin deposition
    (Spain et al, 1950; Rosenberg et al, 1953). The use of steroids for
    corrosive injury in man is a controversial subject which has generated
    a huge amount of literature.

    Middlekamp et al (1969) reported thirty-two cases of oesophageal burns
    due to alkali injury, none of the patients with first degree burns
    (nineteen cases) developed oesophageal stricture. One patient (out of
    six) with second degree burns and all patients with third degree burns
    (seven) developed oesophageal stricture.

    In a study by Anderson et al (1990) on the use of steroids in children
    with corrosive injury of the oesophagus, of one hundred and thirty-one
    children sixty had oesophageal burns. Of these burns, fifty-five were
    caused by known agents, 91% of which were alkaline. Oesophageal
    stricture developed in ten of the thirty-one children treated with
    steroids and in eleven of the twenty-nine controls. Nine of the ten
    patients in the steroid group had third degree burns and one had
    second degree injuries, all eleven patients in the control group had
    third degree burns. Twenty-one patients given steroids and eighteen
    controls did not develop strictures, of these children all but one had
    first or second degree injuries.

    Several authors have found that it is the depth of the initial burn
    rather than the initial treatment which determines the outcome
    (Anderson et al, 1990; Moazam et al, 1987; Webb et al, 1970; Oakes et
    al, 1982).

    The main aim of the management of alkali injury is to reduce stricture
    formation. The role of steroids in alkali injury is still the subject
    of much debate, however most authors agree that patients with first
    degree burns do not require steroids since these burns usually heal
    without stricture formation. The difficulty here is determining the
    severity of the injury from oesophagoscopy since it is difficult to
    determine the depth of the burn and the endoscope cannot be passed
    beyond the first identified burn due to the risk of perforation. Some
    burns are so severe and extensive that strictures may develop despite
    steroid therapy (Haller et al, 1971) and may be delayed (Middlekamp et
    al, 1969).

    In summary, steroids are probably most effective for second degree or
    moderately severe burns (Klein-Schwartz and Oderda, 1983; Hawkins et
    al, 1980; Webb et al, 1970). They are not necessary for first degree
    burns and appear to be ineffective in preventing stricture formation
    following third degree burns. However, there is no clinical evidence
    that steroid therapy is more effective than non-steroid therapy in
    reducing oesophageal stricture in any patient, even those with second
    degree burns. Some authors believe there is no place for steroid
    therapy in the management of corrosive injury (Wijburg et al, 1989; Di
    Costanzo et al, 1980). As Oakes et al (1982) state, clinicians should
    not feel compelled to institute steroid therapy for caustic
    oesophagitis simply because it is considered 'standard therapy'.

     Contraindications and problems in steroid therapy
    It should be noted that there are definite contraindications to the
    use of steroids, these are as follows:-
    a) active infection
    b) perforation of the gastrointestinal tract or secondary
    mediastinitis
    c) significant gastrointestinal bleeding
    d) history of or active ulcer

    Steroids depress the immune system and as a result the patient is more
    susceptible to infection. Also steroids may mask the signs and
    symptoms of infection as well as those of perforation and peritonitis.
    Steroid therapy may also result in a thin-walled oesophagus vulnerable
    to perforation due to reduced wound healing and scar formation
    (Cardona and Daly, 1971).

     When to begin therapy
    Once the decision has been made to use steroids therapy should be
    started within 24-48 hours of the injury because the major
    inflammatory insult occurs within the first 48 hours and after this
    time steroids have little antifibroblastic activity. Therapy started
    later may reduce scar formation but all evidence indicates that the
    best results are obtained with early institution of therapy (Haller et
    al, 1971). The short duration of steroid therapy should not produce a
    significant reduction in intrinsic steroid production or alter the
    metabolic balance.

     Antibiotics
    Antibiotics should be used in all patients with evidence of infection.
    Some authors suggest that prophylactic antibiotics should be given in
    patients on steroid therapy (Howell et al, 1992; Hoffman et al, 1990;
    Adam and Birck, 1982), but others considered this unnecessary (Wijburg
    et al, 1989; Klein-Schwartz and Oderda, 1983; Wijburg et al, 1985)
    since the risk of infection is low (Knopp, 1979).

     Lathyrogens
    Lathyrogens are drugs which decrease the strength of collagen fibres
    through interruption of the covalent crosslinks between newly formed
    collagen molecules. Although animal data has demonstrated that
    lathyrogens are effective in preventing stricture formation (Butler et
    al, 1977) their efficacy in man is as yet undetermined (Moore, 1986).

    Ocular burns

    A number of substances have been investigated in the treatment of
    alkali burns of the eye with the aim of reducing the severity of
    damage.

     Ascorbate (ascorbic acid)
    Ascorbate is required for the formation of collagen and production is
    reduced when the ciliary body is damaged by alkali burns. It may also
    act as a scavenger of free radicals released by the infiltrating
    polymorphonuclear leucocytes.

    Animal experiments with systemic and topical ascorbate has produced
    conflicting results (Grant and Schuman, 1993; Pfister, 198; Pfister et
    al, 1982). Wright (1982) suggests the use of 10% sodium ascorbate
    drops hourly by day for 4-6 weeks. Systemic ascorbate may also be
    given (1g daily), but topical ascorbate is favoured over systemic
    because damage to the ciliary epithelium impairs transport of
    ascorbate across the blood-aqueous barrier.

