PEDIATRICS Vol. 108 No. 2 August 2001, pp. 473-476

EXPERIENCE AND REASON:
Methyl Ethyl Ketone Peroxide Ingestion: Toxicity and Outcome in a 6-Year-Old Child

	ABSTRACT

Top Abstract Introduction Discussion References

A 6-year-old boy developed respiratory distress, metabolic acidosis, severe esophageal and gastric burns, and a coagulopathy after ingestion of an unknown volume of methyl ethyl ketone peroxide (MEKP) in dimethyl phthalate. He was discharged from the pediatric intensive care unit 19 days postingestion but subsequently developed a stricture of the gastroesophageal junction and complete fibrosis of the middle third of the stomach, necessitating gastric resection and reconstruction. He was discharged 93 days postingestion on a program of dilation for the residual esophageal stricture. MEKP acts by initiating lipid peroxidation via free radical production that results in cellular dysfunction and death. Acetylcysteine, a glutathione precursor and possible free radical scavenger, may be of use in severe MEKP poisoning. This case demonstrates the severe effects that some industrial chemicals can have both systemically and locally at the point of contact with the gastrointestinal tract, as well as the long-term management required to ensure good quality of life.

Methyl ethyl ketone peroxide (MEKP; MEK peroxide, 2-butanone peroxide, C₈H₁₈O₄, CAS number 1338-23-4) is an organic peroxide. It is a colorless, odorless liquid and a strong oxidizing agent. It is used as a hardener and curing agent for plastics such as unsaturated polyester and fiberglass resins. Pure MEKP can explode from mechanical shock; consequently, it usually is available as a 40% to 60% solution in dimethyl phthalate (or other phthalates) and often is stored under refrigeration.

Ingestion of MEKP is rare, particularly in children, and reported infrequently. However, ingestion of MEKP carries a high morbidity and mortality. This article documents the clinical course of poisoning after ingestion of MEKP by a 6-year-old child. It illustrates the risks of decanting harmful chemicals into household containers and storing them in the home.

	CASE REPORT

A 6-year-old previously healthy boy ingested an unknown quantity of MEKP that had been stored in a lemonade bottle in the family garage. The liquid, brought home from work by the child's father, was a 36% solution of MEKP in 52% dimethyl phthalate (Butanox AM-50; Akzo Nobel Chemicals, Gillingham, Kent, UK). The father immediately rinsed out the child's mouth, gave him a drink of milk, manually induced vomiting, and took him immediately to the accident and emergency department of a local hospital. The child vomited again on the way to the hospital and on arrival. He was subsequently transferred to a pediatric hospital 7.5 hours postingestion.

On admission, the child had a heart rate of 180/min and a blood pressure of 120/80 mmHg. He appeared pale and agitated with peripheral cyanosis and required intravenous fluid resuscitation with 20 mL/kg of 4.5% albumin solution. Increasing stridor and hypoxemia led to urgent endotracheal intubation and transfer to the pediatric intensive care unit. Arterial blood gas estimation at this time showed a mixed metabolic and respiratory acidosis (pH: 7.03; PaCO₂: 9.2 kPa; PO₂: 35.6 kPa; base deficit: 13 mmol/L; bicarbonate: 21). After intubation, the child had metabolic acidemia (pH: 7.30; PaCO₂: 4.8 kPa; PO₂: 38.5 kPa; base deficit: 7 mmol/L; bicarbonate: 19) that had settled to a mild metabolic acidosis by day 3 (pH: 7.38; PaCO₂: 4.1 kPa; PO₂: 17.1 kPa; base deficit: 5 mmol/L; bicarbonate: 20).

The creatinine on admission was normal at 102 µmol/L, but the urea was elevated at 11.5 mmol/L. Urinalysis confirmed the presence of blood. Serum haptoglobin was normal, but the creatine kinase was elevated at 584 IU (normal: 24-195 IU). Both the serum urea and the creatinine rose and peaked on day 2. The liver function tests were mildly abnormal. They peaked on day 3 and returned to normal at approximately day 6. The bilirubin remained normal, as did the sodium and potassium concentrations. A coagulation screen showed increased prothrombin time, activated partial thromboplastin time, and decreased fibrinogen. The D-dimers were markedly abnormal at >1000 IU/L (normal: <500 IU/L).

