Macroscopic examination during the autopsy revealed a very important asphyxia syndrome with major visceral congestion. Blood, urine, liver, kidney, adrenal, ...
Journal of Analytical Toxicology, Vol. 24, March 2000
Fatal Aluminum PhosphidePoisoning Fran~oise Anger 1, FranCois Paysant2, Florent Brousse3, Isabelle te Normand 4, Patrick Develay 2, Yvan Galliard 3, Alain Baerts, Marie Annick Le Gueut 2, Gilbert Pepin3, and Jean-Pierre Anger 1,* 7Laboratoire de Toxicologie Pharmaceutique and 21nstitut de M6decine L~gale, U.F.R. des Sciences M~dicales et Pharmaceutiques, 2 Avenue du Professeur L6on Bernard, 35043 Rennes, France; 3Laboratoire d'Expertises TOXLAB, 18 Rue Andr~ del Sarte, 75018 Paris, France; and 4Laboratoire des Services de r~animation et des Urgences and 5Centre Anti Poison Centre Hospitalier R~gional et Universitaire de Pontchaillou, 35033 Rennes France
A 39-year-old man committed suicide by ingestion of aluminum phosphide, a potent mole pesticide, which was available at the victim's workplace. The judicial authority ordered an autopsy, which ruled out any other cause of death. The victim was discovered 10 days after the ingestion of the pesticide. When aluminum phosphide comes into contact with humidity, it releases large quantities of hydrogen phosphine (PH3), a very toxic gas. Macroscopic examination during the autopsy revealed a very important asphyxia syndrome with major visceral congestion. Blood, urine, liver, kidney, adrenal, and heart samples were analyzed. Phosphine gas was absent in the blood and urine but present in the brain (94 mL/g), the liver (24 mL/g), and the kidneys (41 mL/g). High levels of phosphorus were found in the blood (76.3 mg/L) and liver (8.22 mg/g). Aluminum concentrations were very high in the blood (1.54 mg/L), brain (36 pg/g), and liver (75 pg/g) compared to the usual published values. Microscopic examination revealed congestion of all the organs studied and obvious asphyxia lesions in the pulmonary parenchyma. All these results confirmed a diagnosis of poisoning by aluminum phosphide. This report points out that this type of poisoning is rare and that hydrogen phosphine is very toxic. The phosphorus and aluminum concentrations observed and their distribution in the different viscera are discussed in relation to data in the literature.
or moisture to form highly toxic phosphine gas (PH3), the fumigating agent. The rate of gas generation is determined by ambient temperature and humidity: AlP + 3 H20~PH3 + AI(OH)3. In France, aluminum phosphide is used for crop protection against various insects and sometimes to fight hamsters by planting tablets in their burrows. Because of its toxicity to humans and domestic animals, the use of AlP is strictly controlled, and it cannot be supplied to ordinary persons. The case of a 39-year-old man who voluntarily poisoned himself and was found dead in his car in June 1998 is reported here. This man worked as an agronomy engineer in a center for vegetable protection. He was depressed and had deserted his home eight days before he was found. His car was parked in a little countryside lane. On the windows of the vehicle, a message for those who would discover his body was displayed: "Be careful, this car has been poisoned." Inside the car, the body was in a state of advanced decay and released an odor of garlic. Some noningested pellets and others that had been vomited were found on the rear seat. The product used was Phosphinon | tablets that the victim managed to obtain at his workplace. The tablets contained 57% aluminum phosphide and are marketed as a very effective mole pesticide.
