UKPID MONOGRAPH ARSENIC ACID SM Bradberry BSc MB MRCP WN Harrison PhD CChem MRSC ST Beer BSc JA Vale MD FRCP FRCPE FFOM National Poisons Information Service (Birmingham Centre), West Midlands Poisons Unit, City Hospital NHS Trust, Dudley Road, Birmingham B18 7QH 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. ARSENIC ACID Toxbase entry Type of product Used in wood preservatives and insecticides. In water arsenic acid forms the arsenate ion. Toxicity Small ingestions of dilute (<3%) arsenate solutions usually are without serious adverse effects. A patient has survived the deliberate ingestion of 10 g arsenate (Mathieu et al, 1992). Features Systemic toxicity may follow arsenic acid ingestion, inhalation or topical exposure. Topical - Causes skin burns. Systemic arsenic poisoning may occur after substantial exposure. Ingestion Minor ingestions (small amounts of dilute (<3%) solutions): - Usually no serious effects. Mild gastrointestinal upset may occur. Substantial ingestions: - Rapid onset (within 1-2 hours) of burning of the mouth and throat, hypersalivation, dysphagia, nausea, vomiting, abdominal pain and diarrhoea. - In severe cases gastrointestinal haemorrhage, cardiovascular collapse, renal failure, seizures, encephalopathy and rhabdomyolysis may occur. - Other features: facial and peripheral oedema, ventricular arrhythmias (notably torsade de pointes), ECG abnormalities (QT interval prolongation, T-wave changes), muscle cramps. - Investigations may show anaemia, leucopenia, thrombocytopenia or evidence of intravascular haemolysis. - Death may occur from cardiorespiratory or hepatorenal failure. The adult respiratory distress syndrome (ARDS) has been reported. - Survivors of severe poisoning may develop a peripheral neuropathy and skin lesions typical of chronic arsenical poisoning. Inhalation - Rhinitis, pharyngitis, laryngitis and tracheobronchitis may occur. Tracheal and bronchial haemorrhage may complicate severe cases. Chronic arsenic exposure - may occur following ingestion, inhalation or topical exposure. Features include weakness, lethargy, gastrointestinal upset, dermal manifestations (hyperkeratosis and "raindrop" pigmentation of the skin), a peripheral (motor and sensory) neuropathy and psychological impairment. - Also reported: peripheral vascular disease (cold sensitivity progressing to ulceration and gangrene), renal tubular or cortical damage and haematological abnormalities (notably pancytopenia). Management Topical 1. Irrigate with copious volumes of water. 2. Treat burns conventionally. 3. Consider the possibility of systemic arsenic poisoning after significant exposure. Ingestion Minor ingestions: 1. Gastrointestinal decontamination is unnecessary. 2. Symptomatic and supportive care only. Substantial ingestions: 1. Most patients will vomit spontaneously but in those who do not, gastric lavage should be considered only if the patient presents within one hour. 2. Supportive measures are paramount. Intensive resuscitation may be required. Ensure adequate fluid replacement and close observation of vital signs including cardiac monitoring. 3. Diarrhoea can be controlled with loperamide. 4. Monitor blood urea, creatinine, electrolytes, liver function and full blood count. 5. Collect blood and urine for arsenic concentration measurements. 6. ECG evidence of QT prolongation may precede atypical ventricular arrhythmias (notably torsade de pointes). Avoid drugs which prolong the QT interval e.g. procainamide, quinidine or disopyramide. Isoprenaline is effective with phenytoin, lignocaine or propranolol as alternatives. 7. Antidotes - chelation therapy with either dimercaprol, DMSA or DMPS should be considered in symptomatic patients where there is analytical confirmation of the diagnosis, but only after specialist advice from the NPIS. References Armstrong CW, Stroube RB, Rubio T, Siudyla EA, Miller GB. Outbreak of fatal arsenic poisoning caused by contaminated drinking water. Arch Environ Health 1984; 39: 276-9. Donofrio PD, Wilbourn AJ, Albers JW, Rogers L, Salanga V, Greenberg HS. Acute arsenic intoxication presenting as Guillain-Barré-like syndrome. Muscle Nerve 1987; 10: 114-20. Engel RR, Hopenhayn-Rich C, Receveur O, Smith AH. Vascular effects of chronic arsenic exposure: a review. Epidemiol Rev 1994; 16: 184-209. Goldsmith S, From AHL. Arsenic-induced atypical ventricular tachycardia. N Engl J Med 1980; 303:1096-7. Greenberg C, Davies S, McGowan T, Schorer A, Drage C. Acute respiratory failure following severe arsenic poisoning. Chest 1979; 76: 596-8. Kersjes MP, Maurer JR, Trestrail JH, McCoy DJ. An analysis of arsenic exposures referred to the Blodgett Regional Poison Center. Vet Hum Toxicol 1987; 29: 75-8. Kingston RL, Hall S, Sioris L. Clinical observations and medical outcome in 149 cases of arsenate ant killer ingestion. Clin Toxicol 1993; 31: 581-91. Kosnett MJ, Becker CE. Dimercaptosuccinic acid as a treatment for arsenic poisoning. Vet Hum Toxicol 1987; 29: 462. Mathieu D, Mathieu-Nolf M, Germain-Alonso M, Neviere R, Furon D, Wattel F. Massive arsenic poisoning - effect of hemodialysis and dimercaprol on arsenic kinetics. Intensive Care Med 1992; 18: 47-50. McWilliams ME. Accidental acute poisoning by a concentrated solution of arsenic acid from percutaneous absorption: a case report. Vet Hum Toxicol 1989; 31: 354. Peterson RG, Rumack BH. D-penicillamine therapy of acute arsenic poisoning. J Pediatr 1977; 91: 661-6. Substance name Arsenic acid Origin of substance Reaction of arsenic trioxide with nitric acid (HSDB, 1995) Synonyms Orthoarsenic acid (DOSE, 1992) Chemical group A compound of arsenic, a group VA element Reference numbers CAS 7778-39-4 (DOSE, 1992) RTECS CG0700000 (RTECS, 1995) UN 1554 (solid) 1553 (liquid) (DOSE, 1992) HAZCHEM 2X (DOSE, 1992) Physicochemical properties Chemical structure H3AsO4 (DOSE, 1992) Molecular weight 141.94 (DOSE, 1992) Physical state at room temperature Solid (HSDB, 1995) Colour White (HSDB, 1995) Odour NIF Viscosity NA pH Weakly acidic (HSDB, 1995) Solubility 3020 g/L in water at 12.5°C (HSDB, 1995) Autoignition temperature NA Chemical interactions Aqueous solutions emit arsine gas when in contact with active metals such as arsenic, iron, aluminium and zinc. (HSDB, 1995) Major products of combustion Produces arsine gas (HSDB, 1995) Explosive limits NA Flammability Not flammable (HSDB, 1995) Boiling point NIF Density 2.2 at 20°C (HSDB, 1995) Vapour pressure NIF Relative vapour density NIF Flash Point NA Reactivity Arsenic acid is a strong oxidizer. (HSDB, 1995) Uses Preparation of arsenate salts Manufacture of pesticides (DOSE, 1992) Hazard/risk classification Index no. 033-005-00-1 Risk phrases Carc. Cat. 1; R45 - May cause cancer T: R23/25 - Also toxic by inhalation and if swallowed Safety phrases S53-45 - Avoid exposure-obtain special instructions before use. In case of accident or if you feel unwell seek medical advice immediately (show label where possible). EEC No: NIF (CHIP2, 1994) INTRODUCTION Arsenic acid is one of the most commercially important pentavalent compounds of arsenic. It is formed from the treatment of arsenic trioxide with nitric acid and used in the manufacture of arsenate salts and pesticides. It is also formed by the slow reaction of arsenic pentoxide in water (Fielder et al, 1986). Aqueous solutions of arsenic acid can liberate arsine gas, the most acutely toxic form of arsenic, when in contact with active metals (HSDB, 1995). It has been suggested that soluble arsenic compounds such as arsenic acid represent a much more acute toxic hazard than insoluble, poorly absorbed forms (Done and Peart, 1971). EPIDEMIOLOGY The main source of arsenic exposure in the world population is drinking water with an high inorganic arsenic concentration (Chiou et al, 1995; Das et al, 1995). In water arsenic acid forms the arsenate ion. Arsenate is the most thermodynamically stable oxide of arsenic in water and as such is believed to be the predominant species, especially under aerobic conditions (IPCS, 1981). Arsenic acid and its salts have been used widely in industry (Fielder et al, 1986; McWilliams 1989) and in pesticide formulations (Done and Peart, 1971; Miller et al, 1980) where they have been the source of acute and chronic, accidental and deliberate intoxication. MECHANISM OF TOXICITY Once absorbed pentavalent arsenic is reduced in vivo to trivalent arsenic. The principle mechanism of arsenic intoxication is disruption of thiol proteins. For example, trivalent arsenic inactivates pyruvate dehydrogenase by complexing with the sulphydryl groups of a lipoic acid moiety (6,8-dithiooctanoic acid) of the enzyme (Jones, 1995). Enhanced cellular destruction of damaged thiol proteins may produce toxic oxygen radicals. Arsenic-induced reduced lymphocyte proliferation (Gonsebatt et al, 1994) and impaired macrophage function also have been described (Lantz et al, 1994). Dong and Luo (1994) have suggested that while arsenic can directly damage DNA, a more important mechanism in arsenic-induced carcinogenicity is enhanced mutagenicity of other compounds via increased DNA-protein crosslinks. The affinity of arsenic for sulphydryl groups is utilized in chelation therapy. TOXICOKINETICS Absorption Soluble arsenical compounds such as arsenic acid are well absorbed after ingestion, when absorption is almost complete. Following inhalation there is significant mucociliary clearance and gastrointestinal absorption of respired particles (Fielder et al, 1986). From very limited animal data arsenic acid appears to be well absorbed through the lungs (Fielder et al, 1986). Although direct evidence of transcutaneous arsenic absorption in man is scarce (Fielder et al, 1986) there are reports of systemic arsenic toxicity following presumed dermal exposure (Garb and Hine, 1977; McWilliams, 1989). Distribution Absorbed arsenic is distributed to all body tissues (Fielder et al, 1986). Once absorbed pentavalent arsenic is reduced in vivo to