     Collagenase inhibitors
    These have been used because there is evidence that collagenases
    released by infiltrating polymorphonuclear leucocytes can cause the
    breakdown of corneal stroma. Various agents have been tried including
    cysteine, acetylcysteine, sodium and calcium EDTA, penicillamine.
    Animal experiments have produced conflicting results (Wright, 1982;
    Grant and Schuman, 1993) and there are currently no recommendations
    for the use of collagenase inhibitors in alkali injury to the eye.

     Sodium citrate
    Citrate is thought to prevent the degranulation of polymorphonuclear
    leucocytes and thereby reduce the production of free radicals and
    proteolytic enzymes, and it has been suggested that the use of topical
    10% citrate may be beneficial in the treatment of alkali-induced
    corneal ulcers (Pfister et al, 1988).

     Steroids
    Steroid therapy in the treatment of alkali-induced ocular injury is
    controversial. Is should only be considered for the treatment of
    severe burns. Wright (1982) suggests topical steroid therapy in cases
    of severe burns but only where stromal melting is absent and then to
    change to a collagenase inhibitor such as sodium citrate. However,
    clinical data to support their use is lacking and Pfister and Koshi
    (1982) advise avoiding topical steroids where possible because they
    interfere with the healing process.

     Prostaglandin inhibitors
    Prostaglandins are important in the recruitment of polymorphonuclear
    leucocytes but an effective topical prostaglandin preparation is not
    readily available and data on the benefit of such therapy is lacking.

    Numerous other treatments have been recommended (as reviewed in Grant
    and Schuman, 1993) but with no outstanding success.

    7  CASE DATA

    Literature

    Case 1: Accidental ingestion, child

    A 16 month old female refused to drink and began drooling after
    ingesting the residue of a sodium hydroxide solution which the mother
    had been using for cleaning. She vomited several times with the
    vomitus containing a small amount of blood. The pharynx was red and
    there was slight bleeding of the upper gums. The chest was initially
    clear but 90 minutes after admission inspiratory and expiratory
    wheezes were present and a chest X-ray suggested aspiration pneumonia.
    At 15 hours post-ingestion laryngoscopy and oesophagoscopy were
    performed. The false cords and epiglottis were found to be red and
    oedematous. The cricopharyngeus was ulcerated and bleeding. The
    oesophageal mucosa was bleeding and circumferential second and third
    degree burns were present. The child required intubation and
    ventilation and was started on methylprednisolone and ampicillin.
    Ventilatory support was necessary for three weeks. Subsequent
    laryngoscopy revealed laryngeal oedema and burns which resulted in
    laryngeal stenosis. An oesophagoscopy at five weeks post-ingestion
    revealed oesophageal narrowing. A barium swallow showed multiple
    oesophageal strictures and hypoperistalsis of the proximal segment of
    the oesophagus. The child required nine oesophageal dilatations, and
    was eventually able to take oral feedings. She was discharged one year
    after the ingestion (Klein-Schwartz and Oderda, 1983).

    Case 2: Accidental ingestion, fatal

    A 14 year old boy took a sodium hydroxide solution (30%) in to his
    mouth. He immediately spat it out. He drank some milk and water and
    vomited. On arrival about 30 minutes later he had retrosternal pain
    and had difficulty swallowing. He was given antibiotics and steroids.
    Oesophagoscopy was performed two days later and revealed mucosal
    lesions in the upper oesophagus. He began to improve and was able to
    take mashed food orally. He then began to develop difficulty in
    swallowing and a X-ray on day 23 revealed a stricture at the level of
    the carina of trachea. On the 38th day oesophagoscopy with dilatation
    of the stricture was performed. About 2 hours later he suffered
    immediate retrosternal pain. An X-ray showed perforation of the
    stricture. This was sown up via a left side thoracotomy. Serious
    inflammatory changes were observed with mediastinal emphysema and a
    purulent pleuritis. A nasogastric tube and three drains were left in
    place. On the 44th day after ingestion profuse bleeding was observed
    through the nasogastric tube and drains was noted. He became shocked
    and the decision was made to operate. He suffered a cardiac arrest
    while general anaesthetic was being given. A right side thoracotomy
    showed a 4-5 mm rupture of the descending part of the aorta with
    bleeding into the left pleura. After cardiac massage, blood
    transfusion and repair of the rupture he stabilised. Part of the
    oesophagus was removed due to inflammation. On day 52 another 

    haemorrhage occurred. He was operated on again and the haemorrhage was
    seen to arise from the aortic rupture. The aorta wall was fragile and
    could not be repaired. The patient died on the operating table. A
    purulent mediastinitis, bilateral purulent pleuritis, lung atalectasis
    and pericarditis were observed at postmortem (Ottoson, 1981). 

