UKPID MONOGRAPH ZINC CHLORIDE SM Bradberry BSc MB MRCP 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 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. ZINC CHLORIDE Toxbase summary Type of product Used in smoke bombs, soldering fluxes, disinfectants, fire-proofing agents, the textile industry and cements. Toxicity Zinc chloride is corrosive by ingestion and highly irritant by inhalation. Features Topical - Topical zinc chloride causes ulceration and burns and chronic exposure has been associated with anorexia, fatigue and weight loss. Ingestion - Zinc chloride is highly corrosive and ingestion of only 10 mL of a 35 per cent solution has caused oropharyngeal and gastric burns, epigastric tenderness, pharyngeal oedema, haematemesis and melaena (Chew et al, 1986). - Respiratory insufficiency and CNS depression may occur in severe cases. Recovery may be complicated by gastrointestinal stricture formation and/or pancreatic insufficiency. - Hyperglycaemia, hypokalaemia, increased alkaline phosphatase, amylase and liver transaminase activities have been reported. Inhalation - Zinc chloride inhalation from smoke screen generators or smoke bombs may cause transient cough, sore throat, hoarseness, a metallic taste and chest pain. - Exposure to high zinc chloride concentrations produces a chemical pneumonitis with marked dyspnoea, a productive cough, fever, chest pain and cyanosis. Pneumothorax and the adult respiratory distress syndrome (ARDS) have been reported. Fatalities have occurred (Hjortsœ et al, 1988). Management Dermal 1. Symptomatic and supportive measures as indicated by the severity of the burn. Ocular 1. Irrigate with copious lukewarm water or 0.9 per cent saline for at least ten minutes. 2. A topical anaesthetic may be required for pain relief and to overcome blepharospasm. 3. Ensure removal of any particles in conjunctival recesses. 4. Fluorescein staining allows identification of corneal burns. 5. Seek ophthalmological advice if symptoms persist or any significant abnormality is detected on examination. Ingestion 1. Most patients will vomit spontaneously. Gastric lavage is contraindicated following zinc chloride ingestion as this salt is corrosive. 2. Symptomatic and supportive measures with adequate fluid resuscitation are paramount. 3. Endoscopic examination may be required (ideally within the first 24 hours). 4. Save blood and urine for zinc concentration estimations. 5. Monitor the blood count and biochemical profile including serum amylase activity. 6. The value of chelation therapy following zinc ingestion has not been confirmed. Discuss with NPIS if patient is symptomatic. Inhalation 1. Remove from exposure after ensuring adequate self-protection. 2. Administer supplemental oxygen by face mask. 3. Symptomatic patients and those with abnormal respiratory physical signs should have a chest X-ray. 4. The possibility of delayed-onset chemical pneumonitis, pulmonary oedema and development of ARDS must be considered. 5. There is no established role for the routine use of inhaled or systemic corticosteriods in patients with pulmonary complications. 6. The use of antibiotics should be reserved for established infection only. 7. The value of chelation therapy following zinc chloride inhalation has not been confirmed. Discussion of individual cases with an NPIS physician is recommended. References Allen MB, Crisp A, Snook N, Page RL. 'Smoke-bomb' pneumonitis. Respir Med 1992; 86: 165-6. Chew LS, Lim HS, Wong CY, Htoo MM, Ong BH. Gastric stricture following zinc chloride ingestion. Singapore Med J 1986; 27: 163-6. Chobanian SJ. Accidental ingestion of liquid zinc chloride: local and systemic effects. Ann Emerg Med 1981; 10: 91-3. Hjortsœ E, Qvist J, Bud MI, Thomsen JL, Andersen JB, Wiberg-Jœgensen F, Jensen NK, Jones R, Reid LM, Zapol WM. ARDS after accidental inhalation of zinc chloride smoke. Intensive Care Med 1988; 14: 17-24. McKinney PE, Brent J, Kulig K. Acute zinc chloride ingestion in a child: local and systemic effects. Ann Emerg Med 1994; 23:1383-7. Potter JL. Acute zinc chloride ingestion in a young child. Ann Emerg Med 1981; 10: 267-9. Substance name Zinc chloride Origin of substance Formed during ignition of zinc oxide/hexachloroethane incendiary devices. (DOSE, 1994) Synonyms Butter of zinc Zinc dichloride Zinc muriate (DOSE, 1994) Chemical group A compound of zinc, a group II B (d block) transition element. Reference numbers CAS 7646-85-7 (DOSE, 1994) RTECS ZH1400000 (RTECS, 1997) UN 2331 (anhydrous) 1840 (solution) (DOSE, 1994) HAZCHEM CODE 2X (solution) (DOSE, 1994) Physicochemical properties Chemical structure ZnCl2 (DOSE, 1994) Molecular weight 136.28 (DOSE, 1994) Physical state at room temperature Solid (MERCK, 1996) Colour White (MERCK, 1996) Odour Fume has an acrid odour. (HSDB, 1997) Viscosity NIF pH The aqueous solution is acid to litmus; pH about 4. (MERCK, 1996) Solubility Water: 432 g/L at 25°C, 614 g/L at 100°C. Soluble in methanol, ethanol, diethyl ether and dimethyl ether. (MERCK, 1996; SAX'S, 1996) Autoignition temperature NIF Chemical interactions Zinc chloride forms an explosive reaction with copper (II) sulphide. Soluble zinc salts are precipitated as zinc hydroxide by ammonium hydroxide. A mixture of potassium and zinc chloride produces a strong explosion on impact. (SAX'S, 1996; HSDB, 1997) Major products of combustion Fumes of chloride and zinc oxide (SAX'S, 1996) Explosive limits NA Flammability Not flammable (HSDB, 1997) Boiling point 732°C (DOSE, 1994) Density 2.907 at 25°C (DOSE, 1994) Vapour pressure 133.322 Pa at 428°C (DOSE, 1994) Relative vapour density NIF Flash point NA Reactivity When heated to decomposition it emits toxic fumes of chloride and zinc oxide. (SAX'S, 1996) Uses In smoke bombs used for crowd dispersal, fire-fighting exercises and military screens. In dentifrices. Treatment of skin complaints. Deodorant. Flux in plating, soldering and welding. Pigment. In wood preservatives and fire-proofing agents. Astringent. (DOSE, 1994) Hazard/risk classification Index no. 030-003-00-2 Risk phrases C; R34. Causes