trivalent arsenic. Trivalent arsenic is methylated in the liver to methylarsonic acid and dimethylarsinic acid (IPCS, 1996). Short-term studies on humans indicate that daily intake in excess of 0.5 mg progressively, but not completely, saturates the capacity to methylate inorganic arsenic (IPCS, 1996). Excretion The half-life of arsenic in blood is about 60 hours with renal excretion predominantly as mono- and dimethyl-derivatives (Waldron and Scott, 1994). The whole body half-life of arsenic in six human volunteers fitted a three compartment system, with 65.9 per cent of orally administered arsenic acid having a half-life of 2.1 days, 30.4 per cent a half-life of 9.5 days and 3.7 per cent a half-life of 38.4 days (mean values) (Pomroy et al, 1980). CLINICAL FEATURES: ACUTE EXPOSURE Dermal exposure Severe foot burns occurred in a patient exposed to concentrated arsenic acid-saturated clothing for eight hours (McWilliams, 1989). Soft tissue deposits believed to be metallic arsenic were noted on X-ray. The patient was transiently encephalopathic and developed a chronic painful foot motor neuropathy. Twenty four hour urine arsenic elimination ranged from 2500 mg falling to 160 mg during an eight week course of penicillamine. Garb and Hine (1977) reported a worker splashed down the left side of his body with arsenic acid in an industrial accident. He immediately removed a contaminated glove and washed his hands but did not notice arsenic acid in his left shoe for some ten minutes. After washing the affected leg and removing a soiled sock he continued to work for a further four hours by which time he had sustained second degree burns in the acid-exposed areas. Eleven hours after the incident he developed pain and swelling at the contact sites plus nausea, vomiting, diarrhoea and abdominal pain necessitating hospital admission. During the next three days his vision became "foggy" and he complained of a sore tongue and "aching teeth". After seven days he developed a burning sensation in the unexposed foot accompanied by paraesthesiae in all extremities. He became progressively weaker and was confined to a wheelchair. Two years after the accident diminished touch sensation and tendon reflexes persisted and he could walk only with a left leg brace and crutches. Chelation therapy was not employed at any time. Ocular exposure Arsenic acid is an eye irritant and may cause burns. Most injuries result from exposure to dusts, causing conjunctivitis, lacrimation, photophobia and chemosis (Grant and Schuman, 1993). Ingestion The oral toxicity of arsenic acid is dependent on the amount ingested and the concentration of the preparation. Tallis (1989) noted that arsenate salts can react with hydrochloric acid in the stomach to liberate arsenic acid and the corresponding metal chloride. In this way poorly soluble salts such as lead arsenate can be a source of arsenic acid which is soluble and well absorbed. Although pentavalent arsenic is reduced in vivo to the generally more toxic trivalent arsenic (Waldron and Scott, 1994) ingestion of dilute arsenic acid salt solutions (less than three per cent) usually are without serious adverse effects (Kingston et al, 1993). In 149 such cases involving sodium arsenate (2.28 per cent) ant killer, 97 per cent of patients were asymptomatic and only one required hospital admission (Kingston et al , 1993). Gastrointestinal toxicity Of 57 cases of ant killer ingestion involving arsenic acid salts (maximum arsenate concentration three per cent) only seven patients were symptomatic. All these vomited, with abdominal pain, diarrhoea and nausea also reported (Kersjes et al 1987). Armstrong et al (1984) reported eight family members (two of whom died) poisoned by well-water which contained arsenic 108 mg/L (form unknown). All had gastrointestinal symptoms including dry mouth, vomiting, dysphagia and/or diarrhoea. Urine arsenic concentrations were directly related to the amount of water consumed. The estimated daily dose of arsenic ingested by the surviving family members ranged from 26-127 mg. An 11 year-old girl and a 27 year-old man died after ingesting approximately 77 mg and 166 mg arsenic daily respectively (duration unknown). Autopsy of the most severely poisoned patient showed diffuse gastrointestinal tract inflammation. A 32 year-old man ingested 900 mg of an arsenic acid salt, vomited within one hour and developed diarrhoea three hours later. His clinical course was complicated by hypotension and renal failure but after 82 days chelation therapy with N-acetylcysteine he fully recovered (Martin et al, 1990). Another patient survived the deliberate ingestion of 10 g of an arsenic acid salt (Mathieu et al, 1992). Severe nausea, vomiting and abdominal tenderness developed within three hours with cardiovascular collapse and subsequent acute renal failure requiring haemodialysis. The patient made a full recovery over three months. Other gastrointestinal features of arsenic poisoning include burning of the mouth and throat with dysphagia (Heyman et al, 1956) and hypersalivation (Schoolmeester and White, 1980). Hepatotoxicity Armstrong et al (1984) reported a 27 year-old man poisoned by well water containing 108 mg/L arsenic. He presented after feeling unwell for six days and on examination was jaundiced with a bilirubin concentration of 120 µmol/L. He collapsed and died later that day and autopsy showed a liver arsenic concentration of 86 mg/kg. He had ingested an estimated 166 mg arsenic daily (duration unknown) and had a urine arsenic concentration of 1.6 mg/L at post mortem. Seven other members of his family with lower arsenic exposure showed transiently elevated serum hepatic transaminase activities and total bilirubin concentrations (values not given). Schoolmeester and White (1980) reported a 16 year-old female who ingested 300 mg sodium arsenate in a suicide attempt. She developed severe abdominal pain and vomiting within 30 minutes. A 24 hour urine collection had an arsenic concentration of 14.2 mg/L (time of collection not stated). Forty-eight hours later serum liver transaminase and alkaline phosphatase activities were elevated (values not given) but these abnormalities resolved within six months. Nephrotoxicity Hypotension (Martin et al, 1990; Mathieu et al, 1992) or rhabdomyolysis following substantial arsenic acid ingestion may precipitate renal failure. Renal cortical necrosis has been described (Gerhardt et al, 1978). Haematuria was reported in one patient in a series of 57 cases of sodium arsenate ingestion (Kersjes et al, 1987). Pyuria, proteinuria and elevated serum creatinine concentrations were reported in members of a family poisoned with arsenic-contaminated well-water (Armstrong et al, 1984). The most severely poisoned patient, who died after a six day illness, developed gross haematuria, a serum creatinine concentration of 390 µmol/L and heavy proteinuria. He had consumed an estimated 166 mg arsenic daily (duration unknown) and had a urine arsenic concentration at post mortem of 1.6 mg/L. No chelation therapy was given. The other family members were treated with dimercaprol and penicillamine. Six eventually recovered. Cardiovascular toxicity Tachycardia is reported frequently following ingestion of arsenic acid salts and is contributed to by anxiety, intravascular fluid depletion and possibly direct arsenic-induced cardiotoxicity (Le Quesne and McCleod, 1977; Martin et al, 1990; Cullen et al, 1995). Ventricular arrhythmias, notably torsade de pointes (Beckman et al, 1991) have been observed. Other ECG abnormalities include prolongation of the QT interval (Goldsmith and From, 1980; Schoolmeester and White, 1980), idioventricular rhythm (Armstrong et al, 1984) and non-specific T wave changes. Sudden onset bradycardia then asystole has been reported following massive acute arsenic ingestion despite vigorous resuscitation and no earlier arrhythmia. Armstrong et al (1984) reported a 27 year-old man who had consumed an unstated quantity of well-water containing 108 mg/L arsenic. After a six day illness (diagnosed as an upper respiratory tract infection) he collapsed and sustained a respiratory arrest and seizures. An idioventricular rhythm was noted. He was resuscitated but remained comatose and died a few hours later. Pericardial effusion with tamponade was reported in another member of the same family who had also drunk arsenic-contaminated water; the outcome in this case was not stated (Armstrong et al, 1984). Neurotoxicity In 57 sodium arsenate ingestions involving solutions containing 1.5-3.0 per cent arsenate, headache, dizziness, lethargy and somnolence were each reported in 2 per cent of cases; 88 per cent of patients were asymptomatic (Kersjes et al 1987). More substantial arsenic ingestions have caused muscle cramps, a sensorineural hearing deficit (Goldsmith and From, 1980), encephalopathy (Jenkins, 1966) and seizures. A peripheral sensory and/or motor neuropathy has been described in survivors of severe acute arsenic poisoning although this is more typical following chronic exposure. Armstrong et al (1984) reported eight family members poisoned with well-water containing 108 mg/L arsenic. All developed gastrointestinal symptoms with "altered mental status" and seizures noted in four. Coma developed in three patients and a peripheral neuropathy in two. Goebel et al (1990) demonstrated acute wallerian degeneration of myelinated nerve fibres in a patient who developed a symmetrical polyneuropathy after attempting suicide by ingesting arsenic. Clinical improvement was associated with microscopic evidence of neurological regeneration. A 46 year-old man developed feet numbness ten days after drinking a solution of sodium arsenate (concentration unknown) in attempted suicide. Two months after ingestion neurological examination demonstrated distal muscle weakness bilaterally, absent knee and ankle reflexes and reduced position and vibration sense with a high-stepping