Esophagoscopy and laryngoscopy at 14 hours postingestion showed superficial mucosal injury with edema of the oropharynx and larynx. The whole esophagus was coated in a white slough, but obvious aortic pulsation suggested residual suppleness of the esophageal wall. The gastroesophageal junction and the body of the stomach were most severely affected, with deep rigid sloughing and no evidence of peristaltic activity. In view of the increasing laryngeal and subglottal edema, a tracheostomy was performed.

Central venous access was established via the right internal jugular vein. Total parenteral nutrition was commenced on day 3. He was repeatedly pyrexial, but blood cultures failed to isolate an organism and blind triple antibiotic therapy (cefotaxime, metronidazole and ampicillin) was given. Hydralazine and nifedipine were required to control several episodes of hypertension of unclear cause. He developed bilateral pleural effusions and required right-sided drainage. An echocardiogram on day 7 and a CT scan on day 8, performed because of the possibility of raised intracranial pressure, were normal.

Repeat esophagoscopy on days 5 and 10 showed improvement in the mucosal appearance, but there was marked sloughing and bleeding at the gastroesophageal junction. The stomach was not entered on these occasions because of the risk of perforation. Sedation was stopped on day 13, and enteral feeding via a nasogastric tube commenced on day 15. The child was discharged from the pediatric intensive care unit on day 19.

Assessment by speech therapists on day 31 confirmed that the child's speech and swallowing mechanisms were intact. Enteral feeding became progressively more difficult, and esophagoscopy on day 34 revealed a tight stricture at the gastroesophageal junction such that the stomach could not be entered. A laparotomy on day 38 revealed complete occlusion of the body and the antrum of the stomach. This required resection and anastomosis of the residual fundus to the prepyloric area, salvaging a stomach of approximately one third the normal volume. The residual esophageal stricture was supple and dilated easily and was considered not to require surgery. Oral feeding was possible on day 80, and the child was discharged from the hospital on day 93 with a scheduled program of endoscopy and esophageal dilation.

The child presented again 1 week later with vomiting. Esophagoscopy confirmed an otherwise normal esophagus except for the strictured area at the cardioesophageal junction, which again dilated easily. The interval between dilations was lengthened gradually. The stricture zone just above the gastroesophageal junction is relatively supple and dilates easily. At 5 years postingestion, the child presents with swallowing difficulty once or twice a year and is treated with a balloon dilation under general anesthesia as a day case. After dilation, his swallowing is normal and his diet is unrestricted, but his intake is poor. He remains thin but is developing along normal centiles and takes iron supplements for a microcytic anemia. His vitamin B₁₂ concentration is within the normal range.

	DISCUSSION

Top Abstract Introduction Discussion References

MEKP is used for hardening fiberglass resins through the production of free radicals. It is postulated that the mechanism of toxicity of MEKP is free radical formation that leads to lipid peroxidation and results in corrosive damage to the gastrointestinal mucosa and liver damage. Lipid peroxidation is a free radical-initiated chain oxidation of unsaturated lipids. The lipid radicals are unstable and react with molecular oxygen to form organic peroxy-free radicals to yield 1 hydroperoxide and 1 new radical. When the substrate is depleted or in the presence of an antioxidant, the reaction is terminated. Subcellular membranes are rich in unsaturated fatty acids and therefore are a target of lipid peroxidation that results in cellular dysfunction and death. In addition, some products of lipid peroxidation, such as aldehydes, also may cause toxicity. Ethane and pentane are the decomposition products of 3-fatty acid hydroperoxides and 6-fatty acid hydroperoxides, respectively. These gases can be measured in expired air as an index of lipid peroxidation. MEKP is a potent initiator of lipid peroxidation and results in pentane production; it has been used as a model for lipid peroxidation in animals. Pentane production after MEKP exposure has been measured in experimental animals¹ but is impractical in the clinical situation (in addition, cell death from any cause can result in production of ethane and pentane).

Mild liver damage consisting of glycogen depletion, fatty changes in liver cells, and infiltration of cells in portal spaces has been reported in rats exposed to organic peroxides.² The liver toxicity from carbon tetrachloride, dibromoethane, halothane, and a number of other substances is known to be caused by free radicals produced during lipid peroxidation. Peripheral zonal hepatic necrosis was found at postmortem examination in a man who died 4 days after ingestion of MEKP with hepatic coma, coagulopathy, and respiratory insufficiency.³ Esophageal burns in this case were only superficial. Postmortem findings in another adult showed superficial necrosis of the hepatic capsule with extensive necrosis of the esophagus and stomach.⁴ However, this patient died within approximately 12 hours of ingestion and maximal liver damage probably did not evolve within this time.