Introduction
Materials and Methods
Aluminum phosphide (ALP)is mainly used as a fumigant in pest control for grain and food preservation. Commercial formulations, which usually contain 55 to 75% active ingredient, are sold in the form of tablets. Aluminum phosphide is available without restriction in some countries. In India, for example, AlP poisoning, which was almost nonexistent a decade ago, has now reached epidemic proportions, and many reports of high mortality (> 50%) have recently been published (1). Pellets of solid aluminum phosphide react rapidly with water
The autopsy was carried out 48 h after the discovery of the body. Powerful aspiration and ventilation systems were installed at the site of the autopsy during the operation. Forensic physicians performing the autopsy wore masks with a specific cartridge for the individuals' protection. Ambient atmosphere was controlled with a Draeger apparatus at each phase of the autopsy: when the stretcher bag was opened, during the external examination, and when the thoracic and abdominal cavities were opened. High surrounding concentrations caused saturation of the cartridge. As a concentration peak was expected, the stomach was not opened. Jayaraman (2)
Abstract
* Author to whom correspondence should be addressed
90
Reproduction(photocopying)of editorial contentof this journal is prohibitedwithout publisher'spermission.
Journal of Analytical Toxicology, Voh 24, March 2000
reported that an autopsy surgeon required medical treatment following exposure to the phosphine released from the excised stomach of a patient who had died from aluminum phosphide poisoning. Systemic samples from the heart, liver, kidney, adrenals, brain, and both lungs were taken for histologic examination. These samples were preserved in diluted formalin (10%). During the autopsy, samples of blood, urine, liver, kidney, adrenals, brain, and heart were taken and immediately placed in small vials (10 mL) that were then sealed for the determination of phosphine concentration by headspace gas chromatography (GC) after incubation at 50~ in a thermostatic bath. Then the vials were unsealed, and the samples, after acid digestion, were analyzed for their phosphorus and aluminum contents by emission spectrometry. Systematic toxicological analysis of the biological samples involved the followingsubstances: ethanol and volatile substances were determined by headspace GC on a CP Sil 19 CB capillary column (25 m x 0.53 mm, Chrompack| and detected by flame ionization; carbon monoxide and cyanide were analyzed by microdiffusion using a Conway cell. Nonvolatile organic compounds (pharmaceuticals,alkaloids,pesticides)were analyzed by liquid-liquid extraction and high-performance liquid chromatographic (HPLC)method on a NovapackWaters C18 column (300 x 3.9-ram i.d., 4 IJm) heated at 30~ with a mobile phase of methanol/tetrahydrofuran/10-2MpH 2.6 KH2PO4(65:5:30v/v/v) and detection by Perkin Elmer UV diode-array spectrophotometer 235~ (HPLC-DAD) (3) or by GC coupled with a mass spectrometer (GC8000 MD800, Thermoquest, les Ulis, France) on a capillary column DB 5 MS (30 m x 0.25-mm i.d.) with a programmed temperature from 60~ (1 min) to 295~ (15 min) at 10~ (injector and transfer line at 280~ Phosphine gas was determined by headspace GC on a Hewlett-Packard 6890 with a capillary column Poraplot Q (30 m x 0.32-ram i.d., 10-1Jmfilm thickness) with a gradient temperature ranging from 45 to 240~ (20~ equipped with an HP 5972 mass selective detector. Pulse splitless injection was done at 180~ over 0.75 rain under the pressure of 65 kPa. Selected ion monitoring detection was done atm/z 31, 33, and 34 with a dwell time of 100 ms for each. Phosphorus and aluminum contents were quantitatively measured by inductively coupled plasma mass spectrometry (ICP-MS, Perkin Elmer Elan 5000 A). Samples were pretreated by heating with ultrapure nitric acid (Merck) at 100~ for 15 rain in PTFE beakers.
Results No lesions were noted during the external examination of the body. Macroscopic examination of the viscera revealed an intensive congestive syndrome. At the level of the esophagus and trachea, some food residues were found.
Histopalhological findings Each organ sample was specifically stained. Microscopic examination revealed an important congestion of the cardiac,
hepatic, and renal tissues and of the leptomeninges. Examination of the two pulmonary parenchyma showed lesions typical of asphyxia with marked congestive characteristics.
