Features of fatal pesticide ingestion in South Australia

Abstract

Pesticides are used to kill, repel or control any animal or plant species which are considered pests, but have also been associated with intentional and unintentional human fatalities. A rapid increase in pesticide suicides was observed during ‘The Green Revolution’ after pesticides were introduced into low- and middle-income rural households without appropriate guidelines for safe use and storage. While national pesticide bans have contributed to a significant decrease in pesticide-related suicides, such cases still comprise a large proportion of all suicides around the world. The aim of the current study was to provide a profile of pesticide suicides in a high-income country as a point of comparison against studies from low- and middle-income countries. Statistical analyses were performed using R (version 4.2.3). Over the 20-year study period, there were a low, yet consistent number of pesticide suicides which were most common among males over the age of 40. Paraquat and methomyl pesticides collectively contributed to almost half (48.8%) of all fatalities. Consistent with the literature, such cases often occurred with little premeditation in response to an acute emotional crisis. While interpretation of autopsy findings was mostly limited, there were some pesticides that demonstrated findings consisted with previously reported characteristics (e.g., gastroesophageal erosions with paraquat). Given the high proportion of cases where paraquat and methomyl pesticides were implicated, it may be appropriate to review the availability and accessibility of such compounds to reduce the occurrence of pesticide suicides in South Australia and potentially the wider Australian population.

Keywords

Pesticide suicide paraquat carbamate methomyl

Introduction

Pesticides (i.e., insecticides, rodenticides, molluscicides and herbicides) are materials which are used to kill, repel or control any animal or plant species which are considered pests. Pesticides are also highly hazardous chemicals which have been associated with fatalities. Agricultural advancements in the mid-90s, referred to as ‘The Green Revolution’, brought new high-yield crops in conjunction with highly hazardous pesticides which were introduced into rural households, often in low- and middle-income countries.¹ However, a lack of awareness and guidelines for safe use and storage saw a rapid increase in pesticide-related deaths, which were both unintentional and intentional.^1–6 While intervention measures such as national pesticide bans have contributed to a significant decrease in pesticide-related suicides, such cases still comprise a large proportion of all suicides around the world. Mew et al. (2017) found that between 2010 and 2014, pesticide self-poisonings accounted for approximately 13.7% of global suicides each year.⁷ However, this study also highlighted significant geographic variations reflecting agricultural farming involvement or socioeconomic status, with a higher prevalence in countries with high rates of agricultural farming and those of low- and middle-income. For example, in countries such as Sri Lanka and Bangladesh where the sale and use of industrial pesticides is widespread, rates of pesticide self-poisonings are high. Therefore, successful intervention measures have focussed on limiting access to pesticides through national bans of highly hazardous materials. These interventions have been shown to be extremely successful in reducing rates of pesticide self-poisoning in countries such as Bangladesh, Sri Lanka and South Korea.^8–10 However, for high-income countries such as the United Kingdom, South Korea and Australia, pesticide self-poisonings are comparatively less common and are more often related to the use of widely available, household products rather than industrial-grade pesticides.

The following study was, therefore, undertaken to analyse the incidence and profile of pesticide self-poisonings in a high-income country to compare with the reported features from low- and middle-income countries with high levels of agricultural activity.

Materials and methods

The database of the Toxicology Group at Forensic Science SA was searched for all post-mortem pesticide detections over a 20-year period from January 2000 to December 2019. Toxicology reports were then cross-referenced against autopsy reports to identify cases where death was attributed to pesticide ingestion. Collected variables included age, sex, medical history, scene findings, location of death, cause of death, manner of death, autopsy and toxicology findings. All cases were subject to full coronial and police investigations with complete autopsies.

Age was separated into the following categories: infant (0–12 months), toddler (1–3 years), child (4–14 years), young person (15–24 years), adult (25–64 years) and elderly (65+ years). Seasons in Australia are summer (December–February), autumn (March–May), winter (June–August) and spring (September–November). Pesticides were separated into the following categories: organophosphates, carbamates, organochlorines, herbicides and rodenticides.

Statistical analyses were performed using R (version 4.2.3). A Poisson regression was used to characterise trends in the observations. A Chi-squared test for given probabilities determined the significance of proportions for the following variables: age category, sex, season and pesticide type. A Welsh two-sample t-test determined the significance of mean age between males and females. Ethics approval for the data used in this study was granted by the University of Adelaide Human Research Ethics Committee (H-2020-033).