Free radicals and lipid peroxidation also have been shown to have a role in corrosive damage. Malondialdehyde (an end product of lipid peroxidation) and glutathione (an antioxidant) were measured at 24, 48, and 72 hours after exposure in rat esophageal tissue treated with sodium hydroxide, a strong alkali. The malondialdehyde concentration was significantly higher in all treated animals compared with the controls. The glutathione concentration was significantly lower in the 48- and 72-hour samples compared with controls.⁵

The toxic dose of MEKP has not been established, but ingestion of any amount should be regarded as potentially serious. In cases in which the dose was known,^3,6,7 toxicity occurred after ingestion of 50 to 100 mL in adults. The median lethal dose in rats² is 484 mg/kg, and a dose of the same order of magnitude (522-597 mg/kg) was fatal in a 47-year-old man.³ MEKP was the most toxic organic peroxide administered by any route to rats.²

Dimethyl phthalate, which often is added to solutions of MEKP as a plasticizer to reduce the risk of explosion, commonly is used as an insect repellent. It generally is thought to be of relatively low toxicity⁸ and is considered to be an irritant rather than corrosive. The median lethal dose in animals varies from 2.4 g/kg (guinea pigs) to 7.2 g/kg (mice). Single doses of up to 1.4 g/kg in mice and dogs caused no effects,⁸ and a toxic dose for humans has not been established. In a review of cases of phthalate ingestion in humans,⁸ 3 adult cases were identified. A dose of 5 g of di-2-ethylhexyl phthalate caused no adverse effects, and 10 g caused mild diarrhea. Nausea, dizziness, and mild renal damage (albuminuria, red and white blood cells in the urine) were reported after ingestion of 10 g of dibutyl phthalate. Dimethyl phthalate may have contributed to the renal injury observed in our case, but equally and more likely, it may have enhanced absorption of MEKP by forming an organic phase immiscible with aqueous media.³

As stated, ingestion of MEKP, although rare, carries a high morbidity and mortality (Table 1). Of the 24 cases of ingestion (21 adults,^3,4,6,79-13 3 children^4,14) in the literature, including this case, 7 adults died (29%).^3,4,6,10,13 However, this is not a true mortality figure because of the bias in reporting. Clinical effects reported from ingestion of MEKP (Table 1) include vomiting, hematemesis, oral and esophageal burns, gastritis (which may be hemorrhagic), drowsiness, and coma. Metabolic acidosis, leukocytosis, respiratory distress, adult respiratory distress syndrome, aspiration pneumonitis, hypotension, hematuria, acute renal failure (secondary to rhabdomyolysis), and liver damage with coagulopathy may be observed in severe cases. Myocarditis with tachycardia, inverted T waves, congestive heart failure and gallop rhythm,⁹ and myocardial infarction¹⁰ have been reported. We did not observe any obvious cardiac toxicity in this child. Early gastrointestinal hemorrhage and perforation, in addition to delayed esophageal stricturing, may occur. Our case confirms the severe corrosive damage to both the esophagus and the stomach occurring particularly at the sites of prolonged contact. There also was evidence of systemic absorption, manifested by coagulopathy, renal and hepatic injury, and acidosis. The mechanism of acidosis is unclear. In this case, the systemic toxicity was relatively mild compared with the corrosive damage. This suggests that this child probably drank only a small quantity, resulting in significant local damage and limited systemic effects.

Gastric lavage and emesis are contraindicated after ingestion of a corrosive substance^15,16 because of the risk of trauma or further injury to the upper gastrointestinal tract on reexposure. This child had already vomited several times before he arrived in hospital, and 7.5 hours had elapsed since ingestion by the time he arrived in the second hospital. Oral fluids should be avoided after ingestion of MEKP because of the risk that perforation of the gastrointestinal tract may have occurred. All patients should be assessed urgently with early and regular endoscopic evaluation of the upper gastrointestinal tract. The ECG; arterial pH; and respiratory, renal, and liver function should be monitored. Nasogastric or jejunostomy tube feeding may be useful for providing adequate enteral intake until the final extent of the corrosive injury is clear. This may avoid the need for total parenteral nutrition and its associated complications. The enteral route of choice would be dictated by the site of the obstructing corrosive lesion. Once past the acute phase, patients will require ongoing follow-up because of the risk of late scarring and stricture formation. The known association between carcinoma and acid or alkali injury to the esophagus^17-20 or the stomach^21,22 makes prolonged review mandatory. Carcinoma has not been reported after ingestion of organic peroxides, but this probably is a reflection of the small number of cases reported. Also, the long period between ingestion and development of carcinoma means that the association between the 2 events may be missed. In addition, most cases of ingestion of MEKP involve adults rather than children, and death from other causes may occur before the development of neoplasm at the injury site.