    Case 3: Accidental ingestion, adult

    A 19 year old male vomited immediately after ingestion of 2 mouthfuls
    of sodium hydroxide. He had drunk it thinking it was wine. He was seen
    in hospital within 6 hours. He had deep oral and hypopharyngeal burns,
    full thickness necrosis of the entire oesophagus and stomach and
    mucosal burns of the bronchus. He required an emergency treacheostomy
    and oesophagogastrectomy with establishment of a cervical
    oesophagostomy. A chimney jejunostomy was constructed to provide
    access for enteral feeding. His post-operative course was complicated
    by recurrent atelectasis and pneumonia secondary to the bronchial
    burns, bowel obstruction and wound complications. He was discharged 6
    weeks after the incident with a tracheostomy and jejeunostomy. He
    developed dense scarring and stenosis of the hypopharynx which
    required multiple laser excisions to restore pharyngeal lumen. After 9
    months he underwent oesphagagogastric reconstruction with an
    antiperistaltic left colon interposition from the hypopharynx to the
    duodenum. Supraglottic scarring required further laser excisions and
    dilatations over the next 18 months. More than 2 years after the
    event, he was eating a normal diet and speaking clearly but still
    occasionally required dilatation of the pharynx (Meredith et al,
    1988).

    This patient was one of nine individuals (aged 15-25 years) who
    sustained severe corrosive injury to the stomach and oesophagus after
    drinking sodium hydroxide solution in mistake for wine. The solution
    had been kept in the fridge for the making of pretzels. The container
    resembled a wine bottle and was labelled in German (Meredith et al,
    1988; Chen et al, 1988; Thompson, 1987).

    Case 4: Clinitest(R) ingestion, fatal

    A 77 year old partially deaf diabetic woman accidentally swallowed 2
    Clinitest(R) tablets. There was no evidence of oropharyngeal
    ulceration. She was given milk to drink and discharged the next day.
    She returned 2 weeks later with increasing dysphagia and could only
    swallow fluids. An endoscopy revealed an oesophageal stricture. She
    died three days later after a massive haematemesis. Postmortem
    revealed an ulcer in the upper oesophagus which had penetrated and
    perforated the aortic wall to form a oesophago-aortic fistula. There
    was no evidence of pharyngeal or gastric ulceration (O'Connor et al,
    1984). 

    Case 5: Clinitest(R) ingestion, child

    A 17 month old child was brought to hospital because of increasing
    dysphagia of 2 weeks duration. Clinitest(R) tablets were kept in the
    home. A barium swallow revealed a short oesophageal stricture.
    Repeated dilatation were of no lasting benefit even after gastrostomy
    and passage of a string through the oesophagus. After 19 dilatations
    in 3 months she underwent resection and primary anastomosis.
    Post-operatively she required 8 dilatations over 12 months before she
    could eat normally (Burrington, 1975).

    Case 6: Ingestion of encapsulated sodium hydroxide

    A 31 year female was admitted to hospital after ingestion that day of
    3 gelatin capsules which she had filled with sodium hydroxide drain
    cleaner. She complained of abdominal cramping. Endoscopy revealed
    ulcerative, exudative oesophagitis in the mid to distal region. There
    were three deep ulcers along the greater curvature of the stomach with
    black necrotic debris in the centre. She was given antibiotics and
    corticosteroids and discharged after 7 days. One month after ingestion
    she returned with haematemesis and melaena. Endoscopy revealed
    persistence of the ulcers in the stomach, one had evidence of recent
    bleeding. She discharged herself after 4 days. Similar gastric
    ulceration was observed in another patient who ingested four capsules
    of the same drain cleaner. These had healed by 6 weeks post-ingestion
    (Gill et al, 1986). 

    Case 7: Dermal exposure, fatal

    A 32 year old man was working in an area of a plant were hot sodium
    hydroxide was being used. It is thought that he was accidentally
    exposed to a spray and aerosol of the alkali which caused him to run
    for the shower. Before he reached it he fell or collapsed into a pool
    of sodium hydroxide. He was found next to the shower face down. He was
    turned over and sprayed with copious amounts of water but was found to
    be dead. A postmortem was performed 40 hours later but chemical damage
    was observed to be progressing 3 days later, just prior to the
    funeral. At postmortem the cotton clothing was found to be in good
    condition but the leather gloves and boots were badly disintegrated.
    The left boot had dissolved, only the steel toe cap and sole remained.
    There were severe burns present, in some cases with total soft tissue
    destruction down to the bone. Both eyes were severely damaged with
    perforation of one. On the left side of the neck the skin and fat was
    totally absent revealing the underlying muscles and vasculature. The
    left side of the body was most severely damaged. There were some
    changes to the respiratory tract indicating a minor degree of
    inhalation. There was extensive epithelial loss of the tongue and the
    stomach wall was found to be neutral indicating ingestion of a small
    amount of alkali. In a re-enactment of the accident it was found that
    the time taken for work boots to become damaged to the same degree as
    the victim's under similar circumstances was about 13 minutes (Lee and
    Opeskin, 1995).

    Case 8: Dermal exposure requiring skin graft

    A 20 year old patient presented 2 hours after accidentally spraying
    herself in the face with an oven cleaner containing 4% sodium
    hydroxide. She had removed the excess liquid but did not irrigate the
    area. She did not experience any pain until nearly 2 hours later. On
    examination she was in moderate distress with no ocular involvement.
    The right side of her face was erythematous and blistering in a
    serpiginous pattern extending from the infraorbital rim to the body
    and angle of her mandible. The area of the right cheek had a bronze
    discolouration. The body surface area involved was about 2%. The area
    was irrigated for 60 minutes. Despite this the burn continued to show
    signs of third degree burn involvement. She was transferred to a burns
    unit and underwent surgical debridement and skin graft. Follow up six
    weeks later revealed good healing and no complications (Lorette and
    Wilkinson, 1988). 