Toxicological findings Classical toxicological analysis revealed neither ethanol nor other volatile organic substances in the blood, normal traces of carboxyhemoglobin, and no cyanides. The only psychotropic drug found was oxazepam within the therapeutic range (0.32 mg/L). Phosphine was present in the brain (94 mL/g), liver (24 mL/g), kidney (41 mL/g), surrenals (2.4 mL/g), and heart (0.9 mL/g) but absent in the blood and urine. The phosphorus and aluminum concentrations determined and compared to their usual values in the biological media examined are given in Table I.
Discussion Deliberate poisoning by metal phosphides, particularly aluminum phosphide, is rare in France. After oral ingestion, AlP reacts with water and stomach acid to produce phosphine gas, which may account in large part for its observed toxicity. Phosphine generated in the gastrointestinal tract is readily absorbed into the blood stream. Phosphine may denature or reduce oxyhemoglobin in addition to enzymes important for respiration and metabolism and may also affect cellular membranes. The mechanism of action of phosphine is still not clear. However, some authors have claimed it is an inhibitor of cytochrome oxydase in animals (4), whereas others have shown reduction in catalase activity leading to free-radical toxicity in animals (5) and humans (6). Phosphides, like phosphine, are toxic to the central nervous system, which is excited and then depressed. There is no adequate explanation for the fact that phosphine does not cause the hemolysis that is characteristic of arsine (ASH3). Symptoms of severe toxicity by AlP ingestion include diarrhea, cyanosis, breathing difficulty,pulmonary edema, respiratory failure, tachycardia and hypotension, dizziness, and death. The most common abnormality seen in the serum of
Table I. Phosphorusand Aluminum Contents in Examined Biological Samples Biological Phosphorus Usual published Aluminum Usualpublished media found values found values Blood Urine Brain Heart Surrenal Liver Kidney
76.3 564 4.3 1.37 4.52 8.22 2.05
mg/L mg/L mg/g mg/g mg/g mg/g mg/g
25~,5 mg/L 1537.5 lag/L 681-1300 mg/L < 0.1 pg/L 2.36 + 0.5mg/g 36 pg/g 1.27 + 0.30 mg/g 4.6 pg/g -* 44 pg/g 200 + 0.05 mg/g 75 pg/g 1.75 _+0.05 mg/g 3 pg/g
9-15 lag/L < 0.1 IJg/L 2 + 1 pg/g < 1 pg/g -* 3 • 1 pg/g 3 • 1 pg/g
* No published resultsin the literature.
91
Journal of Analytical Toxicology,Vol. 24, March 2000
42% of poisoned patients was hypoxemia with severe metabolic acidosis (7). The results of postmortem examinations of victims of phosphine poisoning generally indicate hypoxia with evidence of local trauma in the gastrointestinal tract, liver, kidneys, and central nervous system. The reported lethal oral dose of aluminum phosphide for humans is about 1500 mg (three tablets). Treatment consists in early gastric washing, possible neutralization by potassium permanganate (0.01%), and symptomatic processing. Recently, the antiperoxidant effect of intravenous magnesium sulfate was demonstrated to prevent oxidative stress leading to accelerated lipid peroxidation in the early phase (0-6 h) of AlP poisoning (8,9). Arora et al. (10) reported that histopathological changes revealed varying degrees of congestion, similar to those produced by hypoxic injury. We were probably not able to find phosphine in the blood and urine because we had forgotten to add diluted sulfuric acid to the vials as recommended by Chan et al. (11), but phosphine was detected in the tissues (brain, liver, and kidneys). Phosphine gas was present inside the body because it was found in the environmental atmosphere after the corpse was dissected. Therefore, the phosphorus concentrations was significantly higher in the victim's blood than in normal blood concentrations. In the week before the autopsy, putrefaction may have transformed phosphine in the blood into a nonvolatile phosphorus compound (phosphate, hypophosphite, etc.). It was not easy to interpret these results because we found very few values in the literature, except those presented by Chan et al. (11) or those given by Chugh et al. (12), who used a colorimetric and nonspecific method for the semiquantitative estimation of phosphine blood levels in several patients with severe and moderate poisoning and who also carried out their investigations on fresh corpses. On the other hand, the aluminum and phosphorus concentrations allowed us to confirm the hypothesis of aluminum phosphide poisoning. Aluminum blood concentration has been proposed to confirm a diagnosis of oral poisoning with AlP. Routine modern techniques use atomic absorption spectrometry (AAS)or ICP. Garry et al. (13) reported an accidental death related to phosphine exposure from stored grain fumigated with AlP pellets. The aluminum blood concentration was 713 pg/L, whereas normal laboratory values range from 2 to 42 pg/L. Ten control blood samples from various autopsies have shown Al values between 9 and 15 t~g/L using emission spectrometry, whereas blood in the deceased contained 100 times as much. Similar to the phosphine results, aluminum concentrations were found to be higher in the brain, surrenals, and liver. In urine and the kidneys, no difference from usual published values was found, probably because the death occurred shortly after ingestion. This case of poisoning was probably the result of combined circumstances: availability of this pesticide in an occupational
92
context and a predisposed subject with a latent depression syndrome who had access to this product and knew of its toxicity because of his job.