Vitamin E and acetylcysteine have been suggested as possible therapies for MEKP toxicity¹⁴; however, there are no clinical data to support their use. In addition, the administration of antioxidants, such as vitamin E, after oxidative damage has started actually may promote damage rather than reduce it.²³ Acetylcysteine is a glutathione precursor and also may act as a free radical scavenger. It has been shown to be beneficial in carbon tetrachloride-induced hepatorenal failure.²⁴ The effect of acetylcysteine had been investigated in alkali-induced esophageal injury in animals. Compared with controls, stricture formation was less frequent and less severe in animals that were treated with either acetylcysteine or prednisolone. Acetylcysteine and prednisolone demonstrated similar efficacy in reducing the development of strictures in this animal model.²⁵ The use of acetylcysteine has not been reported in humans with acid- or alkali-induced corrosive injury.²⁶ Vitamin E has been shown in experimental animals to reduce lipid peroxidation (as measured by pentane concentrations in expired air) after exposure to MEKP.¹ In view of these studies and the good safety profile of acetylcysteine, its use should be considered in severe MEKP poisoning.

Our case illustrates the danger of bringing industrial chemicals into the home and highlights the risks of decanting and storing harmful chemicals in household containers, particularly drink containers. Decanting liquids into other containers also results in the loss of the safety information and details on the manufacturer and ingredients. This may lead to a delay in appropriate management after accidental ingestion. Although ingestion of MEKP is rare, most cases reported in the literature involve accidental ingestion from a drink container. This accident was not a typical case of poisoning in a child because this usually involves children who are 1 to 3 years of age and ingest substances while exploring their environment. This 6-year-old child intentionally drank from the lemonade bottle believing that it contained a drink. MEKP has been mistaken for rum,³ vodka,⁶ orange juice,⁷ beer,^11,12 and orangeade.⁶ Another child who developed esophageal stricture after accidental ingestion of MEKP also drank the solution from an unspecified drink container.¹⁴ This case demonstrates the severe corrosive effects that some industrial chemicals can have on the gastrointestinal tract, as well as the long-term management necessary to ensure good quality of life.

	ACKNOWLEDGMENTS

We thank Dr W.J. Tempowski for translating the Polish reference and Sarah McCrea for help in translating the French references. We also thank Dr Paul Dargan for his comments.

Nicola Bates, BSc(Brunel), BSc(open), MSc
National Poisons Information Service
Medical Toxicology Unit
London SE14 5ER United Kingdom

Christopher P. Driver, FRCS(Ed), FRCS(Paed)
Adrian Bianchi, MOM(Matta), MD, FRCS, FRCS(Ed)
Department of Paediatric Surgery
Royal Manchester Children's Hospital
Manchester M27 4HA United Kingdom

	FOOTNOTES

Reprints are not available from the authors.

Received for publication Sep 26, 2000; accepted Jan 2, 2001.

Address correspondence to Nicola Bates, National Poisons Information Service (London), Medical Toxicology Unit, Avonley Rd, London SE14 5ER, United Kingdom.

	ABBREVIATIONS

MEKP, methyl ethyl ketone peroxide.