burns. Safety phrases S(1/2) 7/8-28-45. Keep locked up and out of the reach of children. Keep container tightly closed and dry. After contact with skin, wash immediately with plenty of (to be specified by the manufacturer). In case of accident or if you feel unwell, seek medical advice immediately (show label where possible). EEC No: 231-592-0 (CHIP2, 1994) INTRODUCTION Zinc is an essential trace element required for the function of over 200 metallo-enzymes, including alkaline phosphatase and carbonic anhydrase (Sandstead, 1994). Zinc also plays a critical role in the regulation of DNA and RNA synthesis (via interaction with DNA binding proteins), in hormone-receptor interactions and in the "second-messenger" system of cellular signal transduction (Walsh et al, 1994). Zinc chloride is an acidic salt encountered most frequently in solder fluxes or as a constituent of 'smoke-bomb' emissions when zinc oxide and hexachloroethane are burned together. EPIDEMIOLOGY Inhalational exposure to low zinc chloride concentrations typically occur during military or civilian fire-drill exercises (Schenker et al, 1981; Allen, 1992). In an early report the accidental ignition of several smoke generators at a storage site led to severe zinc chloride poisoning affecting 70 persons, ten of whom died (Evans, 1945). There are several reported cases of accidental or deliberate self-harm involving zinc chloride ingestion, usually as soldering flux (Chobanian, 1981; Hedtke et al, 1989; McKinney et al, 1994). No deaths have occurred but debilitating sequelae include gastrointestinal stricture formation (secondary to corrosive damage) and chronic pancreatic insufficiency (McKinney et al, 1994). MECHANISM OF TOXICITY Excess zinc has been shown to reduce serum free thiol groups and disrupt hepatic enzyme activities (IPCS, 1996). Zinc chloride is directly corrosive to the mucous membranes of the respiratory and gastrointestinal tracts. Anaemia resulting from excess zinc uptake is usually due to zinc-induced copper deficiency via impaired gastrointestinal copper absorption. This probably involves zinc-induced increased metallothionein synthesis with increased copper-metallothionein retention in intestinal mucosal cells (Fischer et al, 1981). TOXICOKINETICS Absorption Zinc chloride exposure occurs primarily via inhalation and ingestion. Available zinc toxicokinetic data involve exposure to non-corrosive zinc salts. Gastrointestinal zinc absorption is a function of a cysteine-rich intestinal protein (CRIP) which sequesters zinc within enterocytes prior to active transport into plasma (Hempe and Cousins, 1992; Walsh et al, 1994). Metallothionein contributes to zinc homeostasis at higher exposures, primarily via retaining excess zinc within mucosal cells which are subsequently shed into the intestinal lumen (Hempe and Cousins, 1992). Zinc absorption is affected by diet; it is inhibited by calcium, phosphorus and phytates and facilitated by dietary protein (Hunt et al, 1991). In one study less than 15 per cent of dietary zinc was absorbed from a high phytate diet compared to 40 per cent from a diet with a high animal protein content (Sandstrom, 1995). In ten healthy volunteers Nève et al (1991) observed a peak serum zinc concentration some 2-3 hours after ingestion of 45 mg zinc (as zinc sulphate). Workers occupationally exposed to zinc fumes may have increased urine zinc concentrations (Hamdi, 1969) as evidence of systemic zinc uptake via the lungs. However, some inhaled zinc is undoubtedly swallowed (and absorbed via the gastrointestinal tract) following clearance via the mucociliary mechanism. Systemic zinc uptake following zinc chloride inhalation may also occur due to damage to the mucosal barrier. In two soldiers who died following zinc chloride inhalation, plasma zinc concentrations gradually increased during their illness although at autopsy zinc concentrations in major organs and tissues were normal save for a slightly increased lung zinc concentration in one case (Hjortsœ et al, 1988). Zinc is absorbed through broken skin when zinc oxide paste is used to treat wounds and burns (Hallmans, 1977). Distribution Most intravascular zinc is contained within erythrocytes. Plasma zinc is bound predominantly to albumin, the remainder bound to other proteins, particularly alpha2-macroglobulin, for distribution to tissues. Excess zinc is stored as a metallothionein complex, mainly in the liver (Abdel-Mageed and Oehme, 1990; IPCS, 1996). Some ninety per cent of total body zinc is in muscle and bone (Wastney et al, 1986). Appreciable amounts of zinc are found also in the kidney, lung, spleen and brain (IPCS, 1996). Zinc crosses the placenta slowly and is found in breast milk (Agency for Toxic Substances and Disease Registry, 1997). Excretion Most ingested zinc is eliminated in faeces via bile, pancreatic fluid and intestinal mucosal cells, with up to ten per cent appearing in urine (Abdel-Mageed and Oehme, 1990). Zinc is also eliminated in sweat and saliva. The kidneys do not play an important role in regulating total body zinc (IPCS, 1996). The whole-body zinc half-life is some 5-16 months (IPCS, 1996). CLINICAL FEATURES: ACUTE EXPOSURE Dermal exposure Topical zinc chloride exposure causes ulceration (Beliles, 1994) and burns (Evans, 1945; Chew et al, 1986). Contact dermatitis also has been described but without reference to original case data (Poisindex, 1996). Ocular exposure Zinc chloride is highly irritant to the eyes causing pain and erythema (Evans, 1945) which may be complicated by corneal oedema, burns and ulceration, iritis, glaucoma and cataract formation. Discrete grey spots on the lens ("glaukomflecken") as typically seen in acute glaucoma also have been described (Grant and Schuman, 1993). A plumber who was accidently splashed in the eye with a 30 per cent zinc chloride solution experienced an immediate reduction in visual acuity with conjunctival haemorrhage and inflammation, corneal opacification, a bullous keratopathy and "glaukomflecken" (Houle and Grant, 1973). Inhalation Pulmonary toxicity Exposure