Results

Overview

During the 20-year study period (2000–2019), 41 cases of fatal pesticide ingestion occurred in South Australia (SA). There was no significant trend in the number of cases over time (i.e., there was a consistent number of cases across the time series, p = 0.09) as shown in Figure 1 (see Appendix A, Table A.1). The largest number of cases occurred in spring (n = 13), followed by summer (n = 11), winter (n = 9) and autumn (n = 8). However, there were no statistically significant trends observed in the seasonality of cases (p = 0.696, see Appendix A, Table A.2). Following police, coronial and pathological investigations, it was determined that all cases involved the intentional ingestion of a pesticide(s). No cases of unintentional pesticide exposure were identified.

Figure 1.

Number of fatal pesticide ingestions in SA between 2000 and 2019.

Of the 41 cases identified, 39 were attributed to pesticide toxicity in isolation, and two cases were attributed to the combined toxic effect of a pesticide in conjunction with another substance (citalopram and methylated spirits, respectively). When divided by pesticide category, herbicides were most frequently implicated (n = 15), followed by carbamates (n = 11), organophosphates (n = 9), rodenticides (n = 6) and organochlorines (n = 1). A summary of the toxicology findings for all cases is presented in Table 1. Note: the values above do not total 41 as one case involved the ingestion of two pesticide products from different pesticide categories.

Table 1.

Pesticide toxicology findings for cases of fatal pesticide ingestion in SA between 2000 and 2019.

Pesticide	No. of cases^a	Range (n)^b	Mean (μ) level	Median (x˜) level
Carbamates	11	-	-	-
Methomyl	10	2–50 mg/L (8)	21.0 mg/L	14 mg/L
Carbaryl	1	NQ	NA	NA
Herbicides	15	-	-	-
Paraquat	10^c	1–80 mg/L (5)	27.1 mg/L	20 mg/L
Diquat	9	0.76–8 mg/L (3)	3.25 mg/L	1 mg/L
Glyphosate	5	250–1100 mg/L (2)	675 mg/L	NA
Organochlorines	1	-	-	-
Endosulfan	1	0.1 mg/L (1)	NA	NA
Organophosphates	9	-	-	-
Malathion	5	0.36–5.5 mg/L (5)	1.7 mg/L	1 mg/L
Dichlorvos	1	NQ	NA	NA
Mevinphos	1	80 mg/L (1)	NA	NA
Omethoate	1	40 mg/L (1)	NA	NA
Dimethoate	1	110 mg/L (1)	NA	NA
Rodenticides	6	-	-	-
Strychnine	5	1.7–5.3 mg/L (4)	3.5 mg/L	3.5 mg/L
Aluminium phosphide	1	0.75 mg/L (1)	NA	NA

One case of mixed pesticide toxicity, therefore, cases do not total 41.

n = number of quantitative analyses, level given if only one case.

Includes nine cases of combined diquat/paraquat toxicity.

NQ: no quantitative data (‘present’).

Demographic characteristics

The decedents included 10 females (mean 71, SD 10.37) and 31 males (mean 60, SD 18.04) with an age range of 18–90 years (mean 63 years). No cases were identified in the infant, toddler and child age categories. The number of cases was highest amongst the elderly (n = 21) and adult (n = 19) age categories, with only one young person identified. There was a much greater proportion of males (76%) compared to females ( $χ_{1}^{2} = 1$ , p = 0.001, see Appendix A, Table A.4). There was also a greater proportion of elderly persons (51%) among the cohort ( $χ_{2}^{2} = 2$ , p < 0.001, see Appendix A, Table A.5). A two-sample t-test also revealed that females were between 1.7 and 20.9 years older than males on average (p = 0.022, see Appendix A, Table A.6).

Decedent histories

More than half (54%, n = 22) of decedents had a history of mental illness, including anxiety/depression (n = 19), schizophrenia (n = 3), dementia (n = 2) and alcoholism (n = 1). Additionally, seven decedents had experienced a domestic dispute or argument shortly prior to death (17%), three of whom also had a history of mental illness. Three decedents had a history of exceedingly poor health at the time of their death. Four decedents made contact with a family member after ingesting pesticide to inform them of their actions. Only one decedent had direct access to industrial-grade pesticides through employment as a fruit grower. No information was provided regarding the decedents’ histories and circumstances in six cases.

Location

The location of pesticide ingestion was most often the home address of the decedent (n = 27, 66%). Seven additional cases occurred in a public area/building and the location of the incident was not documented in the remaining seven cases. Of the decedents who ingested the pesticide(s) at their home address, 16 decedents (59%) died at the same location and the remaining 11 decedents (41%) were transported to hospital where they died. Six of the decedents who ingested pesticides in a public area/building died at the same location, and the remaining decedent died in hospital. Of the seven cases where the location of the incident was not documented, four decedents died in hospital. Overall, 61% (n = 25) of decedents died at the same location as where they ingested the pesticide and 39% (n = 16) died in hospital.