    Case 9: Ocular injury

    A 31 year man had sodium hydroxide blown into his amblyopic left eye
    after an explosion caused by placing solid sodium hydroxide into a
    plugged drain. He washed the eye immediately in a shower and arrived
    at hospital within 5 minutes. On examination the cornea was opaque and
    the lower two thirds of the conjunctiva was ischaemic. Topical
    irrigation was repeated and he was transferred to the operating room
    where intraocular irrigation was commenced. About 100-120ml of
    Ringer's solution was used in this procedure over 90 minutes. At this
    time the cornea was slightly clearer. Methylprednisolone was given by
    retobulbar and subconjunctival injection. Continuous slow topical
    irrigation was continued for a further 24 hours. On the first
    post-operative day visual acuity was present to light, intraocular
    pressure was high and a cataract was present. Topical antibiotics,
    systemic and topical corticosteroids and carbonic anhydrase inhibitors
    were given. Two weeks after the injury aspiration-irrigation of the
    cataract was undertaken with an improvement in visual acuity.
    Acetylcysteine drops were used and a soft contact lens was put in
    place. He was discharged three weeks after the injury but returned
    three days later with severe pain, hypopyon and hyphema. The cornea
    ulcerated and perforated 27 days after the injury. The perforation was
    repaired with a corneoscleral free hand graft. Despite the presence of
    light perception the eye was enucleated at the patient's request 70
    days after the injury (Burns and Hikes, 1979).

    Internal cases

    1) A 63 year old male died within 1 hour of admission after ingestion
    of sodium hydroxide. He had ingested 10 teaspoons in a glass of water.
    He developed pain in the mouth and chest and was given milk to drink.
    On postmortem the pH of the stomach contents was 13.4. There was
    blue-black discolouration of the mucosa from the lips to the pylorus.
    There was no perforation, but milk was found in the lungs (NPIS
    (London) 75/8481).

    2) A 44 year old male was admitted to hospital after ingestion of 6
    tablespoons of sodium hydroxide in a pint of water. He had severe
    burns to the lips, mouth and pharynx. Twelve hours after admission he
    developed laryngeal stricture and a tracheostomy was performed. His
    condition remained satisfactory and the tracheostomy was removed three
    days later. Four days after this he developed haematemesis and
    continued to vomit blood for a month. He required 34 pints of blood by
    transfusion. He began to improve after this time with occasional
    diarrhoea. He then lost a stone in weight and developed electrolyte
    problems, particularly hypokalaemia. It was thought he may have
    developed a gastro-colic fistula. A gastro-graffin swallow revealed
    narrowing of the oesophagus and of the distal part of the stomach with
    dye passing straight through into the ascending colon via a
    gastric-colonic fistula. Seven days after this a laparotomy was
    performed and a feeding jejunostomy tube inserted into the jejunum. He
    began to maintain his weight. About two weeks after this a barium meal
    confirmed the presence of the fistula and he was started on oral
    fluids. Three days later he complained of abdominal pain after
    ingestion of soup. There was abdominal tenderness and rigidity. A
    laparotomy revealed free fluid in the right para-colic gutter with an
    inflamed appendix. An appendectomy and partial gastrectomy were
    performed. Post-operatively he recovered well and was started on oral
    fluids. Three weeks later he developed another episode of severe
    abdominal pain. He deteriorated and another laparotomy showed a closed
    duodenal perforation. His condition remained poor with pyrexia,
    tachycardia and hypotension. Septicaemia was diagnosed and he was
    continued on ampicillin and IV fluids. He continued to deteriorate and
    died 3 days after the last operation, 94 days post-ingestion (NPIS 
    (London) 73/9664)

    8  ANALYSIS

    8.1  Agent/toxin/metabolite

    Not relevant.

    8.2  Sample containers to be used

    8.3  Optimum storage conditions

    8.4  Transport of samples

    8.5  Interpretation of data

    8.6  Conversion factors

    8.7  Other recommendations

    9  OTHER TOXICOLOGICAL DATA

    9.1  Carcinogenicity

    Alkalis are known to increase the risk of oesophageal cancer, which
    can occur years after the initial injury (Isolauri and Markkula, 1989;
    Ti, 1983; Hopkins and Postlethwait, 1981; Benirschke, 1981; Appelqvist
    and Salmo, 1980; Kinnman et al, 1968). The incidence of carcinoma
    following oesophageal injury from sodium hydroxide is 0.8-4%. Of the
    fifteen patients (age range 38-83) in the study by Isolauri and
    Markkula (1989) twelve had accidentally swallowed sodium hydroxide at
    the age of two or three years, one at fifteen years and one at
    twenty-three years of age. The time between ingestion and the
    diagnosis of oesophageal cancer was 22-81 years. Appelqvist and Salmo
    (1980) describe similar results, out of sixty patients with
    oesophageal cancer for which the time of ingestion was known,
    fifty-two had ingested the sodium hydroxide at the age of ten years or
    younger.

    9.2  Genotoxicity

    No data.