to zinc chloride fumes occurs primarily from smoke screen generators or smoke bombs when equal quantities of zinc oxide and hexachloroethane are burnt together. Following detonation of a smoke bomb participants in an airport disaster drill experienced cough, sore throat, hoarseness and chest pain in association with gastrointestinal upset (Schenker et al, 1981). Most symptoms resolved within 48 hours and none experienced permanent injury. In contrast, exposure to high zinc chloride concentrations produces a chemical pneumonitis with productive cough, dyspnoea, fever, chest pain, cyanosis and diffuse pulmonary infiltrates (Allen et al, 1992). In one study these features occurred following exposure to a zinc chloride concentration of approximately 4075 mg/m3 (1955 mg zinc/m3) (Johnson and Stonehill, 1961). The development of emphysematous bullae (which may be complicated by pneumothorax) has been reported (Matarese and Matthews, 1986). Severe cases may develop non-cardiogenic pulmonary oedema (the adult respiratory distress syndrome) (Hjortsœ et al, 1988) sometimes after several days of apparent clinical stability (Homma et al, 1992). Hjortsœ et al (1988) described two soldiers who died following exposure to fumes (for one and two minutes respectively) from smoke ammunition bombs (primarily zinc chloride). The first (exposed for two minutes) experienced immediate cough and dyspnoea but initial chest examination was unremarkable. Inhaled steroids (beclomethasone 100 µg qds) were prescribed. The following day he was pyrexial (38.4°C) with bilateral lower zone pulmonary infiltrates. Oral prednisolone 40 mg daily plus oral antibiotics were instituted. Sputum cultures were negative but the patient gradually deteriorated and some 15 days post exposure required mechanical ventilation for worsening hypoxia. Pulmonary angiography on the 17th day showed diffuse vascular obstruction. On day 19 the patient was commenced on intravenous and nebulized N-acetylcysteine plus intravenous L-3,4-dehydroproline 35 mg/kg/day in an attempt to inhibit pulmonary fibrosis. Despite these measures hypoxaemia and hypercapnia progressed and the patient died 32 days after exposure. A similar clinical course was observed in the second patient who died on the 25th day. Adverse effects occurred in 34 of 70 persons exposed in a tunnel to a mixture of zinc chloride (estimated concentration 33,000 mg zinc/m3), carbon and carbon dioxide following accidental ignition of several smoke bomb generators (Evans, 1945). Immediate symptoms included dyspnoea, hoarseness, chest pain and a productive cough. Two individuals died before reaching hospital; both showed severe respiratory distress and one had aspirated their stomach contents. Among the 15 patients admitted to hospital, all had evidence of upper respiratory tract (and eye) mucosal irritation, with haemoptysis, stridor, cyanosis and persisting chest pain in the majority, often accompanied by neurological, cardiovascular and gastrointestinal features (see below). Initially there were minimal pulmonary auscultatory findings but a chemical pneumonitis with fever and consolidation ensued in several patients during the next few days. Six of 15 patients died. Autopsy showed respiratory tract mucous membrane necrosis with pulmonary haemorrhage and oedema. Autopsy findings in other cases have included diffuse pulmonary microvascular obliteration with widespread pulmonary artery occlusion (Hjortsœ et al, 1988; Homma et al, 1992) and interstitial fibrosis (Milliken et al, 1963; Hjortsœ et al, 1988; Homma et al, 1992). Gastrointestinal toxicity A metallic taste, nausea, vomiting and epigastric pain have been reported following zinc chloride inhalation (Evans, 1945; Schenker et al, 1981). Gastrointestinal inflammation has been noted at autopsy of patients who died following severe zinc chloride poisoning (Evans, 1945). Cardiovascular toxicity Hypotension and tachycardia have accompanied the pulmonary features of significant zinc chloride inhalation (Evans, 1945). Neurotoxicity Agitation and restlessness have been described as early features accompanying pulmonary symptoms in those exposed to zinc chloride smoke (Evans, 1945). Ingestion Gastrointestinal toxicity Zinc chloride ingestion causes corrosive damage. Initial symptoms include burning of the mouth and pharynx with vomiting. An erosive pharyngitis, oesophagitis and gastritis may ensue which may be complicated by gastrointestinal haemorrhage and acute pancreatitis (Chobanian, 1981; Potter, 1981). In one case (Chobanian, 1981) oesophagitis, pharyngitis and acute pancreatitis occurred in a patient who ingested some 85 mL "Ruby Red" solder flux containing an unstated concentration of zinc chloride. An initial serum zinc concentration (some four hours post ingestion) was 1.46 mg/L (normal 0.5-0.9 mg/L). The patient recovered over five days. A 26 year-old woman who ingested 10 mL of a correction fluid containing zinc chloride (35 per cent) and methanol (0.5 per cent) presented four hours post ingestion with oropharyngeal and gastric burns, epigastric tenderness, dysphagia and diarrhoea. Examination revealed gross oral mucosal and pharyngeal oedema with surgical emphysema of the neck and upper chest wall (presumably due to pharyngeal perforation). Serum zinc on the third hospital day was normal (0.9 mg/L). The patient subsequently experienced haematemesis and melaena necessitating multiple blood transfusions. Recovery was delayed further by formation of a gastric stricture requiring subtotal gastrectomy (Chew et al, 1986). Hedtke et al (1989) reported a 13 month-old child who ingested 9300 mg of a moss killer containing 68 per cent zinc chloride. He presented with vomiting, gastric mucosal erosions, hyperglycaemia and increased alkaline phosphatase and amylase activities. The peak serum zinc concentration was 14.2 mg/L (normal range 0.68 - 1.36 mg/L) one hour post ingestion. A 16 month-old child who ingested approximately one tablespoon of soldering flux liquid (containing zinc chloride 22.5 per cent and ammonium chloride 5.5 