Scene findings

Unfortunately, scene findings were not available in 16 cases, most often in cases in the earlier part of the study or where death had occurred in hospital. For the remaining 25 cases where scene findings were documented, a suicide note was only found in six cases (24%). However, there were an additional four cases where the decedent informed a family member of their actions prior to death either in person or via telephone. The presence of drinking utensils used to ingest the pesticide product was noted in eight cases (e.g., water bottle, cup).

Pesticide-specific findings

Herbicides

Herbicides collectively contributed to the largest number of deaths overall (n = 15, 37%). The proportion of suicides attributed to the use of herbicides was statistically significant ( $χ_{4}^{2} = 4$ , p = 0.009, see Appendix A, Table A.3). While the use of herbicides was less consistent over time compared to other compounds, there were two obvious peaks around 2004–2008 and 2013 with a sparse number of deaths in the remaining years. Most herbicide cases involved the use of a product containing paraquat (±diquat) (n = 10), with the remaining five cases involving a glyphosate-containing product. There was no clear pattern or evidence of product preference, with alternation between paraquat and glyphosate-containing products over the study period.

The ten paraquat (and diquat) cases showed pulmonary oedema and congestion at autopsy, with early acute bronchopneumonia in four cases and intra-alveolar haemorrhage in five cases. Gastrointestinal erosions were observed inconsistently between cases, with definitive evidence of oropharyngeal/oesophageal ulcerations seen in five cases and gastrointestinal ulcerations in two cases. The trachea and bronchi were noted to show acute inflammatory changes with or without erythema in the remaining cases. Renal and hepatic necrosis was seen in four cases, with three additional cases only showing organ congestion. The remaining cases showed organs of normal appearance. Cerebral congestion and swelling were observed in two cases, with areas of acute haemorrhage observed in one.

The four cases of glyphosate toxicity where an autopsy was performed showed pulmonary oedema and congestion, with evidence of focal bronchopneumonia in one case. Oesophageal or gastrointestinal erosions were only observed in two cases, noted as erythematous corrosive gastritis and erosive epiglottitis, respectively. An additional case showed green discolouration of the oesophageal mucosa and evidence of corrosive gastritis in the stomach. Renal and hepatic necrosis were present in one case, with the remaining cases demonstrating generalised organ congestion without necrosis. Two cases had cerebral congestion and swelling, one of which also showed microscopic acute haemorrhages on histology of the brainstem with loss of Purkinje cells in the cerebellum. A summary of post-mortem findings for all cases is presented in Table 2.

Table 2.

Case details for cases of fatal pesticide ingestion in SA between 2000 and 2019.