    9.3  Mutagenicity

    9.4  Reprotoxicity

    No data.

    9.5  Teratogenicity

    No data.

    9.6  ADI

    The acceptable daily intake is not determined.

    9.7  MRL

    9.8  AOEL

    9.9  TLV

    TLV = 2 mg m-3.
    long term exposure limit = 5 mg m-3 (EH 40/95).

    9.10  Relevant animal data

    Leape et al (1971) demonstrated that 5ml of a 30.5% solution of sodium
    hydroxide was uniformly fatal when introduced into a cat's oesophagus,
    even if neutralised 3 seconds later. Exposure of cat oesophagus to
    8.3% sodium hydroxide for 30 seconds resulted in destruction of the 
    superficial layer of squamous mucosa along with submucosal and
    transmural thrombosis in blood vessels (Ashcraft and Padula, 1974).

    9.11  Relevant  in vitro data

    10  ENVIRONMENTAL DATA

    10.1  Ecotoxicological data

    No data.

    10.2  Behaviour

    No data.

    10.3  Biodegradation

    No data.

    10.4  Environmentally important metabolites

    None.

    10.5  Hazard warnings

    10.5.1  Aquatic life

    10.5.2  Bees

    10.5.3  Birds

    10.5.4  Mammals

    10.5.5  Plants

    10.5.6  Protected species

    10.6  Waste disposal data

    Sodium hydroxide may occur in waste from the following industries:
    soaps, cleaning compounds, pulp and paper, petroleum, mineral and
    vegetable oils, leather, dyes and textile dyeing, cotton, coal
    distillation, etc.

    Author

    Nicky Bates

    National Poisons Information Service (London Centre)
    Medical Toxicology Unit
    Guy's & St Thomas' Hospital Trust
    Avonley Road
    London
    SE14 5ER
    UK

    This monograph was produced by the staff of the London Centre of the
    National Poisons Information Service in the United Kingdom. The work
    was commissioned and funded by the UK Departments of Health, and was
    designed as a source of detailed information for use by poisons
    information centres.

    Peer review was undertaken by the Directors of the UK National Poisons
    Information Service.

    June 1996

    REFERENCES

    Adam JS and Birck HG. 1982
    Pediatric caustic ingestion.
    Ann Otol Rhinol Laryngol 91:656-658

    Ali Zargar S, Kochhar R, Mehta S and Mehta SK. 1991
    The role of fibreoptic endoscopy in the management of corrosive
    ingestion and modified endoscopic classificatiion of burns.
    Gastrointest Endosc 37:165-169

    Allen RE, Thoshinsky MJ, Stallone RJ and Hunt TK. 1970
    Corrosive injuries of the stomach.
    Arch Surg 100:409-413

    Anderson KD, Rouse TM and Randolph JGH. 1990
    A controlled trial of corticosteroids in children with corrosive
    injury of the oesophagus.
    N Eng J Med 323 (10):637-640

    Andreoni B, Marini A, Gavinelli M, Biffi R, Tiberio G, Farina ML and
    Rossi A. 1995
    Emergency management of caustic ingestion in adults.
    Surg Today 25:119-124

    Appelqvist P and Salmo M. 1980
    Lye corrosion carcinoma of the oesophagus. A review of 63 cases.
    Cancer 45:2655-2658

    Arif A and Karetzky MS. 1991
    Complications of caustic ingestion.
    N J Med J 88 (3):201-204

    Ashcraft KW and Padula RT. 1974
    The effect of dilute corrosives on the esophagus.
    Pediatrics 53 92):226-232

    Balasegaram M. 1975
    Early management of corrosive burns of the oesophagus.
    Br J Surg 62:444-447

    Beare JDL. 1990a
    Eye injuries from assault with chemicals.
    Br J Opthalmol 74:514-518

    Beare J. 1990b
    Management of chemical burns to the eye.
    In Major Chemical Disasters - Medical Aspects of Management.
    Murray V (ed).
    International Congress and Symposium Series 155
    Royal Society of Medicine

    Benirschke K. 1981
    Time bomb of lye ingestion?
    Am J Dis Child 135:17-18

    Bomberg BE, Sung CL and Walden RH. 1965
    Hydrotherapy for chemical burns.
    Plast Reconstr Surg 35:85-95

    Burns RP and Hikes CE. 1979
    Irrigation of the anterior chamber for the treatment of alkali burns.
    Am J Ophthalmol 88 (1):119-120

    Burrington JD. 1975
    Clinitest burns of the esophagus.
    Ann Thorac Surg 20 (4):400-404

    Butler C, Madden JW, Davis WM and Peacock EE. 1977
    Morphological aspects of experimental esophageal lye strictures. II.
    Effect of steroid hormones, bougienage and induced lathyrism on acute
    lye burns.
    Surgery 81:431-435

    Cardona JC and Daly JF. 1971
    Current mangement of corrosive esophagitis. An evaluation of results
    of 239 cases.
    Ann Otol 80:521-527

    Carroll L, Agnone F and Vance M. 1994
    Laproscopic evaluation after lye ingestion (abstract 140).
    Vet Hum Toxicol 36 (4):373

    Cello JP, Fogel RP and Boland R. 1980
    Liquid caustic ingestion. Spectrum of injury.
    Arch Intern Med 140:501-504