per cent, pH 2) immediately vomited then developed dysphagia with inability to swallow his own saliva (McKinney et al, 1994). Severe gastrointestinal corrosive effects ensued (oropharyngeal, oesophageal and gastric burns). Severe gastric scarring and outlet obstruction necessitated a gastric antrectomy. The peak plasma zinc concentration was 12 mg/L (normal range 0.6-1.0 mg/L) 14 hours post ingestion (McKinney et al, 1994). Five months later pancreatic exocrine insufficiency was diagnosed with an on-going requirement for pancreatic enzyme supplements (McKinney et al, 1995). Pulmonary toxicity An asthmatic patient who ingested correction fluid developed an acute episode of bronchospasm and severe oropharyngeal and laryngeal inflammation leading to stridor and dysphonia (Chew et al, 1986). She eventually recovered. A child who immediately started coughing after ingesting a tablespoon of soldering flux (see above) was noted 20 minutes later to have widespread coarse breath sounds and wheeze in addition to features of severe gastrointestinal toxicity. A chest X-ray some five hours later revealed a right pleural effusion with basal atelectasis. The pulmonary features resolved with supportive care although the child had ongoing gastrointestinal problems (McKinney et al, 1994). Nephrotoxicity Microscopic haematuria without associated renal failure (Chobanian, 1981; Chew et al, 1986) and mild albuminuria (Chew et al, 1986) have occurred following zinc chloride ingestion. Neurotoxicity Within 20 minutes of ingesting one tablespoon of solder flux (containing zinc chloride 22.5 per cent and ammonium chloride 5.5 per cent, pH 2) a 16 month-old became comatose and later developed agitation and lethargy which persisted for several days (McKinney et al, 1994). Somnolence and lethargy occurred in a 13 month-old (Hedtke et al, 1989) and a two year-old (Potter, 1981) following zinc chloride ingestion. Cardiovascular toxicity Premature atrial beats were reported in a 13 month-old child following the supposed ingestion of 9.3 g moss killer containing 68 per cent zinc chloride (Hedtke et al, 1989). Hypotension secondary to intravascular volume depletion is likely in those with severe corrosive damage following zinc chloride ingestion. The 16 month-old child described above who ingested solder flux containing zinc chloride presented with clinical signs consistent with hypovolaemic shock (pulse 120 beats per minute, blood pressure 88/62 mmHg) but following intravenous fluid resuscitation and an upper gastrointestinal endoscopy, became hypertensive (figure not given) requiring hydralazine and labetalol therapy. The cause of hypertension in this case was not known but occurred prior to commencement of chelation therapy. Hepatotoxicity Severe gastrointestinal corrosive effects following zinc chloride ingestion have been associated with transiently increased liver enzyme activities (McKinney et al, 1994). Haemotoxicity Gastrointestinal ulceration and burns following zinc chloride ingestion may precipitate an acute fall in the haemoglobin concentration. Intravascular haemolysis may occur in severely poisoned patients (McKinney et al, 1994). CLINICAL FEATURES: CHRONIC EXPOSURE Dermal exposure In an early report chronic topical zinc chloride exposure in the pillow manufacturing industry was claimed to result in anorexia, weight loss, fatigue and leg pains, microcytic anaemia and thrombocytopenia (du Bray, 1937). The causative role of zinc chloride could not be confirmed but the patient's symptoms and abnormal investigation findings resolved when exposure ceased. Inhalation Occupational asthma associated with raised serum IgE concentrations occurred in two men after several months working with soldering fluxes containing zinc chloride and ammonium chloride (Weir et al, 1989) although in one patient there was no reduction in FEV1 after exposure to zinc chloride alone. MANAGEMENT Dermal exposure Decontaminate with soap and water ensuring adequate self-protection. Ideally, patients should remove their own contaminated clothing. Treat burns conventionally. Consider the possibility of systemic zinc uptake, particularly where the skin is broken. Ocular exposure Irrigate with copious amounts of lukewarm water for at least ten minutes. A topical anaesthetic may be required for pain relief and to overcome blepharospasm. Ensure removal of any particles lodged in the conjunctival recesses. The instillation of fluorescein allows detection of corneal damage. Specialist ophthalmological advice should be sought if any significant abnormality is detected on examination and in those whose symptoms do not resolve rapidly. Inhalation Remove the subject from exposure after ensuring adequate self-protection. Symptomatic and supportive measures are the mainstay of management. Symptomatic patients and those with abnormal respiratory signs should have a chest X-ray, receive supplemental oxygen and bronchodilators if necessary and be observed until symptoms resolve. The possibility of developing a chemical pneumonitis, delayed-onset pulmonary oedema and/or ARDS must be considered. In these circumstances high flow oxygen is mandatory but the value of inhaled or systemic corticosteriods has not been confirmed. Steroid therapy did not favourably influence the course of a soldier who died from ARDS 32 days after a two minute exposure to zinc chloride smoke (Hjortsœ et al, 1988). He received inhaled beclomethasone 100 µg qds within hours of exposure substituted on the second hospital day with a two week course of oral prednisolone 40 mg daily. In chemical pneumonitis and ARDS it generally is recommended that antibiotics should be reserved for established infection only. The patient described by Hjortsœ et al (1988) (see above) also received oral penicillin the day following zinc chloride exposure but sputum cultures were negative and the antibiotics were discontinued after seven days. Intravenous L-3,4-dehydroproline (35 mg/kg/day) was administered to a patient with zinc chloride smoke-induced ARDS in an attempt to inhibit pulmonary