	REFERENCES

Top Abstract Introduction Discussion References

1. Litov RE, Matthews LC, Tappel AL Vitamin E protection against in vivo lipid peroxidation initiated in rats by methyl ethyl ketone peroxide as monitored by pentane. Toxicol Appl Pharmacol 1981; 9:96-106
2. Floyd EP, Stokinger HE Toxicity studies of certain organic peroxides. Am Ind Assoc J 1958; 19:205-212
3. Karhunen PJ, Ojanpera I, Lalu K, Vuori E Peripheral zonal hepatic necrosis caused by accidental ingestion of methyl ethyl ketone peroxide. Hum Exp Toxicol 1990; 9:197-200 [Medline]
4. Mittleman RE, Romig LA, Gressman E Suicide by ingestion of methyl ethyl ketone peroxide. J Forensic Sci 1986; 31:312-320 [Medline]
5. Günel E, Çalayan F, Çalayan O, Akilliolu I Reactive oxygen radical levels in caustic esophageal burns. J Pediatr Surg 1999; 34:405-407 [Medline]
6. Wojdya Z, Pach J, Koodiej J Ostre zatrucia Ketonoxem. [Fatal acute poisoning with Ketonox]. Arch Med Sadowej Kryminol 1979; 29:199-205
7. Deisher JB Poisoning with liquid plastic catalyst. NW Med J 1958; 57:46
8. Krauskopf LG Studies on the toxicity of phthalates via ingestion. Environ Health Perspect 1973; 3:61-72 [Medline]
9. Dines DE, Shipman K Toxic myocarditis. Report of a case arising from ingestion of liquid plastic catalyst. Angiology 1962; 13:297-302
10. Litovitz TL, Schmitz BF, Bailey KM. Annual report of the American Association of the Poison Control Center National Data Collection System. Clin Toxicol. 1990;8:1989:394-442
11. Chataigner D, Garnier R, Reygagne A, Blasquez M, Efthymiou M-L Brûlures chimiques du tractus digestif supérieur et hépatite cytolytique secondaries à l'ingestion d'un mélange de peroxyde de méthyléthylcétone, de diacétone alcool et de phtalate de diiosobutyl. [Chemical burns to the upper digestive tract and cytolytic hepatitis secondary to ingestion of a mixture of methyl ethyl ketone peroxide, diacetone alcohol and diisobutyl phthalate]. J Toxicol Clin Exp 1992; 12:205-206 [Medline]
12. Blazquez M, Laurent-Puig P, Chataigner D, Delchier J-C Les lésions caustiques oesogastriques et hépatiques induites par l'ingestion d'un durcisseur de résine. [Caustic oesophogastric and liver lesions caused by ingestion of a plastic resin hardener]. Gastroenterol Clin Biol 1991; 15:154-555
13. Burger LM, Chandor SB Fatal ingestion of plastic resin catalyst. Arch Environ Health 1971; 3:402-404
14. Prez-Martnez A, Gutirrez-Junquera C, Gonzlvez-Piera J, Marco-Macin A, Rubio-Guijarro JR, Moya-Marchante M. Oesophageal stenosis in a child caused by ingestion of methyl ethyl ketone peroxide. Eur J Pediatr. 1997;156:976-977 (letter)
15. American Academy of Clinical Toxicology, European Association of Poisons Centres and Clinical Toxicologists Position statement: ipecac syrup. Clin Toxicol 1997; 35:699-709 [Medline]
16. American Academy of Clinical Toxicology, European Association of Poisons Centres and Clinical Toxicologists Position statement: gastric lavage. Clin Toxicol 1997; 35:711-719 [Medline]
17. Isolauri J, Markkula H Lye ingestion and carcinoma of the esophagus. Acta Chir Scand 1989; 59:269-271
18. Appelqvist P, Salmo M Lye corrosive carcinoma of the esophagus. A review of 63 cases. Cancer 1980; 45:2655-2658 [CrossRef][Medline]
19. Benirscke K Time bomb of lye ingestion? Am J Dis Child 1981; 135:17-18 [Medline]
20. Kinnman J, Shin HI, Wetteland P Carcinoma of the oesophagus after lye ingestion. Report of a case in a 15 year old Korean male. Acta Chir Scand 1968; 134:489-493 [Medline]
21. O'Donnell CH, Abbott WE, Hirshfeld JW Surgical treatment of corrosive gastritis. Am J Surg 1949; 78:251-255 [Medline]
22. Eaton H, Tennekoon GE Squamous carcinoma of the stomach following corrosive acid burns. Br J Surg 1972; 59:382-386 [Medline]
23. Halliwell B The antioxidant paradox. Lancet 2000; 355:1179-1180 [CrossRef][Medline]
24. Ruprah M, Mant TG, Flanagan RJ Acute carbon tetrachloride poisoning in 19 patients: implication for diagnosis and treatment. Lancet 1985; 1:1027-1029 [CrossRef][Medline]
25. Liu AJ, Richardson MA Effects of N-acetylcysteine on experimentally induced esophageal lye injury. Ann Otol Rhinol Laryngol 1985; 94:477-482 [Medline]
26. Chyka PA, Butler YA, Holliman BJ, Herman MI Utility of acetylcysteine in treating poisonings and adverse drug reactions. Drug Saf 2000; 22:123-148 [Medline]