    Chen YM, Ott DJ, Thompson JN and Gelfand DW. 1988
    Progressive roentgenographic appearance of caustic esophagitis.
    South Med J 81 (6):724-744

    Christesen HBT. 1995
    Prediction of complications following unintentional caustic ingestion
    in children. Is endoscopy always necessary?
    Acta Paediatr 84:1177-1182

    Christesen HBT. 1994a
    Epidemiology and prevention of caustic ingestion in children.
    Acta Paediatr 83:212-215

    Christesen HBT. 1994b
    Caustic ingestion in adults - epidemiology and prevention.
    Clin Toxicol 32 (5):557-568

    Clausen JO, Nielsen TLF and Fogh A. 1994
    Admission to Danish hospitals after suspected ingestion of corrosives.
    A nationwide survey (1984-1988) comprising children aged 0-14 years.
    Dan Med Bull 41:234-237

    Crain EE, Gershel JC and Mezey AP. 1984
    Caustic ingestions. Symptoms as predictors of esophageal injury.
    Am J Dis Child 138:863-865

    Di Costanzo J, Noirclerc M, Jouglard J, Escoffier JM, Cano N, Martin J
    and Gauthier A. 1980
    New therapeutic approach to corrosive burns of the upper
    gastrointestinal tract.
    Gut 21:370-375

    Edmonson MB. 1987
    Caustic ingesion by farm children.
    Pediatrics 79 (3):413-416

    EH 40/95
    Occupational exposure limits. 1995
    HSE

    Estrera A, Taylor W, Mills LJ and Platt MR. 1986
    Corrosive burns of th esophagus and stomach: a recommendation for an
    aggressive surgical approach.
    Ann Thorac Surg 41:276-283

    Ferguson MK, Migliore M, Staszak VM and Little AG. 1989
    Early evaluation and therapy for caustic esophageal injury.
    Am J Surg 157:116-119

    Ford M. 1991
    Alkali and acid injuries of the upper gastrointestinal tract.
    in Contemporary management in critical care 1 (3):225-249
    Tobin MJ and Grenvik A (eds).
    Churchill Livingstone

    Friedman EM. 1987
    Caustic ingestions and foreign body aspirations: an overlooked form of
    child abuse.
    Ann Otol Rhinol Laryngol 96:709-712

    Gandhi RP, Cooper A and Barlow BA. 1989
    Successful management of esophageal strictures without resection or
    replacement.
    J Pediatr Surg 24 (8):745-750

    Gaudreault P, Parent M, McGuigan MA, Chicoine L and Lovejoy FH. 1983
    Predicability of eosophageal injury from signs and symptoms: a study
    of caustic ingestions in 378 children.
    Pediatrics 71 (5):767-770

    Gill RA, Pentel PR and Merritt VE. 1986
    Encapsulated alkali ingestion has a characteristic endoscopic
    appearance.
    Gastrointest Endoscopy 32 (2):121-122

    Gossot D, Sarfati E and Celerier M. 1987
    Early blunt esophagectomy in severe caustic burns of the upper
    digestive tract.
    J Thorac Cardiovasc Surg
    94:188-191

    Grant WM and Schuman JS. 1993
    Toxicology of the eye 4th ed.
    Charles C Thomas, Springfield Illinois

    Grenga TE. 1983
    A new risk of lye ingestion by children.
    N Engl J Med 308 (3):156-157

    Gundogdu HZ, Tanyel FC, Buyukpamukcu N and Hicsonmez A. 1992
    Conservative treatment of caustic esophageal strictures in children.
    J Pediatr Surg 27 (6):767-770

    Guth AA, Pachter HL, Albanese C and Kim U. 1994
    Combined duodenal and colonic necrosis. An unusual sequela of caustic
    ingestion.
    J Clin Gastroenterol 19 (4):303-305

    Hawkins DB, Demeter MJ and Barnett TE. 1980
    Caustic ingestion: controversies in management. A review of 214 cases.
    Laryngoscope 90:98-109

    Hendrickx L, Hubens A and Van Hee W. 1990
    Emergency oesophageal stripping, an aggressive approach to acute,
    necrotic caustic burns of the oesophagus and stomach.
    Acta Chir Belg 90:46-49

    Herbert K and Lawrence JC. 1989
    Chemical burns.
    Burns 15 (6):381-384

    Hoffman RS, Howland MA, Kamerow HN and Goldfrank LR. 1989
    Comparison of titratable acid/alkaline reserve and pH in potentially
    caustic household products.
    Clin Toxicol 27 (4&5):24261

    Homan CS, Maitra SR, Lane BP, Thode HC and Sable M. 1994
    Therapeutic effects of water and milk for acute alkali injury of the
    esophagus.
    Ann Emerg Med 241:14-20

    Hopkins RA and Postlethwait RW. 1981
    Caustic burns and carcinoma of the esophagus.
    Ann Surg 194:146-148

    Howell JM, Dalsey WC, Hartsell FW and Butzin CA. 1990
    Steroids for the treatment of corrosive esophageal injury. A
    statistical analysis of past studies.
    Am J Emerg Med 10:421-425