fibrosis but without modification of the progressively deteriorating course. The patient died 32 days after exposure. Ingestion Effective management primarily involves rapid appropriate symptomatic and supportive care. The role of chelating agents is discussed below. Decontamination and dilution Vomiting is likely to occur spontaneously following significant zinc chloride ingestion. Gastric lavage is contraindicated due to the corrosive nature of the salt. There may be some benefit in attempting oral dilution, if performed immediately, though this is controversial. Burkhart et al (1990) advocated whole-bowel irrigation as an effective gut decontamination method following zinc sulphate ingestion but there are no controlled data to support this view. Fluids should not be offered if the patient is not fully conscious, is unable to swallow or protect his/her own airway, has respiratory difficulty or severe abdominal pain. Possible complications of fluid administration include vomiting, aspiration, perforation of the gastrointestinal tract and worsening of oesophageal or gastric injuries. Supportive measures Airway support and analgesia should be provided as required. Treat hypovolaemic shock with intravenous colloid/crystalloid or blood. Monitor biochemical and haematological profiles and acid/base status. Antibiotics should be reserved for established infection only. If corrosive oesophageal or gastric damage is suspected panendoscopy should be carried out, ideally within 12-24 hours, to gauge the severity of injury (Knapp et al, 1994). A severity score based on acid ingestions may be useful: Grade 0: Normal examination 1: Oedema, hyperaemia of mucosa 2a: Superficial, localized ulcerations, friability, blisters 2b: Grade 2a findings and circumferential ulceration 3: Multiple, deep ulceration, areas of necrosis (Zargar et al, 1989) Zargar et al (1989) illustrated the important prognostic value of this grading system in the management of acid ingestions. In their series of 41 patients endoscoped within the first 36 hours of acid ingestion, those with grade 0 and 1 burns were discharged within one or two days, those with grade 2a burns required only supportive care for a little longer, whereas those with grade 2b and 3 burns required nutritional support via jejunostomy feeding (total parenteral nutrition is an alternative). All patients with grade 0, 1 and 2a injury recovered without sequelae. Acute complications and death were confined to those with grade 3 burns although several patients with grade 2b burns developed oesophageal or gastric strictures. An early surgical opinion should be sought if there is any suspicion of pending gastrointestinal perforation or where endoscopy reveals evidence of grade 3 burns. Corticosteroids There is no evidence to suggest any role for corticosteroid therapy in the management of zinc chloride ingestion. In a controlled trial of steroid use among 60 children with oesophageal burns following corrosive ingestion (alkalis in the majority) the use of steroids (intravenous prednisolone 2 mg/kg within 24 h and daily until oral intake was resumed then prednisolone 2.5 mg/kg orally daily for at least three weeks) did not influence outcome (Anderson et al, 1996). Smaller case series have also concluded that systemic corticosteroids confer no benefit following acid ingestion and may exacerbate or mask symptoms of pending perforation (Hawkins et al, 1980). Antidotes Animal studies Domingo et al (1988) investigated the antidotal potential of DTPA (trisodium calcium diethylene triamine-pentaacetate), CDTA (cyclohexane diamine tetraacetate), d-penicillamine, sodium calciumedetate, DMSA (dimercapto-succinic acid) and the sodium salt of DMPS (dimercaptopropanesulphonate) on reducing mortality in rodents poisoned with intraperitoneal zinc acetate 66-330 mg/kg. This dose range was selected to span the previously calculated intraperitoneal LD50 (108 mg/kg) and LD99 (216 mg/kg) for this zinc salt. Chelation therapy (or 0.9 per cent saline as control) was administered intraperitoneally to ten mice ten minutes after dosing with zinc. Antidote doses and outcome (measured as survival at 14 days) are summarized in Table 1. The very high antidote doses employed in this study necessitate caution in interpreting the results with regard to potential clinical value. Nevertheless at a zinc acetate dose in excess of the LD99 there was 100 per cent survival following treatment with DTPA, d-penicillamine and DMPS. DMSA was not an impressive antidote under these conditions. The same research group (Llobet et al, 1988) undertook a similar study. Intraperitoneal zinc acetate 0.49 mmol/kg (a dose approximately equivalent to its LD50) and 1.15 mmol/kg (approximately the LD99) was administered immediately before antidote administration to ten mice. Outcome was measured as survival ratio. Sodium calciumedetate (2152 mg/kg), DTPA (2262 mg/kg) and d-penicillamine (857 mg/kg), completely protected against mortality at the LD99 for zinc acetate. Under exactly the same conditions (each chelating agent at a dose of 5.75 mmol/kg) the survival ratios for CDTA, DMPS and DMSA were 80, 50 and 70 per cent respectively. Table 1. Survival (%) following parenteral chelation therapy in mice after a single zinc acetate injection (after Domingo et al, 1988) Chelating agent Antidote dose Zinc acetate dose (mg/kg) (mg/kg) 66 153 241 Control - 40 20 0 Na2CaCDTA 1360 100 90 90 Na2CaEDTA 1644 100 90 90 Na2CaDTPA 1569 100 100 100 d-Penicillamine 857 100 100 100 DMSA 619 80 30 10 DMPS 273 100 100 100 Several antidotes have been investigated for their potential to enhance zinc elimination. Domingo et al (1988) measured 24 hour faecal and urinary zinc excretion in mice administered 88 mg/kg intraperitoneal zinc acetate followed ten minutes later by an intraperitoneal dose of chelating agent (or saline in the control group). The same chelating agents and doses were used as in Table 1, again with ten mice in each treatment group. In the control group urine zinc elimination was less than half faecal excretion. CDTA