Acknowledgment The authors thank Guy Bouer for his assistance in the translation of this article.
References 1. S. Singh, D. Sing, N. Wig, I. Jit, and B.K. Sharma. Aluminium phosphide ingestion--a clinico-pathologic study. Clin. Toxicol. 34(6):703-706 (1996). 2. K.S. Jayaraman. Death pills from pesticide. Nature 353(6343): 377 (1991). 3. A. Tracqui, P. Kintz, and P. Mangin. Systematic toxicological analysis using HPLC/DAD. J. Forensic Sci. 40(2)" 112-120 (1995). 4. W. Chefurka, K.P. Kashi, and E.J. Bond. The effect of phosphine on electron transport in mitochondria. Pestic. Biochem. Physiol. 6:350-362 (1976). 5. C. Bolter and W. Chefurka. Extra mitochondrial release of hydrogen peroxide from insects and mouse liver mitochondria using respiratory inhibitors--phosphine, nyxothiazol, antimycin and spectral analysis of inhibited cytochromes. Arch. Biochem. Biophys. 278(1): 65-72 (1989). 6. S.N. Chugh, V. Arora, A. Sharma, and K. Chugh. Free radical scavengers and lipid peroxidation in acute aluminium phosphide poisoning. Indian J. Med. Res. 104:190-193 (1996). 7. S.N. Khosla, R. Handa, and P. Khosla. Aluminium phosphide poisoning. Trop. Doct. 22(4): 155-157 (1992). 8. S.N. Chugh, T. Kolley, R. Kakkar, K. Chugh, and A. Sharma. A critical evaluation of anti-peroxidant effect of intravenous magnesium in acute aluminium phosphide poisoning. Magnesium Res. 10(3): 225-230 (1997). 9. U.K. Singh, B. Chakraborty, and R. Prasad. Aluminium phosphide poisoning: a growing concern in pediatric population. Indian Pediat~ 34:650-651 (1997). 10. B. Arora, R.S. Punia, R. Kalra, S.N. Chugh, and D.R. Arora. Histopathological changes in aluminium phosphide poisoning. J. Indian Med. Assoc. 93(10): 380-381 (1995). 11. L.T.F. Chart, R.J. Crowley, D. Delliou, and R. Geyer. Phosphine analysis in post mortem specimens following ingestion of aluminium phosphide. J. Anal. Toxicol. 7:165-167 (1983). 12. S.N. Chugh, R. Pal, V. Singh, and S. Seth. Serial blood phosphine levels in acute aluminium phosphide poisoning. J. Assoc. Physicians India 44(3): 184-185 (1996). 13. V.F. Garry, P.F.Good, J.C. Manivel, and D.P. Perl. Investigation of a fatality from nonoccupational aluminium phosphide exposure: measurement of aluminium in tissue and body fluids as a marker of exposure. J. Lab. Clin. Med. 122:739-747 (1993).
Manuscript received April 13, 1999; revision received June 4, 1999.