gait. Sixteen months later there was improvement in both sensory and motor deficits although residual disability was evident at eight year follow-up (Le Quesne and McCleod, 1977). Dermal toxicity Le Quesne and McCleod (1977) described a patient who developed a papular erythematous rash and generalized epidermal desquamation one week after drinking 10 mL of an arsenate solution (concentration unknown). Striate leukonychia (Mees' lines) and hyperkeratotic or hyperpigmented skin lesions are characteristic of chronic arsenic intoxication but have been described also following substantial acute ingestion (Heyman et al, 1956; Kyle and Pease, 1965; Jenkins, 1966). Facial and peripheral oedema have also been reported following arsenic ingestion (Heyman et al, 1956; Kyle and Pease, 1965). Haemotoxicity In moderate or severe arsenic poisoning investigations typically show anaemia, leucopenia or pancytopenia (Kyle and Pease, 1965; Armstrong et al, 1984). There may be evidence of intravascular haemolysis and the blood film often shows basophilic stippling (Kyle and Pease, 1965). Mathieu et al (1992) described a 30-year-old male who ingested 10 g sodium arsenate with suicidal intent. He developed severe gastrointestinal features of arsenic poisoning within hours and required haemodialysis for management of acute renal failure. Five days after ingestion he developed thrombocytopenia and anaemia. Bone marrow examination showed maturation arrest but recovery ensued over 10 days. Multi-organ toxicity Severe acute arsenic poisoning may result in death from cardiorespiratory or hepatorenal failure (Jenkins, 1966; Armstrong et al, 1984; Campbell and Alvarez, 1989; Morton and Dunnette, 1994). The adult respiratory distress syndrome (ARDS) has been described (Bolliger et al, 1992). Inhalation Inhalation of arsenic compounds causes rhinitis, pharyngitis, laryngitis and tracheobronchitis (Morton and Dunnette, 1994). CLINICAL FEATURES: CHRONIC EXPOSURE Dermal exposure Occupational exposure may lead to chronic arsenical toxicity. Contact dermatitis has been reported in workers exposed to arsenic acid salts used in crystal manufacture (Barbaud et al, 1995). Ingestion Ingestion of arsenic-contaminated drinking water (Feinglass, 1973; Chiou et al, 1995), illicit whisky (Moonshine) (Gerhardt et al, 1980) "tonics" or traditional remedies have caused chronic arsenical poisoning. Inhalation Occupational exposure may lead to chronic arsenical poisoning. Perforation of the nasal septum has been reported. Systemic arsenic acid toxicity The systemic features observed are similar for each source of exposure which are considered together. General toxic effects Patients with chronic arsenic acid poisoning may present with general debility, progressive weakness (Feinglass, 1973; Gerhardt et al, 1980) fever and sweats (Heyman et al, 1956). Dermal toxicity The characteristic dermal manifestations are hyperkeratosis and "raindrop" pigmentation of the skin (Heyman et al, 1956; Kyle and Pease, 1965; Shannon and Strayer, 1989). Hyperkeratoses appear as multiple small nodules which may coalesce to form plaques and are found most commonly on the palms and soles. By contrast, hyperpigmentation is more prominent in the axilla, groin, areola and around the waist, typically with mucosal sparing (Shannon and Strayer, 1989). These changes seem to be exacerbated by poor nutritional status (Das et al, 1995). Hyperkeratotic lesions may develop into squamous cell carcinomas which are notable for their occurrence on non light-exposed areas of the upper extremities and trunk (Shannon and Strayer, 1989). The fingernails may become brittle with transverse white striae (Mees' lines) (Mees, 1919; Heyman et al, 1956; Kyle and Pease, 1965; Gerhardt et al, 1980). Exfoliative dermatitis (Nicolis and Helwig, 1973) has been reported. Neuropsychological toxicity A symmetrical peripheral neuropathy is typical. Sensory symptoms predominate with paraesthesiae, numbness and pain, particularly of the soles of the feet, extending in a "glove and stocking" distribution (Jenkins, 1966; Gerhardt et al, 1980). Motor involvement with symmetrical distal limb weakness, muscle atrophy and loss of deep tendon reflexes is recognized (Heyman et al, 1956; Gerhardt et al, 1980; Bansal et al, 1991). Complete respiratory muscle paralysis (Greenberg et al, 1979; Gerhardt et al, 1980), a phrenic neuropathy (Bansal et al, 1991) and cranial nerve involvement (Schoolmeester and White, 1980) have been reported. The neuropathy may be confused with the Guillain-Barré syndrome (Kyle and Pease, 1965; Donofrio et al, 1987). Gastrointestinal symptoms and skin manifestations suggest arsenic poisoning, while a high CSF protein concentration and cranial nerve involvement are more typical of the Guillain-Barré syndrome. Electromyelography may show reduced peripheral nerve conduction velocities in the absence of symptoms. Psychological impairment is widely reported in chronic arsenical poisoning with defects of verbal learning ability and memory and personality changes (Heyman et al, 1956; Schoolmeester and White, 1980). Hutton et al (1982) described a case of chronic self-intoxication with sodium arsenate ant poison. The patient was initially admitted with gastrointestinal symptoms and pancytopenia. He subsequently developed severe peripheral neuropathy and myelopathy. Urinalysis revealed an arsenic concentration of 3600 mg/L. The patient eventually admitted self-administering arsenic in order to secure early retirement on medical grounds. Gastrointestinal toxicity Nausea is common in chronic arsenical poisoning. In the German literature Reinl (1970) reported a 53 year-old man who had been exposed to arsenic acid used as an oxidizing agent in chemical manufacture. During his ten year employment he developed symptoms including diarrhoea and vomiting. He died of lung cancer believed to be related to arsenic exposure. Hepatotoxicity Abnormal liver enzyme activities (Schoolmeester and White, 1980) have been observed in chronic arsenic poisoning. Arsenic-induced cirrhosis has been described but may be explained by concomitant excess ethanol consumption (Morton and Dunnette, 1994). Narang (1987) suggested increased arsenic consumption as a contributing factor in the aetiology of liver disease in the Indian population when he found significantly increased hepatic arsenic concentrations at autopsy in 178 patients dying from cirrhosis, non cirrhotic portal fibrosis, fulminant hepatitis, Wilson's disease or alcoholic liver disease. In the German literature cirrhosis and splenomegaly were reported at autopsy of a 53 year-old man occupationally exposed to arsenic acid for ten years. Dermal and gastrointestinal symptoms occurred during his employment and arsenic induced lung cancer was cited as the cause of death (Reinl, 1970). Nephrotoxicity Renal manifestations probably reflect capillary damage and include haematuria, proteinuria with casts and acute tubular or cortical necrosis (Morton and Dunnette, 1994). Peripheral vascular and cardiovascular toxicity "Black foot disease" refers to a severe form of peripheral vascular disease seen in Taiwan in those who drink artesian well water with an high arsenic concentration. Initial paraesthesiae and cold sensitivity progress to ulceration and gangrene (Chiou et al, 1995). It has been suggested that mortality due to all vascular diseases may be increased in these populations (Chen and Lin, 1994; Engel et al, 1994). Raynaud's syndrome has also been described in those chronically exposed to arsenic dust. Several authors refer to the myocardial toxicity of arsenic (Schoolmeester and White, 1980; Hall and Harruff, 1989) which has been attributed to impaired oxidative metabolism of myocardial tissue plus a direct arsenic-induced inflammatory process. A 42 year-old agricultural worker presented with systemic features of chronic arsenic poisoning (neuropathy and skin lesions) and had a 24 hour urine arsenic excretion of 7000 µg (Hall and Harruff, 1989). He received a 15 day course of dimercaprol with some improvement in motor function. On the 26th day of hospital admission he suddenly collapsed and died following a cardiac arrest. At post-mortem he had a diffuse interstitial myocarditis which was assumed to have triggered a fatal arrhythmia. Haemotoxicity Pancytopenia has been reported in cases of chronic intoxication by arsenic acid salts (Schoolmeester and White, 1980; Hutton et al, 1982). Anaemia, neutropenia (Heyman et al, 1956; Kyle and Pease, 1965), or evidence of haemolysis (Kyle and Pease, 1965) have also been reported as have macrocytosis without anaemia (Heaven et al, 1994) and a myelodysplastic syndrome (Rezuke et al, 1991). Chronic arsenic exposure complicated by aplastic anaemia may predispose to acute myeloid leukaemia (Kjeldsberg and Ward, 1972). Disrupted haem metabolism with altered urinary porphyrin excretion (Garcia-Vargas et al, 1994) has been reported. Pulmonary toxicity An irritating cough and haemoptysis have been reported in chronic arsenic intoxication (Heyman et al, 1956). Endocrine toxicity Epidemiological evidence from Taiwan (Lai et al, 1994) has recently associated chronic arsenic exposure with the development of diabetes mellitus. MANAGEMENT Dermal exposure Surface decontamination should be attempted where necessary. Treat burns conventionally. Consider the possibility of systemic arsenic poisoning and the need for chelation therapy (see below). Ocular exposure Irrigate the eye with copious lukewarm water. A topical anaesthetic may be necessary for pain relief. Seek an ophthalmic opinion if symptoms persist or examination is abnormal. Ingestion Decontamination After acute ingestion of a substantial quantity of arsenic acid most patients will vomit spontaneously but, in those who do not, gastric lavage should be considered only if it is possible to undertake the procedure within the first hour. Supportive measures Severe acute arsenic acid poisoning requires