enhanced urine zinc excretion some six fold and DTPA, DMSA and DMPS each some four fold. d-Penicillamine enhanced 48 hour urine zinc elimination by some 80 per cent. Sodium calciumedetate did not enhance urine zinc excretion but was the only antidote to increase 24 hour faecal zinc elimination (by some 20 per cent). The effect of increasing the time delay between zinc dosing and DTPA or CDTA administration was investigated by Llobet et al (1989). Intraperitoneal zinc acetate 0.4 mmol/kg (LD50 0.49 mmol/kg) was followed 0-24 hours later by a single intraperitoneal dose of chelating agent to give a chelating agent: zinc acetate molar ratio of approximately 10:1. Urine and faecal zinc elimination were monitored for 48 hours with five mice in each treatment group. Urine and faecal zinc elimination were increased significantly (p < 0.05) by both chelating agents when administered up to two hours after poisoning but only DTPA significantly enhanced the 48 hour urine and faecal zinc excretion when the antidotes were given 12 hours after zinc dosing. When there was a 24 hour delay between zinc acetate injection and antidote administration, DTPA significantly (p < 0.05) enhanced urine but not faecal 48 hour zinc elimination (CDTA was not effective). In summary, animal studies suggest DTPA is the most effective zinc antidote as judged by improved survival (100 per cent following a zinc acetate dose in excess of the LD99) and increased zinc elimination. CDTA produced 90 per cent survival and increased urine zinc excretion some six fold. d-Penicillamine also gave excellent results in mortality studies, with 100 per cent survival following administration of zinc acetate at a dose exceeding the LD99. However, d-penicillamine did not enhance zinc elimination significantly. Sodium calciumedetate improved survival (90-100 per cent) and in one study (Domingo et al, 1988) was the only chelating agent to increase faecal zinc elimination. DMPS 273 mg/kg completely protected mice against the lethal effects of zinc acetate at a dose in excess of the LD99 and the same antidote dose enhanced urine zinc excretion some four fold following administration of 88 mg/kg zinc acetate. By contrast DMSA 619 mg/kg resulted in only 10 per cent survival following 241 mg/kg zinc acetate. Clinical studies Although increased renal zinc excretion has been noted during chelation therapy instituted to enhance elimination of other toxic heavy metals, there is no convincing evidence of benefit in human zinc poisoning. In health, most zinc is eliminated via the gastrointestinal tract with only a small contribution made by renal excretion. No data have been found regarding the effect of chelation therapy on biliary zinc excretion in man. Dimercaprol McKinney et al (1994) reported improved mental status in a patient with severe zinc chloride poisoning (by ingestion) following administration of intramuscular dimercaprol 12 mg/kg/day for 24 hours and intravenous sodium calciumedetate 1 g/m2 for five days. This treatment was instituted 74 hours post ingestion. Chelation therapy was not associated with increased urine zinc elimination. Intramuscular dimercaprol 4 mg/kg qds was instituted less than four hours after the ingestion of 3 g each of zinc sulphate and copper sulphate by an 86 year-old woman. This patient also received oral d-penicillamine (see below) but both antidotes were discontinued after 48 hours due to deteriorating renal function. The patient made a full recovery over 20 days. A 16 year-old who ingested 12 g elemental zinc was treated some nine days later with intramuscular dimercaprol 2.3-9.2 mg/kg daily. Chelation was associated with clinical improvement and a reduction in the blood zinc concentration but urine zinc concentrations were not measured (Murphy, 1970). Sodium calciumedetate A 24 year-old man who developed erosive pharyngitis and oesophagitis, hyperamylasaemia, microscopic haematuria and a serum zinc concentration of 1.46 mg/L (normal range 0.5-0.9 mg/L) after ingesting liquid zinc chloride, made an uneventful recovery following supportive care and intravenous sodium calciumedetate 45 mg/kg in divided doses over 36 hours. No zinc excretion data were given (Chobanian, 1981). Potter (1981) utilized intravenous sodium calciumedetate 150 mg in the management of a 28 month-old child who had ingested a zinc chloride solution; no urine zinc excretion data were given. The patient reported by McKinney et al (1994) who was severely poisoned after ingesting one tablespoon of a zinc chloride-containing soldering flux was treated with intravenous sodium calciumedetate 1 g/m2 for five days. The urine zinc excretion in the eight hours preceding chelation was 950 µg. Urine zinc excretion was not increased by sodium calciumedetate with only 1000 µg/24 h removed on the fourth day of treatment (no interim data were given). N-acetylcysteine In response to a rising serum zinc concentration, a soldier who developed adult respiratory distress syndrome following two minutes inhalation of zinc chloride smoke was administered intravenous (140 mg/kg/day for three days) and nebulized (100 mg qds for 13 days) N-acetylcysteine between days 19 and 32 in an attempt to enhance zinc elimination. The urine zinc excretion increased from some 125 µmol/24h on day 20 to 260 µmol/24h on day 22 (coinciding with intravenous N-acetylcysteine administration) then rose to nearly 300 µmol/day (the maximum observed zinc excretion) on day 24. Unfortunately no pre-chelation zinc excretion measurements were made. There was no clinical improvement with therapy and the patient died in respiratory and renal failure on day 32. d-Penicillamine An 86 year-old woman who developed chemical pneumonitis, gastritis, cardiac and renal failure following the ingestion (and partial aspiration) of 3 g each of zinc and copper sulphate, received 250 mg oral d-penicillamine qds (in addition to intramuscular dimercaprol 4 mg/kg qds) commenced less than four hours after zinc ingestion (Hantson et al, 1996). Unfortunately there were no pre-chelation urine zinc excretion data and treatment was discontinued after 48 hours due to deteriorating renal function. The maximum 24 hour urine zinc excretion, achieved on the first day of chelation therapy was some 6000 µg. Another patient with zinc chloride poisoning by inhalation survived following treatment with oral penicillamine 125 mg twice daily (Allen et al, 1992). No blood or urine zinc concentrations were measured. Antidotes: Conclusions and recommendations 1. There are no controlled clinical data of chelation therapy in zinc poisoning and animal studies must be interpreted with caution in view of the extremely high antidote doses employed. Nevertheless, animal studies suggest that of the antidotes readily available for clinical use, sodium calciumedetate is the preferred agent with d-penicillamine or DMPS potential alternatives. 