    Howell JM. 1986
    Alkaline ingestions.
    Ann Emerg Med 15:820-825

    Ingraham HJ, Perry HD and Donnerfeld ED. 1991
    Air-bag keratitis.
    N Engl J Med 324 (22):1599-1600

    Isolauri J and Markkula H. 1989
    Lye ingestion and carcinoma of the oesophagus.
    Acta Chir Scand 155:269-271

    Kinnman J, Shin HI and Wetteland P. 1968
    Carcinoma of the oesophagus after lye ingestion. Report of a case in
    15 year old Korean male.
    Acta Chir Scand 134:489-493

    Kirsh MM, Peterson A, Brown JW, Orringer MB, Ritter F and Sloan H.
    1978
    Treatment of caustic injuries of the esophagus.
    Ann Surg 188:675-678

    Klein-Schwartz W and Oderda GM. 1983
    Management of corrosive ingestions.
    Clin Toxicol Consult 5 (2):39-55

    Knopp R. 1979
    Caustic ingestions.
    JACEP 8:329-336

    Krenzelok EP and Clinton JE. 1979
    Caustic esophageal and gastric erosion without evidence of oral burns
    following detergent ingestion.
    JACEP 8:194-196

    Kynaston JA, Patrick MK, Shepherd RW, Raivadera PV and Cleghorn GJ.
    1989
    The hazards of automatic dishwasher detergent.
    Med J Aust 151:5-7

    Lacoutre PG, Gaudrealt and Lovejoy FH Jr. 1986
    Clintest tablet ingestion: an in vitro investigation concerned with
    intial emergency management.
    Ann Emerg Med 15:143-146

    Leape LL, Ashcraft KW, Scarpelli DG and Holder TM. 1971
    Hazard to health - liquid lye.
    N Engl J Med 284:578-581

    Lee KAP and Opeskin K. 1995
    Fatal alkali burns.
    Forensic Sci Int 72:219-227

    Leonard LG, Scheulen JJ, Munster AM. 1982
    Chemical burns: effects of prompt first aid.
    J Trauma 22 (5):420-423

    Lorette JJ and Wilkinson JA. 1988
    Alkaline chemical burn to the face requiring full thickness skin
    grafting.
    Ann Emerg Med 17:739-741

    Lowe JE, Graham DY, Boisaubin EV and Lanza FL. 1979
    Corrosive injury to the stomach: the natural history and role of
    fiberoptic endoscopy.
    Am J Surg 137:803-806

    Mallory A and Schaefer JW. 1977
    Clinitest ingestion.
    Br Med J 2:105-107

    Matenga JA, Russell J, Paul B and Thomas JEP.
    Suicide due to caustic soda - a case report with some unusual
    complications.
    Cent Af J Med 33 (3):77-79

    Mauger T. 1988
    Sodium hydroxide masquerading as a contact lens solution.
    Arch Ophthamol 106:1037

    McCabe RE, Scott JR and Knox WG. 1969
    Fistulisation between the esophageal, aorta and trachea as a
    complication of acute corrosive esophagitis: report of a case.
    Am Surg 35 (6):450-454

    Middlekamp JN, Ferguson TB, Roper CL and Hoffman FD. 1969
    The management and problems of caustic burns in children.
    J Thorac Cardiovasc Surg 57:341-347

    Moazam F, Talbert JL, Miller D and Mollitt DL. 1987
    Caustic ingestion and its sequelae in children.
    South Med J 80:187-190

    Moon MEL and Robertson IF. 1983
    Retrospective study of alkali burns of the eye.
    Aust J Opthalmol 11:281-286

    Moore WR. 1986
    Caustic ingestions. Pathophysiology, diagnosis and treatment.
    Clin Pediatr 25 (4):192-196

    Mutaf O, Avanoglu A, Mevsim A and Ozok G. 1995
    Management of tracheosophageal fistula as a complication of esophageal
    dilatations in caustic esophageal burns.
    J Pediatr Surg 30 (6):823-826

    Nelson JD and Kopietz LA. 1987
    Chemical injuries to the eyes. Emergency, intermediate and long-term
    care.
    Postgrad Med 81 (4):62-75

    Nelson R, Walson P and Kelley M. 1983
    Caustic ingestion.
    Ann Emerg Med 12:559-562

    Nuutinen M, Uhari M, Karvali T and Kouvalainen K. 1994
    Consequences of caustic ingestions in children.
    Acta Paediatr 83:1200-1205

    Oakes DD, Sherck JP and Mark JBD. 1982
    Lye ingestion. Clinical patterns and therapeutic implications.
    J Thorac Cardiovasc Surg 83:194-204

    O'Connor HJ, Dixon MF, Grant AC, Sooltan MA, Axon ATR and Henry JA.
    1984
    Fatal ingestion of Clinitest in an a adult.
    J R Soc Med 77:963-965

    O'Donoghue JM, Al-Ghazal SK and McCann JJ. 1996
    Caustic soda burns to the extremities: difficulties in management.
    BJCP 50 (2):108-109

    Okonek S, Bierbach H and Atzpodien W. 1981
    Unexpected metabolic acidosis in severe lye poisoning.
    Clin Toxicol 18 (2):225-230

    Ottoson A. 1981
    Late aortic rupture after lye ingestion.
    Arch Toxicol 47:59-62