prompt intensive resuscitation with adequate fluid replacement and close observation of vital signs including cardiac monitoring. Diarrhoea may be treated symptomatically with loperamide. Chelation therapy should be considered in symptomatic cases. Obtain blood and urine for arsenic concentration determination. Electrocardiographic evidence of QT prolongation in arsenic poisoning may precede atypical ventricular arrhythmias, notably torsade de pointes, and in these circumstances drugs which themselves prolong the QT interval, such as procainamide, quinidine or disopyramide, should be avoided. Isoprenaline is effective; phenytoin, lignocaine or propranolol are alternatives (Goldsmith and From, 1980). Inhalation Immediate management involves removal from exposure and administration of supplemental oxygen if necessary. Evidence of systemic arsenic uptake should be sought and chelation therapy considered as discussed below. Antidotes Chelating agents used in the treatment of arsenic poisoning are dithiol compounds which can remove arsenic from endogenous sulphydryl groups, the targets of arsenic toxicity (Jones, 1995). Traditionally, dimercaprol (British anti-lewisite, BAL) has been the recommended chelator in arsenic intoxication (Jenkins, 1966; Greenberg et al, 1979; Roses et al, 1991). However, dimercaprol may produce unpleasant adverse effects and must be administered by deep intramuscular injection. There is increasing evidence that dimercaptosuccinic acid (DMSA, Succimer) (Aposhian et al, 1984; Graziano, 1986; Fournier et al, 1988; Inns et al, 1990) and dimercaptopropane sulphonate (DMPS, Unithiol) (Aposhian, 1983; Aposhian et al, 1984; Hruby and Donner, 1987; Inns et al, 1990) are less toxic and may be preferable. DMSA and DMPS are more effective in reducing the arsenic content of tissues, they increase biliary as well as urinary arsenic elimination and, unlike dimercaprol, do not appear to cause arsenic accumulation in the brain (Kreppel et al, 1990; Moore et al, 1994). On the other hand, arsenic mercaptide (the chelation complex of dimercaprol and arsenic) is dialysable and hence dimercaprol may be preferred in the presence of renal failure (Sheabar et al, 1989; Mathieu et al, 1992) The importance of an increased urine arsenic concentration in determining the need for chelation therapy is disputed. Kersjes et al (1987) suggested a spot urine concentration greater than 200 µg/L should be taken as an indication of "significant" arsenic exposure but Kingston et al (1993) emphasised that arsenic concentrations significantly higher than this (3500 µg/24 h and 5819 µg/24 h in two of their patients) may be observed in the acute phase following pentavalent arsenic ingestion without severe sequelae. Dimercaprol (British anti-lewisite; BAL) Dimercaprol was developed during the Second World War as an antidote for lewisite (dichloro(2-chlorovinyl) arsine) poisoning (Peters et al, 1945). It possesses two sulphydryl groups and forms a stable mercaptide ring with arsenic. The alcohol group on dimercaprol confers some degree of water solubility, thereby enhancing excretion from the body. As the chelation complex tends to dissociate it is necessary to maintain a constant excess of dimercaprol. Unlike DMSA and DMPS, dimercaprol is also lipid soluble and increases the brain arsenic concentration in arsenic-intoxicated animals (Jones, 1995). Though increasingly superseded by the less toxic thiol chelating agents, intramuscular dimercaprol remains useful in severe arsenic poisoning where vomiting prevents oral antidote administration, supplies of DMSA or DMPS are not rapidly available (Jolliffe et al, 1991) or renal failure requires haemodialysis; dimercaprol but not DMSA chelates can cross the dialysis membrane (Sheabar et al, 1989; Mathieu et al, 1992). Animal studies Stocken and Thompson (1946) demonstrated increased urine arsenic excretion (up to 33.5 per cent of the amount applied) in the 24 hours following cutaneous application of lewisite to rodents, when dimercaprol (dose not stated) was spread over the affected area up to one hour later. Dimercaprol also prevented arsenic-induced diarrhoea observed in control animals. Intravenous injection of dimercaprol glucoside 1.5 g/kg prevented death in two rabbits poisoned with cutaneous lewisite (12 mg/kg). Eleven control animals died, as did two treated with subcutaneous dimercaprol 0.07 g/kg (Danielli et al, 1947). A recent study has demonstrated that intramuscular dimercaprol protects rabbits against the lethal systemic effects of intravenously administered lewisite. No appreciable difference was found between the protective effect of dimercaprol and that of water soluble analogues DMPS and DMSA (Inns et al, 1990). Clinical studies In a case series, 12 men were exposed to smoke containing diphenylcyano-arsenic (1.6 mg/m3), "other forms of organic arsenic" (0.5 mg/m3) and "inorganic arsenic" (1.8 mg/m3) for six minutes. They were treated with 3.5 mg/kg intramuscular dimercaprol 6.5-78 hours post exposure. Urine arsenic excretion increased by an average of 40 per cent between two and four hours after the injection. The largest increase, both absolute and relative, was observed in those treated earliest (6.5 hours after exposure) (Wexler et al, 1946). Giberson et al (1976) described the treatment of a 44 year-old male who ingested 400 mg sodium arsenite. Intramuscular dimercaprol 250 mg was administered every four hours. Haemodialysis was initiated in response to renal failure with 3.3 mg arsenic removed over four hours. By the sixth day, when renal function had recovered, arsenic excretion had reached 75 mg/24h with at least 115 mg arsenic excreted between days two and six. A four year-old boy who had ingested an unknown amount of arsenic trioxide rat poison was treated with dimercaprol 5 mg/kg every four hours for 16 hours. The urine contained 2,120 µg arsenic over the first 12 hours. He developed an urticarial rash over the lower extremities which subsided with the discontinuation of dimercaprol. The urine arsenic concentration decreased gradually during d-penicillamine treatment (Peterson and Rumack, 1977). Schoolmeester and White (1980) reported a 16 year-old female who ingested 300 mg sodium arsenate in a suicide attempt. She received intramuscular dimercaprol 125 mg every four hours for the first 24 hours, then twice daily for 24 hours. A 24 hour urine arsenic concentration (starting time not specified) was 14,200 µg/L. The effect of chelation therapy on arsenic excretion is not known but the patient fully recovered. Mahieu et al (1981) described a 44 year-old male who ingested an unknown amount of arsenic trioxide which had been mistaken for sugar. The dose "certainly exceeded 1000 mg". Intramuscular dimercaprol 2.5-4 mg/kg tds was administered for 21 days. Initial arsenic excretion was low due to renal insufficiency but increased to 10 mg/24h from three to seven days post ingestion. The patient excreted a total of 129 mg arsenic during his 26 days in hospital. A 40 year-old woman poisoned at the same time and treated with the same regimen for 17 days excreted 16.7 mg arsenic on the first day, the amount decreasing on subsequent days. Seventy three milligrams arsenic were eliminated over three weeks. A 32 year-old man who ingested 900 mg sodium arsenate in a suicide attempt commenced treatment with intramuscular dimercaprol 5 mg/kg four hourly five hours later (Bansal et al, 1991). Dimercaprol was stopped on day four. This patient also received oral d-penicillamine and intravenous then oral N-acetylcysteine between days two and 82 post ingestion. The urine arsenic concentration rose on the second hospital day then declined progressively during the next week although the data were incomplete and uninterpretable. A 22 month-old female who developed diarrhoea, vomiting and lethargy after ingesting approximately 0.7 mg sodium arsenate was treated initially with one intramuscular dose of dimercaprol 3 mg/kg nine hours post ingestion (Cullen et al, 1995). Three hours later the infant was asymptomatic and dimercaprol therapy discontinued although she subsequently received oral d-penicillamine then oral DMSA to treat persisting high urine arsenic concentrations (4880 µg/L in the first 24 hours after admission). On the third hospital day the urine arsenic concentration (from a 24 hour collection) was 1355 µg/L and fell progressively to 96 µg/L on day 12. These data do not enable any conclusions to be drawn regarding enhanced arsenic elimination. No benefit from dimercaprol was reported by McCutchen and Utterback (1966) in the treatment of severe chronic arsenic poisoning. Other authors have reported disappointing results with dimercaprol in the management of arsenic neuropathy (Heyman et al, 1956) although Jenkins (1966) described "no detectable disability" 18 months after acute sodium arsenite ingestion in a patient who developed a peripheral neuropathy and received "a full course of dimercaprol" (details not given). Marcus (1987) described a 16 year-old male who survived ingestion of 56 mg arsenic trioxide following treatment with intramuscular dimercaprol 4 mg/kg every four hours (duration not stated). The maximum urine arsenic excretion was "over 50 mg/day" falling to 20 µg/day by day 31. At twelve month follow-up neurological effects persisted. Mahieu et al (1981) suggested that a high (greater than 90 per cent) proportion of methylated arsenic in the urine of poisoned patients could be used to indicate a late presentation with less likelihood of benefit from chelation therapy. Treatment protocol Dimercaprol must be given by deep intramuscular injection. After injection 90 per cent of an administered dose is absorbed and Cmax is attained within one hour (Peters et al, 1947). Dimercaprol is distributed throughout the intracellular space and metabolic degradation and excretion is complete in less than four hours. Depending on severity, 2.5-5 mg/kg should be