2. Although case reports claim clinical benefit following parenteral administration of dimercaprol, sodium calciumedetate and d- penicillamine, urine and/or faecal zinc excretion data to support these claims are lacking. 3. Chelation therapy cannot be advocated routinely in the management of zinc poisoning; symptomatic cases should be discussed with the NPIS. AT RISK GROUPS Patients with haemochromatosis are at greater risk of zinc toxicity due to iron-induced increased metallothionein concentrations since metallothionein has a greater affinity for zinc than iron. MEDICAL SURVEILLANCE Serum zinc concentrations are increased in acute zinc poisoning. The 24 hour urine zinc excretion is useful when monitoring chronic exposure although there is no well established relationship between the extent of exposure and urine zinc concentration. Hair zinc concentrations are not useful (Agency for Toxic Substances and Disease Registry, 1997). Normal zinc concentrations in biological fluids Plasma and serum: 1.1-1.3 mg/L (IPCS, 1996). Whole blood: 6.8-10.8 mg/L (IPCS, 1996). 24 hour urine excretion: less than 500 µg (IPCS, 1996). OCCUPATIONAL DATA Occupational exposure standard Zinc chloride, fume: Long-term exposure limit (8 hour TWA reference period) 1 mg/m3. OTHER TOXICOLOGICAL DATA Carcinogenicity There is no evidence to suggest zinc chloride is carcinogenic in humans (Leonard and Gerber, 1989) and the Environmental Protection Agency has concluded zinc is not classifiable with regard to its human carcinogenicity (Agency for Toxic Substances and Disease Registry, 1997). Marrs et al (1988) demonstrated a high incidence of pulmonary carcinoma in experimental zinc oxide/hexachloroethane smoke poisoning. However, a recent in vitro and in vivo study failed to show a significant genotoxic effect and the authors concluded the smoke was "not .... a major health hazard" (Anderson et al, 1996). Severe gastric burns following corrosive ingestion are associated with an increased risk of gastric carcinoma. Reprotoxicity There is no conclusive evidence regarding the reprotoxicity of zinc chloride in humans (Reprotext, 1996). There were no adverse effects following the administration of oral zinc sulphide (providing 20 mg elemental zinc daily) to 494 women during the last two trimesters of pregnancy (Mahomed et al, 1989). A relationship between high amniotic fluid or maternal serum zinc concentrations and foetal neural tube defects has been proposed, but evidence for this is inconsistent (Reprotext, 1996; Reprotox 1996). Pre-eclampsia, abnormal deliveries, anencephaly, and an increased incidence of stillbirths have been associated with low maternal serum zinc concentrations. Zinc deficiency also has been associated with delayed sexual maturity (Reprotext, 1996; Reprotox, 1996). Low seminal fluid zinc concentrations have been implicated in male infertility but the use of zinc supplements to treat this condition remains controversial (Reprotext, 1996; Reprotox, 1996). Genotoxicity Escherichia coli WP2 (lambda) microscreen assay-positive. Salmonella typhimurium TA98, TA100, TA102, TA1537, TA2637-negative. In vitro human lymphocytes, chromosomal aberrations-positive. In vitro human lymphocytes, low concentrations stimulated DNA synthesis, high concentrations inhibited DNA synthesis. In vivo mouse bone marrow cells, chromosomal aberrations were induced in calcium deficient mice but not in normal calcium supplemented mice (DOSE, 1994). Fish toxicity Lethal to fathead minnow at less than 80 mg/L as zinc (exposure unspecified) (DOSE, 1994). EC Directive on Drinking Water Quality 80/778/EEC Zinc: Guide level 100 µg/L at supply works, 5000 µg/L after 12 hour contact with consumers' pipework. Chlorides: Guide level 25 mg/L (DOSE, 1994). WHO Guidelines for Drinking Water Quality NIF AUTHORS SM Bradberry BSc MB MRCP 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 28/1/98 REFERENCES Abdel-Mageed AB, Oehme FW. A review of the biochemical roles, toxicity and interactions of zinc, copper and iron: I. Zinc. Vet Hum Toxicol 1990; 32: 34-9. Agency for Toxic Substances and Disease Registry (ATSDR). Zinc. In: ATSDR Toxicological Profiles. Atlanta: CRC Press, Inc., 1997. Allen MB, Crisp A, Snook N, Page RL. 'Smoke-bomb' pneumonitis. Respir Med 1992; 86: 165-6. Anderson D, Blowers SD, Marrs TC, Rice P. An in vitro and an in vivo unscheduled DNA synthesis assay with a zinc oxide/hexachloroethane (Zn/HCE) smoke. Hum Exp Toxicol 1996; 15: 38-44. Beliles RP. The metals: Zinc. In: Clayton GD, Clayton FE eds. Patty's industrial hygiene and toxicology. Vol 2. 