    Payten RJ. 1972
    Clintest tablet stricture of the oesophagus.
    Br Med J 4:728-729

    Pfister RR. 1983
    Chemical injuries to the eye.
    Ophthalmology 90:1246-1253

    Pfister RR and Koshi J. 1982
    Alkali burns of the eye: pathophysiology and treatment.
    South Med J 75 (4):417-422

    Pfister RR, Haddox JL and Lank KM. 1988
    Citrate or ascorbate/citrate treatment of established corneal ulcers
    in the alkali-injured rabbit eye.
    Invest Ophthalmol Vis Sci 29 (7):1110-1115

    Pfister RR, Paterson CA and Hayes SA. 1982
    Effect of topical 10% ascorbate solution on established corneal ulcers
    after severe alkali injury.
    Invest Ophthamol Vis Sci 22:382-385

    Postlethwait RW. 1983
    Chemical burns of the esophagus.
    Surg Clin N Am 63 (4):915-924

    Rabinovitz M, Udekwu AO, Campbell WL, Kumar S, Razack A and Van Thiel
    DH. 1990
    Tracheosophageal-aortic fistula complicating lye ingestion.
    Am J Gastroenterol 85 (7):868-871

    Rappert P, Preier L, Korab Wa nd Neubauer T. 1993
    Diagnostic and therapeutic management of esophageal and gastric
    caustic burns in childhood.
    Eur J Pediatr Surg 3:202-205

    Ribet M, Chambon JP and Pruvot FR. 1990
    Oesophagectomy for severe corrosive injuries: is it always legitimate?
    Eur J Cardiovasc Surg 4:347-350

    Rosenberg N, Kunderman PJ, Vroman L and Moolten SE. 1953
    Prevention of experimental esophageal stricture by cortisone. II.
    Arch Surg 66:593-598

    Rubin AE, Bentur L and Bentur Y. 1992
    Obstructive airway disease associated with occupational sodium
    hydroxide inhalation.
    Br J Ind Med 49:213-214

    Rubin MM, Jui VJ and Cozzi GM. 1989
    Treatment of caustic ingestion.
    J Oral Maxillofac Surg 47:286-290

    Rumack BH and Burrington JD. 1977
    Caustic ingestion: a rational look at diluents.
    Clin Toxicol 11:27-34

    Shaffer RT, Carrougher JG, Kadakia SC and Levine SM.1994
    Update on caustic ingestions: how therapy has changed.
    J Crit Ill 9 (2):161-173

    Schlatter-Lanz I. 1985
    Veratzungen durch Testtabletten zur semiquantitativen
    Harnzuckerbestimmung (Clinitest).
    Schweiz Rundschau Med 74 (50):1402-1406

    Shikovitz MJ, Levy J, Villano D, Graver LM and Pochaczevsky R. 1996
    Speech and swallowing rehabilitation following devastating caustic
    ingestion: techniques and indicators for success.
    Laryngoscope 106 (Suppl 78):1-12

    Smally AJ, Binzer A, Dolin S and Viano D. 1992
    Alkaline chemical keratitis: eye injury from air bags.
    Ann emerg med 21 (11):1400-1402

    Spain DM, Molomut N and Haler A. 1950
    The effect of cortisone on the formation of granulation tissue in
    mice.
    Am J Pathol 26:710-711

    Swanson-Biearman B, Mvros R, Dean BS and Krenzelok EP. 1993
    Air bags: lifesaving with toxic potential?
    Am J Emerg Med 11:38-39

    Tewfik TL and Schloss MD. 1980
    Ingestion of lye and other corrosive agents - a study of 86 infant and
    child cases.
    J Otolaryngol 9 (1):72-77

    Thompson JN. 1987
    Corrosive esophageal injuries. I. A study of nine cases of concurrent
    accidental caustic ingestion.
    Laryngoscope 97:1060-1068

    Ti TK. 1983
    Oesophageal carcinoma associated with corrosive injury - prevention
    and treatment by oesophageal resection.
    Br J Surg 70:223-225

    Vilogi J, Whitehead B and Marcus SM. 1985
    Oven cleaner pads: new risk for corrosive injury.
    Am J Emerg Med 3:412-414

    Webb WR, Koutras P, Ecker RR and Sugg WL. 1970
    An evaluation of steroids and antibiotics in caustic burns of the
    esophagus.
    Ann Thorac Surg 9:95-102

    Whiter JE, McClafferty K, Orton RB, Tokarewicz AC and Nowak ES. 1995
    Ocular alkali burn associated with motor car air-bag activation.
    Can Med Assoc J 153 (7):933-934

    Wijburg FA, Beukers MM, Heymans HS, Bartelsman JF and den Hartog Jager
    FC. 1985
    Nasogastric intubation as sole treatment of caustic esophageal
    lesions.
    Ann Otol Rhinol Laryngol
    94:337-341

    Wijburg FA, Heymans HSA and Urbanus NAM. 1989
    Caustic esophageal lesions in childhood: prevention of stricture
    formation.
    J Pediatr Surg 24 (2):171-173

    Winek CL, Rozin L, Wahba WW and Rafizadeh V. 1995
    Ingestion of lye.
    Forensic Sci Int 73:143-147

    Wright P. 1982
    The chemically injured eye.
    Trans Ophthal Soc UK 102:85-87