administered four hourly for two days. This is to ensure that a constant excess of dimercaprol is always present as the chelation complex dissociates. Traditionally, this initial treatment is followed by 2.5 mg/kg bd intramuscularly for one to two weeks. However, this is an empirical recommendation and may be insufficient in severe cases. Dosage and duration should be adjusted therefore, depending on urine arsenic removal. Adverse effects The most common adverse effect of dimercaprol is dose-related hypertension (with an increase in systolic pressure of up to 50 mmHg) which usually resolves within three hours of administration (Dollery, 1991) but may be associated with nausea, headache, sweating and abdominal pain. Gastrointestinal disturbance may also occur without hypertension. Conjunctivitis, paraesthesiae and fever have been described. Dimercaprol is contraindicated in severe liver disease since it is metabolized by glucuronidation with subsequent biliary excretion. DMSA DMSA is commercially available in some countries (though not the UK) mainly as meso-DMSA, although a DL-form also exists. Animal studies Aposhian et al (1984) demonstrated that DMSA was moderately more effective than DMPS (and substantially more effective than dimercaprol) in protecting mice from the lethal effects of sodium arsenite. DMSA mobilizes arsenic from tissues, increasing urine arsenic excretion without a rise in brain arsenic concentrations (Aposhian et al, 1984). Mice administered subcutaneous arsenic trioxide (5 mg/kg) followed immediately by intraperitoneal DMSA 100 mg/kg, showed significantly increased urine arsenic excretion (p<0.01) in the first 12 hours post chelation although the 48 hour urine arsenic elimination was not significantly different between DMSA-treated mice and controls (Maehashi and Murata, 1986). In animal studies DMSA protected against the embryotoxic effects of sodium arsenite but only when given within one hour of exposure (Bosque et al, 1991). Recent experiments suggest that oral monoester DMSA analogues may offer renal protection in arsenic poisoning by increasing the enteral arsenic content to enhance faecal rather than renal elimination (Hannemann et al, 1995). In other animal studies lipophilic DMSA analogues were inferior to the parent compound as arsenic antidotes (Kreppel et al, 1993). Clinical studies Lenz et al (1981) described a 46 year-old man who ingested 200 mg arsenic and survived following treatment with oral DMSA 300 mg qds for three days. Kosnett and Becker (1987) reported an increase in the 24 hour urine arsenic excretion from 26 µg to a maximum of 340 µg on the second day of oral DMSA 660 mg tds in a patient who presented more than 30 days after malicious acute arsenic ingestion. Nine days after ingesting approximately 0.7 mg of a soluble arsenic salt a 22 month-old female was treated with oral DMSA 30 mg/kg/day for at least four days (Cullen et al, 1995). The child had already received chelation therapy with dimercaprol and d-penicillamine, but further treatment was instituted because of a persistently raised urine arsenic concentration (650 µg/L on day five). Four days later the urine arsenic concentration had fallen to 96 µg/L. The authors reported an overall urine arsenic half-life of 2.6 days. Although the child initially experienced vomiting, diarrhoea and lethargy these features resolved within 12 hours and renal and hepatic function remained normal throughout (Cullen et al, 1995). There was no objective improvement in the neurological manifestations of chronic arsenic poisoning in a man poisoned by an ethnic remedy despite two weeks therapy with oral DMSA 400 mg tds (Kew et al, 1993). No urine arsenic excretion data were given. A 33 year-old woman with acute-on-chronic lead and arsenic poisoning from a herbal remedy clinically recovered following two one-week courses of oral DMSA 270 mg tds, though the effect of chelation therapy on urine arsenic excretion is difficult to interpret (Mitchell-Heggs et al, 1990). Treatment protocol DMSA is given orally in a dose of 30 mg/kg body weight per day; an intravenous preparation is available in some countries and may be preferable if the patient is vomiting (Hantson et al, 1995). Adverse effects Side-effects following treatment with DMSA are rare but include skin rashes, gastrointestinal disturbance, elevation of serum transaminase activities and flu-like symptoms (Reynolds, 1993). DMSA should be used with caution in patients with impaired renal function or a history of hepatic disease (Reynolds, 1993). DMPS Animal studies DMPS is commercially available as a racemic mixture of the dextro-rotatory and levo-rotatory forms which appear to be equally effective arsenic chelators (Aposhian, 1983), though animal studies suggest DMSA may be superior to either (Aposhian et al, 1984). Urine arsenic elimination of arsenic-poisoned rats in the 48 hours post treatment with DMPS 100 mg/kg intraperitoneally was significantly lower (p<0.05) than in either control (5 mg/kg subcutaneous arsenic trioxide only) or DMSA-treated mice (Maehashi and Murata, 1986). However DMPS significantly increased (p<0.01) faecal arsenic elimination in the 24 hours post chelation compared to control or DMSA treated mice, suggesting biliary excretion of the DMPS-arsenic chelate (Maehashi and Murata, 1986). Other authors have noted enhanced biliary but not faecal arsenic excretion following parenteral DMPS administration to arsenic-poisoned experimental animals. This suggests enterohepatic circulation of the chelate, which Reichl et al (1995) attempted to block using oral cholestyramine. They demonstrated enhanced faecal arsenic elimination (p<0.05) when intraperitoneal DMPS 0.1 mmol/kg and subcutaneous arsenic trioxide (0.02 mmol/kg) administration was followed by an oral combination of cholestyramine (0.2 g/kg) and DMPS 0.1 mmol/kg (Reichl et al, 1995). Domingo et al (1992) demonstrated a protective effect of DMPS 150-300 mg/kg, but not dimercaprol, against experimental arsenite-induced embryotoxicity and teratogenicity as judged by the incidence of foetal malformation or death in mice administered intraperitoneal sodium arsenite (12 mg/kg) on day nine of gestation. Clinical studies Two men inadvertently ingested 1 g and 4 g arsenic trioxide respectively (Moore et al, 1994). The more severely poisoned patient developed acute renal failure and 26 hours post ingestion had a blood arsenic concentration of 400 µg/L. He received intravenous DMPS 5 mg/kg every four hours for six days then oral DMPS 400 mg every four hours for one week. The other patient had a blood arsenic concentration of 98 µg/L, 36 hours post ingestion and received a shorter course of intravenous then oral DMPS. Both patients recovered fully but quantitative data showing the effect of chelation therapy on urine arsenic elimination were documented poorly. In another report there was no objective improvement in the neurological manifestations of chronic arsenic poisoning in a patient treated with oral DMPS 100 mg tds for three weeks (Kew et al, 1993). Treatment protocol DMPS is given orally or parenterally in a dose of 30 mg/kg body weight per day. Adverse effects Side effects following treatment with DMPS are infrequent but have included allergic skin reactions, nausea, vertigo and pruritis (Aposhian, 1983). d-Penicillamine Animal studies d-Penicillamine has been reported to be as effective as dimercaprol and NAC in prolonging the survival time of mice injected with a lethal dose of sodium arsenite (Shum et al, 1981). Other studies have disputed the validity of these results and have failed to demonstrate d-penicillamine as a useful chelator (Aposhian, 1982; Kreppel et al, 1989). Clinical studies Peterson and Rumack (1977) described three children who shared a bottle of rat poison containing arsenic trioxide 1.75 per cent. One died within hours following a rapidly deteriorating course of coma, convulsions and cardiac arrhythmias. The second, a four year-old male, presented with lethargy, a sinus tachycardia and tachypnoea. Oral d-penicillamine 25 mg/kg qds replaced dimercaprol treatment after 16 hours when the patient developed an urticarial rash over the lower extremities. The first twelve-hour urine collection during dimercaprol treatment contained 2,120 µg arsenic with the urine arsenic concentration decreasing during the five days d-penicillamine therapy. The child made a full recovery. The third patient (Peterson and Rumack, 1977) had no severe features of toxicity at presentation. He received the same chelation therapy regimen as patient 2. On the second day post ingestion the 24 hour urine arsenic excretion was 300 µg, increasing in the next 24 hours (the second day of d-penicillamine therapy) to approximately 800 µg. This patient also recovered fully. A one year-old child ingested 15-20 mg sodium arsenate (as ant poison) and was treated within six hours with 5 mg/kg intramuscular dimercaprol (Peterson and Rumack, 1977). The chelating agent was then changed to oral d-penicillamine 100 mg/kg/day and continued for five days. An initial 12 hour urine collection (commenced approximately six hours post ingestion) contained 192 µg arsenic, increasing to 2000 µg arsenic in the next 24 hours before falling to approximately 200 µg/24 h on day two. These authors advocated d-penicillamine 100 mg/kg/day as the treatment of choice in arsenic poisoning (where oral therapy is possible). They recommended d-penicillamine should be continued until the 24 hour urine arsenic excretion is less than 50 µg (Peterson and Rumack, 1977). A 16 month-old child was given a five day course of oral d-penicillamine 250 mg qds 14 hours after ingesting 9-14 mg arsenic trioxide (Watson et al, 1981). Clinical features of toxicity (diarrhoea, vomiting and lethargy) resolved within 24 hours and the child was discharged on day three. The arsenic concentration in urine collected during the first day of treatment was 560 µg/L. However, no