4th ed. New York: John Wiley & Sons, Inc., 1994; 2332-42. Burkhart KK, Kulig KW, Rumack B. Whole-bowel irrigation as treatment for zinc sulfate overdose. Ann Emerg Med 1990; 19: 1167-70. Chew LS, Lim HS, Wong CY, Htoo MM, Ong BH. Gastric stricture following zinc chloride ingestion. Singapore Med J 1986; 27: 163-6. CHIP2/Chemicals (Hazard Information and Packaging for Supply) Regulations 1994. Health and Safety Commission. Sudbury: Health and Safety Executive, 1994. Chobanian SJ. Accidental ingestion of liquid zinc chloride: local and systemic effects. Ann Emerg Med 1981; 10: 91-3. Domingo JL, Llobet JM, Paternain JL, Corbella J. Acute zinc intoxication: comparison of the antidotal efficacy of several chelating agents. Vet Hum Toxicol 1988; 30: 224-8. DOSE/Dictionary of substances and their effects. Vol 7. Cambridge: Royal Society of Chemistry, 1994. du Bray ES. Chronic zinc intoxication. An instance of chronic zinc poisoning from zinc chloride used in the pillow manufacturing industry. JAMA 1937; 108: 383-5. Evans EH. Casualties following exposure to zinc chloride smoke. Lancet 1945; 2: 368-70. Fischer PWF, Giroux A, L'Abbé MR. The effect of dietary zinc on intestinal copper absorption. Am J Clin Nutr 1981; 34: 1670-5. Grant WM, Schuman JS. Toxicology of the eye. 4th ed. Illinois: Charles C Thomas, 1993. Hallmans G. Treatment of burns with zinc-tape. Scand J Plast Reconstr Surg Hand Surg 1977: 11: 155-61. Hamdi EA. Chronic exposure to zinc of furnace operators in a brass foundry. Br J Ind Med 1969; 26: 126-34. Hawkins DB, Demeter MJ, Barnett TE. Caustic ingestion: Controversies in management. A review of 214 cases. Laryngoscope 1980; 90: 98-109. Hedtke J, Daya MR, Neace G, Burton BT. Local and systemic toxicity following zinc chloride (ZnCl2) ingestion. Vet Hum Toxicol 1989; 31: 342. Hempe JM, Cousins RJ. Cysteine-rich intestinal protein and intestinal metallothionein: An inverse relationship as a conceptual model for zinc absorption in rats. J Nutr 1992; 122: 89-95. Hjortsœ E, Qvist J, Bud MI, Thomsen JL, Andersen JB, Wiberg-Jœgensen F, Jensen NK, Jones R, Reid LM, Zapol WM. ARDS after accidental inhalation of zinc chloride smoke. Intensive Care Med 1988; 14: 17-24. Homma S, Jones R, Qvist J, Zapol WM, Reid L. Pulmonary vascular lesions in the adult respiratory distress syndrome caused by inhalation of zinc chloride smoke: a morphometric study. Hum Pathol 1992; 23: 45-50. Houle RE, Grant WM. Zinc chloride keratopathy and cataracts. Am J Opthalmol 1973; 75: 992-6. HSDB/Hazardous Substances Data Bank. In: Tomes plus. Environmental Health and Safety Series 1. Vol 32. National Library of Medicine, 1997. Hunt JR, Lykken GI, Mullen LK. Moderate and high amounts of protein from casein enhance human absorption of zinc from whole wheat or white rolls. Nutr Res 1991; 11: 413-8. IPCS/International Programme on Chemical Safety. IPCS Chemical information CD-Rom (Inchem). Geneva: IPCS/Canadian Centre for Occupational Health and Safety (CCOHS), 1996. Johnson FA, Stonehill RB. Chemical pneumonitis from inhalation of zinc chloride. Dis Chest 1961; 40: 619-23. Knapp JF, Kennedy C, Wasserman GS, Lelli J. Case 01-1994: A toddler with caustic ingestion. Pediatric Emerg Care 1994; 10: 54-58. Léonard A, Gerber GB. Zinc toxicity: does it exist? J Am Coll Toxicol 1989; 8: 1285-1290. Llobet JM, Domingo JL, Corbella J. Antidotes for zinc intoxication in mice. Arch Toxicol 1988; 61: 321-3. Llobet JM, Domingo JL, Corbella J. Comparison of the antidotal efficacy of polyamincarboxylic acids (CDTA and DTPA) with time after acute zinc poisoning. Vet Hum Toxicol 1989; 31: 25-8. Mahomed K, James DK, Golding J, McCabe R. Zinc supplementation during pregnancy: a double blind randomised controlled trial. Br Med J 1989; 299: 826-30. Marrs TC, Colgrave HF, Edginton JAG, Brown RFR, Cross NL. The repeated dose toxicity of a zinc oxide/hexachloroethane smoke. Arch Toxicol 1988; 62: 123-32. Matarese SL, Matthews JI. Zinc chloride (smoke bomb) inhalational lung injury. Chest 1986; 89: 308-9. McKinney PE, Brent J, Kulig K. Acute zinc chloride ingestion in a child: local and systemic effects. Ann Emerg Med 1994; 23:1383-7. McKinney PE, Brent J, Kulig K. Zinc chloride ingestion in a child: Exocrine pancreatic insufficiency. Ann Emerg Med 1995; 25: 562. MERCK/The Merck Index. Zinc chloride. In: Budavari S. ed. An encyclopedia of chemicals, drugs, and biologicals. 12th ed. New Jersey: Merck and Co., Inc., 1996; 1733. Milliken JA, Waugh D, Kadish ME. Acute interstitial pulmonary fibrosis caused by a smoke bomb. Can Med Assoc J 1963; 88: 36-9. Murphy JV. Intoxication following ingestion of elemental zinc. JAMA 1970; 212: 2119-20. Nève J, Hanocq M, Peretz A, Abi Khalil F, Pelen F, Famaey JP. Pharmacokinetic study of orally administered zinc in humans. Evidence for an enteral recirculation. Eur J Drug Metab Pharmacokinet 1991; 16: 315-23. Poisindex. In: Micromedex International Healthcare Series. Vol 90. Colorado: Micromedex, Inc., 1996. Potter JL. Acute zinc chloride ingestion in a young child. Ann Emerg Med 1981; 10: 267-9. Reprotext. In: Tomes plus. Environmental Health and Safety Series I. Vol 31. Colorado: Micromedex, Inc., 1996. Reprotox. In: Tomes plus. Environmental Health and Safety Series I. Vol 31. Washington DC: Fabro S, Scialli AR. Reproductive Toxicology Center, Columbia Hospital for Women, 1996. RTECS/Registry of Toxic Effects of Chemical Substances. In: Tomes plus. Environmental Health and Safety Series 1. Vol 32. National Institute for Occupational Safety and Health (NIOSH), 1997. Sandstead HH. Understanding zinc: Recent observations and interpretations. J Lab Clin Med 1994; 124: 322-7. Sandstrom B. Consideration in estimates of requirements and critical intake of zinc. Adaption, availability and interactions. Analyst 1995; 120: 913-5. SAX'S/Lewis RJ. Sax's dangerous properties of industrial materials. 9th ed. Vol 3. New York: Van Nostrand Reinhold, 1996. Schenker MB, Speizer FE, Taylor JO. Acute upper respiratory symptoms resulting from exposure to zinc chloride aerosol. Environ Res 1981; 25: 317-24. Walsh CT, Sandstead HH, Prasad AS, Newberne PM, Fraker PJ. Zinc: health effects and research priorities for the 1990s. Environ Health Perspect 1994; 102: 5-46. Wastney ME, Aamodt RL, Rumble WF, Henkin RI. Kinetic analysis of zinc metabolism and its regulation in normal humans. Am J Physiol 1986; 251: 398-408. Weir DC, Robertson AS, Jones S, Sherwood Burge P. Occupational asthma due to soft corrosive soldering fluxes containing zinc chloride and ammonium chloride. Thorax 1989; 44: 220-3. Zargar SA, Kochhar R, Nagi B, Mehta S, Mehta SK. Ingestion of corrosive acids. Spectrum of injury to upper gastrointestinal tract and natural history. Gastroenterology 1989; 97: 702-7.