earlier urine arsenic concentrations were measured and prior to d-penicillamine therapy the patient had received 185 mg dimercaprol over 18 hours (Watson et al, 1981). DiNapoli et al (1989) instituted d-penicillamine therapy in a patient unable to tolerate intramuscular dimercaprol following intravenous sodium arsenite injection. d-Penicillamine 500 mg tds was administered and after ten days a 24 hour urine arsenic excretion of 2 mg was reported. There were no symptoms of bone marrow depression, haemolysis or peripheral neuropathy. After a further ten days treatment the urine arsenic concentration was 20 µg/L. Bansal et al (1991) described a 35 year-old man with severe arsenic polyneuropathy involving the phrenic nerves bilaterally, who recovered following d-penicillamine therapy 250 mg tds for two weeks (route of administration was not stated). However, the 24 hour urine arsenic excretion only rose to 82.4 µg/g creatinine in the first 72 hours of chelation compared to a pretreatment value of 73.5 µg/g creatinine. Cullen et al (1995) reported a 22 month-old child who ingested some 0.7 mg sodium arsenate. Following a single dose of dimercaprol 3 mg/kg, oral d-penicillamine therapy was commenced, 250 mg qds for nine doses. By day four the 24 hour urine arsenic concentration had dropped from 4880 to 682 µg/L. The child was discharged on day six on oral d-penicillamine therapy (dose not stated) but readmitted three days later due to a persistently high urine arsenic excretion (650 µg/L on day five). At this stage d-penicillamine was replaced by DMSA since the child had developed an erythematous rash. Oral d-penicillamine 250 mg qds for seven days failed to increase urinary arsenic elimination in a patient with chronic arsenic poisoning whose initial 24 hour urine arsenic excretion was 342 µg (normal <5 µg/24 h) (Heaven et al, 1994). In another report the urine arsenic concentration in a 67 year-old man with arsenic-associated aplastic anaemia had risen to 20,246 µg/L after four days penicillamine therapy 500 mg qds compared to a pretreatment concentration of 7840 µg/L (Kjeldsberg and Ward, 1972). The patient died from acute myeloid leukaemia some six months later. N-acetylcysteine Animal studies The survival time of mice injected subcutaneously with a lethal dose of sodium arsenite (25 mg/kg) was increased significantly (p<0.05) if intraperitoneal N-acetylcysteine (NAC) 100 mg/kg was administered 30 minutes later. There was no significant difference between this dose of NAC, dimercaprol 5 mg/kg and d-penicillamine 50 mg/kg as an antidote under these conditions (Shum et al, 1981). Clinical studies Martin et al (1990) reported "remarkable clinical improvement" in a 32 year-old man with severe arsenic poisoning following ingestion of a soluble salt when he was administered intravenous NAC 70 mg/kg four hourly after dimercaprol had "failed to improve his condition". However urinary arsenic excretion data were poorly documented and dimercaprol was continued during treatment with NAC. Antidotes: Conclusions and recommendations 1. There are no controlled clinical trials of chelation therapy in arsenic poisoning and no conclusive evidence that dithiol antidotes reverse arsenic-induced neurological damage. On the present evidence it is difficult to recommend a single preferred antidote, though in the absence of renal failure DMSA may offer some advantages over other agents; if renal failure supervenes dimercaprol and haemodialysis should be employed. 2. Chelation therapy should be considered in symptomatic patients where there is analytical confirmation of the diagnosis. 3. Although urine arsenic concentrations are useful to confirm the diagnosis of arsenic poisoning chelation therapy should not be instituted on the basis of an increased urine arsenic concentration alone. Haemodialysis Haemodialysis removes arsenic from the blood but achieves less effective arsenic clearance than chelation therapy when normal renal function is present. It is indicated therefore only in the presence of renal failure. Giberson et al (1976) reported an arsenic dialysis clearance of 87 mL/min. During four hours of dialysis 3360 µg arsenic was removed in a patient with acute arsenic poisoning complicated by renal failure who was also receiving 250 mg intramuscular dimercaprol six times daily. The 24 hour urine arsenic excretion on the same day was 2030 µg though this increased to 75,000 µg/24 h on the sixth hospital day when renal function had recovered. A similar haemodialysis arsenic clearance of 76-87 mL/min was demonstrated in another patient with acute sodium arsenite intoxication complicated by acute renal failure (Vaziri et al, 1980). Levin-Scherz et al (1987) instituted haemodialysis promptly in a patient who presented 26 hours after ingesting 2 g arsenic trioxide. The patient also received intramuscular dimercaprol, 300 mg initially then 180 mg every four hours, but died within 72 hours of ingestion. The maximum amount of arsenic removed in the dialysate was 2.9 mg. Mathieu et al (1992) demonstrated a haemodialysis clearance comparable to some 40-77 per cent of the daily arsenic renal elimination on the day following diuresis recovery. In this case the total blood haemodialysis clearance (210 mL/min) exceeded the instantaneous plasma haemodialysis clearance (mean 85 mL/min), suggesting that some arsenic removed by haemodialysis originated in erythrocytes. These authors showed similar haemodialysis arsenic clearance with or without prior administration of intramuscular dimercaprol 250 mg, and advocated dimercaprol as the chelating agent of choice in arsenic poisoning complicated by renal failure, since it does not impair arsenic dialysis clearance. Experimental evidence in dogs (Sheabar et al, 1989) suggests DMSA- arsenic chelates do not pass through the dialyser membrane. Haemoperfusion A 37 year-old man presented within four hours of ingesting 90 mL of a 1.5 per cent arsenic trioxide solution (Smith et al, 1981). Although initially only tachycardic he subsequently became hypotensive and oliguric. For the first 48 hours he received 200 mg intramuscular dimercaprol four hourly then d-penicillamine 500 mg qds. Charcoal haemoperfusion was instituted 11 hours after admission followed by two hours haemodialysis. These therapies were repeated over the next four days but "discontinued because of continued good renal function and lack of clinical response". Serum arsenic concentrations immediately post haemoperfusion were slightly higher than pre-haemoperfusion values, suggesting no benefit. MEDICAL SURVEILLANCE Blood arsenic concentrations correlate poorly with exposure but may be useful in chronic poisoning (Morton and Dunnette, 1994). Arsenic concentrations in hair and nails have been used to indicate chronic systemic absorption, although their use as biological monitors of occupational exposure to airborne arsenic is limited by difficulty in excluding external contamination (Yamamura and Yamauchi, 1980). Urine arsenic concentrations are the most useful biomonitoring tool, ideally as 24 hour urine arsenic excretion collection, although spot urine arsenic concentrations have been proposed in screening asymptomatic patients with a history of possible acute arsenic ingestion (Grande et al, 1987). Since certain marine organisms (especially mussels) may contain large amounts of organoarsenicals, it is advisable that workers refrain from eating seafood for at least 48 hours before urine collection (Buchet et al, 1994). Analytical speciation methods capable of separating inorganic arsenic and its methylated derivatives from dietary organoarsenicals partially overcome this problem (Farmer and Johnson, 1990; Buchet et al, 1994). However, Vahter (1994) has suggested that under certain circumstances these compounds are released from seafood which can invalidate assessment of inorganic arsenic exposure. Farmer and Johnson (1990) found that high urine concentrations of inorganic arsenic plus its mono- and dimethyl derivatives corresponded to the possible workplace atmospheric arsenic concentrations for those involved in arsenic production or glass manufacture. Increased urine arsenic concentrations have also been noted in timber treatment workers using an arsenic-based wood preservative (Gollop and Glass, 1979). Telolahy et al (1993) suggested a potential role for increased urine coproporphyrins as an indicator of chronic occupational arsenic exposure since arsenic is known to disrupt haem metabolism. Regular examination of the skin should be included in an occupational health surveillance programme. Workers with evidence of excessive arsenic exposure should be offered long-term monitoring for the development of skin, bladder or lung cancer, though in practise this may be difficult to execute. OCCUPATIONAL DATA Maximum exposure limit Long-term exposure limit (8 hour TWA reference period) 0.1 mg/m3 (Health and Safety Executive, 1995). OTHER TOXICOLOGICAL DATA Carcinogenicity Individuals who chronically ingest arsenic have an increased risk of developing skin cancer, usually squamous cell carcinoma but also basal cell carcinomas (Schoolmeester and White, 1980; Chen et al, 1988; Shannon and Strayer, 1989; Chiou et al, 1995). Squamous cell carcinomas may arise in areas of arsenic-induced Bowen's disease (Shannon and Strayer, 1989). Hsueh et al (1995) demonstrated a significant dose-response relation between skin cancer prevalence and arsenic exposure from artesian well water. These authors identified chronic hepatitis B carriage and malnutrition as risk factors for arsenic-induced dermatological malignancy. Skin cancer has also been documented among vineyard workers and farmers exposed to inhaled inorganic arsenic in pesticides (Chen and Lin, 1994) although skin and gastrointestinal absorption probably contributed to arsenic toxicity in these cases. There is an association between chronic arsenic exposure and cancer of the urinary tract (Chen et al, 1988; Chen and Lin, 1994), lung (Chen and Lin, 1994) and liver, both hepatic angiosarcoma and hepatocellular carcinoma (Chen and Lin, 1994). Reinl (1970) reported the death of a 53 year-old man who had been exposed to arsenic acid for ten years. The compound had been used an as oxidizing agent in a chemical manufacturing process. He suffered gastrointestinal and dermal symptoms of arsenic toxicity during his employment and at autopsy arsenic induced metastatic bronchial carcinoma was diagnosed as the cause of death. Smoking exerts a synergistic effect with ingested and inhaled arsenic in the development of pulmonary malignancy. There is limited evidence that other internal cancers, particularly of the gastrointestinal tract and haematological malignancies, are linked aetiologically to arsenic exposure (Chen and Lin, 1994). Reprotoxicity Animal studies suggest arsenic is embryotoxic and teratogenic but reliable human data are scarce (Council on Scientific Affairs, 1985). Daya et al (1989) reported a 22 year-old female who ingested 340 mg sodium arsenate while 20 weeks pregnant. Treatment with dimercaprol 150 mg four hourly was commenced two hours post ingestion, the maximum 24 hour urine arsenic excretion was 3030 µg/L and a healthy infant was delivered at 36 weeks. A woman in the third trimester of pregnancy developed acute renal failure after ingesting a large quantity of an arsenical rat poison. Her baby was delivered on the fourth day post ingestion but died within a few hours from hyaline membrane disease. At autopsy the infant showed significant arsenic accumulation in the liver, brain and kidneys (liver arsenic concentration 0.74 mg/100 g tissue) (Lugo et al, 1969). Genotoxicity (sodium arsenate) Cultured human peripheral lymphocytes: Induced chromosomal aberrations and sister chromatid exchanges. Syrian hamster cells and human lymphocytes: Induced sister chromatid exchanges and chromosomal aberrations. Chinese hamster ovary cells: Induced chromosomal aberrations. Drosophilia melanogaster: Wing spot test negative (sodium arsenate is highly toxic to Drosophilia and hence could only be tested at very low concentrations) (DOSE, 1994). Fish toxicity (arsenic) EC50 (96 hr) fathead minnow 141-144 mg/L. LC50 96 hr) knifefish 31 mg/L. Oral administration (0.52 mg/kg/day for 24 weeks) to rainbow trout caused chronic inflammatory changes in subepithelial tissues of the gall bladder wall in 71 per cent of the group. LC50 (96 hr) striped base 30 mg (DOSE, 1992). EC Directive on Drinking Water Quality 80/778/EEC Maximum admissible concentration 50 µg/L, as arsenic (DOSE, 1992). WHO Guidelines for Drinking Water Quality Guideline value 10 µg/L, as arsenic (WHO, 1993). AUTHORS SM Bradberry BSc MB MRCP WN Harrison PhD CChem MRSC ST Beer BSc JA Vale MD FRCP FRCPE FRCPG FFOM National Poisons Information Service (Birmingham Centre), West Midlands Poisons Unit, City Hospital NHS Trust, Dudley Road, Birmingham B18 7QH UK This monograph was produced by the staff of the Birmingham 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. Date of last revision 17/1/97 REFERENCES Aposhian HV. Biological chelation: 2,3-Dimercapto-propanesulfonic acid and meso-dimercaptosuccinic acid. Adv Enzyme Regul 1982; 20: 301-19. Aposhian HV. DMSA and DMPS - water soluble antidotes for heavy metal poisoning. Ann Rev Pharmacol Toxicol 1983; 23: 193-215. Aposhian HV, Carter DE, Hoover TD, Hsu C-A, Maiorino RM, Stine E. DMSA, DMPS, and DMPA - as arsenic antidotes. Fundam Appl Toxicol 1984; 4: S58-70. Armstrong CW, Stroube RB, Rubio T, Siudyla EA, Miller GB. Outbreak of fatal arsenic poisoning caused by contaminated drinking water. Arch Environ Health 1984; 39: 276-9. Bansal SK, Haldar N, Dhand UK, Chopra JS. Phrenic neuropathy in arsenic poisoning. Chest 1991; 100: 878-80. Barbaud A, Mougeolle JM, Schmutz JL. Contact hypersensitivity to arsenic in a crystal factory worker. Contact Dermatitis 1995; 33: 272-3. Beckman KJ, Bauman JL, Pimental PA, Garrard C, Hariman RJ. Arsenic-induced torsade de pointes. Crit Care Med 1991; 19: 290-2. Bolliger CT, van Zijl P, Louw JA. Multiple organ failure with the adult respiratory distress syndrome in homicidal arsenic poisoning. Respiration 1992; 59: 57-61. Bosque MA, Piera V, Domingo JL, Corbella J. Meso-2,3-Dimercaptosuccinic acid (DMSA) alleviation of arsenite-induced embryofetotoxicity in mice: effectiveness with time. Toxicologist 1991; 11: 295. Buchet JP, Pauwels J, Lauwerys R. Assessment of exposure to inorganic arsenic following ingestion of marine organisms by volunteers. Environ Res 1994; 66: 44-51. Campbell JP, Alvarez JA. Acute arsenic intoxication. Am Fam Physician 1989; 40: 93-7. Chen C-J, Kuo T-L, Wu M-M. Arsenic and cancers. Lancet 1988; 1: 414-5. Chen C-J, Lin L-J. Human carcinogenicity and atherogenicity induced by chronic exposure to inorganic arsenic. In: Nriagu JO, ed. Arsenic in the environment. Part II: Human health and ecosystem effects. Vol 27. New York: John Wiley & Sons, Inc., 1994; 109-31. Chiou H-Y, Hsueh Y-M, Liaw K-F, Horng S-F, Chiang M-H, Pu Y-S, Lin J S-N, Huang C-H, Chen C-J. Incidence of internal cancers and ingested inorganic arsenic: a seven- year follow-up study in Taiwan. Cancer Res 1995; 55: 1296-300. CHIP2/Chemicals (Hazard Information and Packaging for Supply) Regulations1994. Health and Safety Commission. Sudbury: Heath and Safety Executive, 1994. Council on Scientific Affairs. Effects of toxic chemicals on the reproductive system. JAMA 1985; 253: 3431-7. Cullen NM, Wolf LR, St Clair D. Pediatric arsenic ingestion. Am J Emerg Med 1995; 13: 432-5. Danielli JF, Danielli M, Fraser JB, Mitchell PD, Owen LN, Shaw G. BAL-intrav: a new non-toxic thiol for intravenous injection in arsenical poisoning. Biochem J 1947; 41: 325-33. Das D, Chatterjee A, Mandal BK, Samanta G, Chakraborti D. Arsenic in ground water in six districts of West Bengal, India: the biggest arsenic calamity in the world. Part 2. Arsenic concentration in drinking water, hair, nails, urine, skin-scale and liver tissue (biopsy) of the affected people. Analyst 1995; 120: 917-24. Daya MR, Irwin R, Parshley MC, Harding J, Burton BT. Arsenic ingestion in pregnancy. Vet Hum Toxicol 1989; 31: 347. DiNapoli J, Hall AH, Drake R, Rumack BH. Cyanide and arsenic poisoning by intravenous injection. Ann Emerg Med; 1989: 308-11. Dollery C, ed. Therapeutic drugs. London: Churchill Livingstone, 1991; D148-51. Domingo JL, Bosque MA, Llobet JM, Corbella J. Amelioration by BAL (2,3-dimercapto-1-propanol) and DMPS (sodium 2,3-dimercapto1-propanesulfonic acid) of arsenite developmental toxicity in mice. Ecotoxicol Environ Safety 1992; 23: 274-81. Done AK, Peart AJ. Acute toxicities of arsenical herbicides. Clin Toxicol 1971; 4: 343-55. Dong J-T, Luo X-M. Effects of arsenic on DNA damage and repair in human fetal lung fibroblasts. Mutat Res 1994; 315: 11-5. Donofrio PD, Wilbourn AJ, Albers JW, Rogers L, Salanga V, Greenberg HS. Acute arsenic intoxication presenting as Guillain-Barré-like syndrome. Muscle Nerve 1987; 10: 114-20. DOSE/Dictionary of substances and their effects. Vol 1. Cambridge: Royal Society of Chemistry, 1992. DOSE/Dictionary of substances and their effects. Vol 7. Cambridge: Royal Society of Chemistry, 1994. Engel RR, Hopenhayn-Rich C, Receveur O, Smith AH. Vascular effects of chronic arsenic exposure: a review. Epidemiol Rev 1994; 16: 184-209. Farmer JG, Johnson LR. Assessment of occupational exposure to inorganic arsenic based on urinary concentrations and speciation of arsenic. Br J Ind Med 1990; 47: 342-8. Feinglass EJ. Arsenic intoxication from well water in the United States. N Engl J Med 1973; 288: 828-30. Fielder RJ, Dale EA, Williams SD. Toxicity review 16. Inorganic arsenic compounds. London: HMSO, 1986. Fournier L, Thomas G, Garnier R, Buisine A, Houze P, Pradier F, Dally S. 2,3-dimercaptosuccinic acid treatment of heavy metal poisoning in humans. Med Toxicol 1988; 3: 499-504. Garb LG, Hine CH. Arsenical neuropathy: residual effects following acute industrial exposure. J Occup Med 1977; 19: 567-8. Garcia-Vargas GG, Del Razo LM, Cebrián ME, Albores A, Ostrosky-Wegman P, Montero R, Gonsebatt ME, Lim CK, De Matteis F. Altered urinary porphyrin excretion in a human population chronically exposed to arsenic in Mexico. Hum Exp Toxicol 1994; 13: 839-47. Gerhardt RE, Hudson JB, Rao RN, Sobel RF. Chronic renal insufficiency from cortical necrosis induced by arsenic poisoning. Arch Intern Med 1978; 138: 1267-9. Gerhardt RE, Crecelius EA, Hudson JB. Moonshine-related arsenic poisoning. Arch Intern Med 1980; 140: 211-3. Giberson A, Vaziri ND, Mirahamadi K, Rosen SM. Hemodialysis of acute arsenic intoxication with transient renal failure. Arch Intern Med 1976; 136: 1303-4. Goebel HH, Schmidt PF, Bohl J, Tettenborn B, Krämer G, Gutmann L. Polyneuropathy due to acute arsenic intoxication: biopsy studies. J Neuropathol Exp Neurol 1990; 49: 137-49. Goldsmith S, From AHL. Arsenic-induced atypical ventricular tachycardia. N Engl J Med 1980; 303: 1096-7. Gollop BR, Glass WI. Urinary arsenic levels in timber treatment operators. NZ Med J 1979; 89: 10-1. Gonsebatt ME, Vega L, Montero R, Garcia-Vargas G, Del Razo LM, Albores A, Cebrian ME, Ostrosky-Wegman P. Lymphocyte replicating ability in individuals exposed to arsenic via drinking water. Mutat Res1994; 313: 293-9. Grande GA, Rogers AA, Ling LJ, Googins MK. Urine spot test as guide to treatment in acute pentavalent arsenic ingestion. Vet Hum Toxicol 1987; 29: 73-4. Grant WM, Schuman JS. Toxicology of the eye. 4th ed. Illinois: Charles C Thomas, 1993. Graziano JH. Role of 2,3-dimercaptosuccinic acid in the treatment of heavy metal poisoning. Med Toxicol 1986; 1: 155-62. Greenberg C, Davies S, McGowan T, Schorer A, Drage C. Acute respiratory failure following severe arsenic poisoning. Chest 1979; 76: 596-8. Hall JC, Harruff R. Fatal cardiac arrhythmia in a patient with interstitial myocarditis related to chronic arsenic poisoning. South Med J 1989; 82: 1557-60. Hannemann M, Kreppel H, Kraus M, Szinicz L, Kauth I, Reichl FX, Singh PK, Jones MM. BAL, DMPS, DMSA and new DMSA analogues as arsenic antidotes. Przeglad Lekarski 1995; 52: 175. Hantson Ph, Mahieu P, Lauwerys R. Tolerance of intravenous dimercaptosuccinic acid (DMSA) therapy for acute arsenic poisoning. Przeglad Lekarski 1995; 52: 175. Health and Safety Executive. EH40/95. Occupational exposure limits 1995. Sudbury: Heath and Safety Executive, 1995. Heaven R, Duncan M, Vukelja SJ. Arsenic intoxication presenting with macrocytosis and peripheral neuropathy, without anemia. Acta Haematol 1994; 92: 142-3. Heyman A, Pfeiffer JB, Willett RW, Taylor HM. Peripheral neuropathy caused by arsenical intoxication. A study of 41 cases with observations on the effects of BAL (2,3-dimercapto-propanol). N Engl J Med 1956; 254: 401-9. Hruby K, Donner A. 2,3-dimercapto-1-propanesulphonate in heavy metal poisoning. Med Toxicol 1987; 2: 317-23. HSDB/Hazardous Substances Data Bank. In: Tomes plus. Environmental Health and Safety Series I. Vol 26. National Library of Medicine, 1995. Hsueh Y-M, Cheng G-S, Wu M-M, Yu H-S, Kuo T-L, Chen C-J. Multiple risk factors associated with arsenic-induced skin cancer: effects of chronic liver disease and malnutritional status. Br J Cancer 1995; 71: 109-14. Hutton JT, Christians BL, Dippel RL. Arsenic poisoning [letter]. N Engl J Med 1982; 307: 1080. Inns RH, Rice P, Bright JE, Marrs TC. Evaluation of the efficacy of dimercaprol chelating agents for the treatment of systemic organic arsenic poisoning in rabbits. Hum Exp Toxicol 1990; 9: 215-20. IPCS. Environmental Health Criteria 18. Arsenic. Geneva: WHO, 1981. IPCS. Guidelines for drinking-water quality. 2nd ed. Vol 2. Health criteria and other supporting information. Geneva: WHO, 1996. Jenkins RB. Inorganic arsenic and the nervous system. Brain 1966; 89: 479-98. Jolliffe DM, Budd AJ, Gwilt DJ. Massive acute arsenic poisoning. Anaesthesia 1991; 46: 288-90. Jones MM. Chemistry of chelation: Chelating agent antagonists for toxic metals. Chapter 13. In: Goyer RA, Cherian MG, eds. Toxicology of metals: biochemical aspects. Handbook of Experimental Pharmacology Vol 115. Heidelberg: Springer Verlag 1995; 279-304. Kersjes MP, Maurer JR, Trestrail JH, McCoy DJ. An analysis of arsenic exposures referred to the Blodgett Regional Poison Center. Vet Hum Toxicol 1987; 29: 75-8. Kew J, Morris C, Aihie A, Fysh R, Jones S, Brooks D. Arsenic and mercury intoxication due to Indian ethnic remedies. Br Med J 1993; 306: 506-7. Kingston RL, Hall S, Sioris L. Clinical observations and medical outcome in 149 cases of arsenate ant killer ingestion. Clin Toxicol 1993; 31: 581-91. Kjeldsberg CR, Ward HP. Leukemia in arsenic poisoning. Ann Intern Med 1972; 77: 935-7. Kosnett MJ, Becker CE. Dimercaptosuccinic acid as a treatment for arsenic poisoning. Vet Hum Toxicol 1987; 29: 462. Kreppel H, Reichl FX, Forth W, Fichtl B. Lack of effectiveness of d-penicillamine in experimental arsenic poisoning. Vet Hum Toxicol 1989; 31: 1-5. Kreppel H, Reichl F-X, Szinicz L, Fichtl B, Forth W. Efficacy of various dithiol compounds in acute As2O3 poisoning in mice. Arch Toxicol 1990; 64: 387-92. Kreppel H, Paepcke U, Thiermann H, Szinicz L, Reichl FX, Singh PK, Jones MM. Therapeutic efficacy of new dimercaptosuccinic acid (DMSA) analogues in acute arsenic trioxide poisoning in mice. Arch Toxicol 1993; 67: 580-5. Kyle RA, Pease GL. Hematologic aspects of arsenic intoxication. N Engl J Med 1965; 273: 18-23. Lai M-S, Hsueh Y-M, Chen C-J, Shyu M-P, Chen S-Y, Kuo T-L, Wu M-M, Tai T-Y. Ingested inorganic arsenic and prevalence of diabetes mellitus. Am J Epidemiol 1994; 139: 484-92. Lantz RC, Parliman G, Chen GJ, Carter DE. Effect of arsenic exposure on alveolar macrophage function. I. Effect of soluble As(III) and As(V). Environ Res 1994; 67: 183-95. Lenz K, Hruby K, Druml W, Eder A, Gaszner A, Kleinberger G, Pichler M, Weiser M. 2,3-dimercaptosuccinic acid in human arsenic poisoning. Arch Toxicol 1981; 47: 241-3. Le Quesne PM, McCleod JG. Peripheral neuropathy following a single exposure to arsenic. J Neurol Sci 1977; 32: 437-51. Levin-Scherz JK, Patrick JD, Weber FH, Garabedian C Jr. Acute arsenic ingestion. Ann Emerg Med 1987; 16: 702-4. Lugo G, Cassady G, Palmisano P. Acute maternal arsenic intoxication with neonatal death. Am J Dis Child 1969; 117: 328-30. Maehashi H, Murata Y. Arsenic excretion after treatment of arsenic poisoning with DMSA or DMPS in mice. Jpn J Pharmacol 1986; 40: 188-190. Mahieu P, Buchet JP, Roels HA, Lauwerys R. The metabolism of arsenic in humans acutely intoxicated by As2O3. Its significance for the duration of BAL therapy. Clin Toxicol 1981; 18: 1067-75. Marcus SM. Survival after massive arsenic trioxide ingestion. Vet Hum Toxicol 1987; 29: 481. Martin DS, Willis SE, Cline DM. N-acetylcysteine in the treatment of human arsenic poisoning. J Am Board Fam Pract 1990; 2: 293-6. Mathieu D, Mathieu-Nolf M, Germain-Alonso M, Neviere R, Furon D, Wattel F. Massive arsenic poisoning - effect of hemodialysis and dimercaprol on arsenic kinetics. Intensive Care Med 1992; 18: 47-50. McCutchen JJ, Utterback RA. Chronic arsenic poisoning resembling muscular dystrophy. South Med J 1966; 59: 1139-45. McWilliams ME. Accidental acute poisoning by a concentrated solution of arsenic acid from percutaneous absorption: a case report. Vet Hum Toxicol 1989; 31: 354. Mees RA. The nails with arsenical polyneuritis. JAMA 1919; 72: 1337. Miller CS, Hoover WL, Culver WH. Exposure of pesticide applicators to arsenic acid. Arch Environ Contam Toxicol 1980; 9: 281-88 Mitchell-Heggs CAW, Conway M, Cassar J. Herbal medicine as a cause of combined lead and arsenic poisoning. Hum Exp Toxicol 1990; 9: 195-6. Moore DF, O'Callaghan CA, Berlyne G, Ogg CS, Alban Davies H, House IM, Henry JA. Acute arsenic poisoning: absence of polyneuropathy after treatment with 2,3-dimercaptopropanesulphonate (DMPS). J Neurol Neurosurg Psychiatry 1994; 57: 1133-5. Morton WE, Dunnette DA. Health effects of environmental arsenic. In: Nriagu JO, ed. Arsenic in the environment. Part II: Human health and ecosystem effects. Vol 27. New York: John Wiley & Sons, Inc., 1994; 17-34. Narang APS. Arsenicosis in India. Clin Toxicol 1987; 25: 287-95. Nicolis GD, Helwig EB. Exfoliative dermatitis. A clinicopathologic study of 135 cases. Arch Dermatol 1973; 108: 788-97. Peters RA, Stocken LA, Thompson RHS. British anti-lewisite (BAL). Nature 1945; 156: 616-9. Peters RA, Spray GH, Stocken LA, Collie CH, Grace MA, Wheatley GA. The use of British anti-lewisite containing radioactive sulphur for metabolism investigations. Biochem J 1947; 41: 370-3. Peterson RG, Rumack BH. D-penicillamine therapy of acute arsenic poisoning. J Pediatr 1977; 91: 661-6. Pomroy C, Charbonneau SM, McCullough RS, Tam GKH. Human retention studies with 74As. Toxicol Appl Pharmacol 1980; 53: 550-7. Reichl F-X, Hunder G, Liebl B, Fichtl B, Forth W. Effect of DMPS and various adsorbents on the arsenic excretion in guinea-pigs after injection with As2O3. Arch Toxicol 1995; 69: 712-7. Reinl VW. Lungenkrebs durch arsen-einwirkung bei einem industriearbeiter. [Pulmonary cancer caused by arsenic in an industrial worker.] Zentralbl Arbeitsmed Arbeitsschutz 1970; 20: 75-8. Reynolds JEF, ed. Martindale: The Extra Pharmacopoeia. London: The Pharmaceutical Press, 1993. Rezuke WN, Anderson C, Pastuszak WT, Conway SR, Firshein SI. Arsenic intoxication presenting as a myelodysplastic syndrome: a case report. Am J Hematol 1991; 36: 291-3. Roses OE, García Fernández JC, Villaamil EC, Camussa N, Minetti SA, Martínez de Marco M et al. Mass poisoning by sodium arsenite. Clin Toxicol 1991; 29: 209-13. RTECS/Registry of Toxic Effects of Chemical Substances. In: Tomes plus. Environmental Health and Safety Series I. Vol 26. National Institute for Occupational Safety and Health (NIOSH), 1995. Schoolmeester WL, White DR. Arsenic poisoning. South Med J 1980; 73: 198-208. Shannon RL, Strayer DS. Arsenic-induced skin toxicity. Hum Toxicol 1989; 8: 99-104. Sheabar FZ, Yannai S, Taitelman V. Efficiency of arsenic clearance from human blood in vitro and from dogs in vivo by extracorporeal complexing hemodialysis. Pharmacol Toxicol 1989; 64: 329-33. Shum S, Skarbovig J, Habersang R. Acute lethal arsenite poisoning in mice: effect of treatment with N-acetylcysteine, d-penicillamine and dimercaprol on survival time. Vet Hum Toxicol 1981; 23: 39-42. Smith SB, Wombolt DG, Venkatesan R. Results of hemodialysis and hemoperfusion in the treatment of acute arsenic ingestion. Clin Exp Dial Apheresis 1981; 5: 399-404. Stocken LA, Thompson RHS. British anti-lewisite. Arsenic and thiol excretion in animals after treatment of lewisite burns. Biochem J 1946; 40: 548-54. Tallis GA. Acute lead arsenate poisoning. Aust N Z J Med 1989; 19: 730-2. Telolahy P, Javelaud B, Cluet J, de Ceaurriz J, Boudene C. Urinary excretion of porphyrins by smelter workers chronically exposed to arsenic dust. Toxicol Lett 1993; 66: 89-95. Vahter M. What are the chemical forms of arsenic in urine, and what can they tell us about exposure? Clin Chem 1994; 40: 679-80. Vaziri ND, Upham T, Barton CH. Hemodialysis clearance of arsenic. Clin Toxicol 1980; 17: 451-6. Waldron HA, Scott A. Metals - Arsenic. In: Raffle PAB, Adams PH, Baxter PJ, Lee WR, eds. Hunter's disease of occupations. 8th ed. London: Edward Arnold, 1994; 109-12. Watson WA, Veltri JC, Metcalf TJ. Acute arsenic exposure treated with oral d-penicillamine. Vet Hum Toxicol 1981; 23: 164-6. Wexler J, Eagle H, Tatum HJ, Magnuson HJ, Watson EB. Clinical uses of 2,3-dimercaptopropanol (BAL). II. The effect of BAL on the excretion of arsenic in normal subjects and after minimal exposure to arsenical smoke. J Clin Invest 1946; 25: 467-73. WHO. Guidelines for drinking-water quality. 2nd Edition. Vol 1. Recommendations. Geneva: WHO, 1993. Yamamura Y, Yamauchi H. Arsenic metabolites in hair, blood and urine in workers exposed to arsenic trioxide. Ind Health 1980; 18: 203-10.