Type	Case no.	Age/sex	Pesticide (product)	Amount ingested	Survival period	Toxicology^	Autopsy findings	Biochemistry
Carbamate	1	40M	Carbaryl (Carbaryl 800WP)	Unknown	No timings documented – found dead at scene	Carbaryl: NQ	Externally, white froth was observed around the mouth which continued into the trachea and bronchi. Faecal staining was seen over the thighs and legs. Gross oedema of the larynx and intra-alveolar oedema of the lungs was present. The stomach contained ∼450 mL of white liquid with a pungent odour, with underlying corrosive gastritis of the gastric mucosa. Histology showed pulmonary oedema in several sections, was congested elsewhere.	Not performed.
	2	66M	Methomyl (Lannate®)	Unknown	2 days in hospital – died of multi-organ failure	Methomyl: 42 mg/L	Diffuse purplish bruising of the right side of the tongue and inner aspect of lower lip. Laryngeal and tracheal mucosal surfaces congested and focally haemorrhagic, pulmonary congestion and marked oedema, the stomach contained small amount of clotted blood and bloodstained watery fluid, a 4 mm diameter area of superficial ulceration on posterior wall of distal gastric body. Histology showed pulmonary oedema and congestion, patchy early bronchopneumonia and intra-alveolar haemorrhage, liver showed widespread centrilobular haemorrhagic necrosis and haemorrhage, kidneys showed acute tubular necrosis, stomach showed autolytic changes and focal mucosal haemorrhage (no acute or chronic ulceration).	Not performed.
	3	70F	Methomyl (Lannate®)	Unknown	<24 h in hospital – died of metabolic acidosis	Methomyl: 39 mg/L Methanol: 0.023%	The mucosal surfaces of the larynx and trachea were normal, the lungs were congested and moderately oedematous. There was no evidence of gastrointestinal erosions and the stomach contained 500 mL of black, granular liquid. Histology showed the lungs to be congested with patchy early acute bronchopneumonia, the liver showed early centrilobular necrosis with interstitial oedema and early acute inflammatory infiltrate, the kidneys showed minor focal glomerular obsolescence but were otherwise normal. The brain was normal	Not performed.
	4#	77M	Methomyl (Lannate®) Dichlorvos (unknown)	Unknown	No timings documented – found dead at scene	Methomyl: 2 mg/L Dichlorvos: NQ	Strong-smelling green mucoid material was contained in the stomach, but no gastro-oesophageal erosions were present. Both lungs, the larynx and deep soft tissues of the neck were congested and oedematous. Mild swelling and cerebellar grooving were observed in the brain but were otherwise normal. Histology showed prominent vascular congestion in the lungs.	Vitreous humour: Sodium: 134 mmol/L Potassium: 20.1 mmol/L Urea: 9.7 mmol/L Creatinine: 63 µmol/L Glucose: < 0.6 mmol/L Betahydroxybutyrate: 0.15 mol/L Acetoacetate screen: Negative Cholinesterase studies not performed
	5	76M	Methomyl (Lannate®)	Unknown	<24 h – found dead at scene	Methomyl: 7 mg/L	The lungs demonstrated mild to moderate congestion and severe oedema without haemorrhage. The mucosa of the trachea and bronchi appeared normal without inflammation. The stomach contained a moderate quantity of green-coloured fluid, and the mucosal surface was unremarkable. Histology was not contributory.	Not performed.
	6	84F	Methomyl (Lannate®)	Unknown	<24 h – found dead at scene	Methomyl: 12 mg/L Methanol: NQ	The lungs showed mild to moderate congestion and severe oedema but no haemorrhage. Green fluid lined the oesophagus. The stomach contained 150 mL of greenish/blue-coloured fluid and the mucosal surface was congested with green colouration. A 1.5 cm area of erosion was present in first part of duodenum where green fluid was present and there was some fluid in the proximal jejunum. Histology was not contributory.	Not performed.
	7	73F	Methomyl (Lannate®)	Unknown	<3 h – found dead at scene	Methomyl: 3.8 mg/L Methanol: NQ	Moderate pulmonary congestion and oedema but no haemorrhage. Approx. 250 mL of green, watery fluid was contained in the stomach and a normal mucosal lining. The liver appeared congested. Histology showed mild pulmonary oedema and the other organs were unremarkable.	Vitreous Humour: Sodium: 129 mmol/L Glucose: 3.3 mmol/L Urea: 5.2 mmol/L Osmolality: 310 mosmol/kg BOH Butyrate: < 0.10 mmol/L Cholinesterase studies not performed
	8	84F	Methomyl (unknown)	Unknown	<4 h – found dead at scene	Methomyl: NQ Methanol: NQ	The trachea and main bronchi contained frothy oedema fluid and both lungs were oedematous. Blue-stained mucoid material was present in the posterior pharynx and upper oesophagus. The stomach contained 300 mL of green/grey watery liquid and the mucosa appeared normal. Histology showed mild patchy alveolar wall thickening and clubbing but no significant inflammation.	Vitreous Humour: Sodium: 127 mmol/L Glucose: 4.0 mmol/L Creatinine: 100 µmol/L Urea: 8.6 mmol/L Osmolality: 348 mosmol/kg Betahydroxybutyrate: < 0.10 mmol/L Cholinesterase studies not performed
	9	86M	Methomyl (unknown)	<1L	No timings documented – found dead at scene	Methomyl: 50 mg/L Methanol: NQ	Externally, blue liquid was seen on the lips and within the mouth of the deceased. The lungs showed dependent congestion. The mucosal surface of oesophagus was coated with a blue material, the stomach contained 150 mL of blue-green liquid with a chemical odour as did the proximal duodenum. Histology showed congestion of the lungs and the liver with an increase in inflammatory cells.	Not performed.
	10	90M	Methomyl (Lannate®)	Unknown	∼60–90 min – died at scene	Methomyl: 16 mg/L	Mucosal surfaces of the larynx, trachea and bronchi were congested. Watery blood-stained froth was present in the trachea and bronchi and both lungs were oedematous and congested. Food material was present in the oesophagus and had a slightly blue hue. 100 mL of brown fluid was present in the stomach which contained partly digested food and the mucosa appeared normal. Histology showed the lungs to be oedematous and congested.	RBC cholinesterase: 6 U/g Hb Plasma cholinesterase: 0.42 kU/L
	11	86M	Methomyl (Lannate®)	Unknown	∼30–90 min – died at scene	Methomyl: 17 mg/L	Externally, green stains were noted around the mouth and a strong odour of the body. The mucosal surfaces of larynx, trachea and bronchi were congested and contained a small amount of liquid gastric contents. Both lungs were oedematous and congested but there was no obvious evidence of gastric aspiration. The stomach contained 200 mL of greenish/grey watery fluid containing small fragments of partly digested food material with a bluish hue. On histology, the lungs were congested and oedematous.	RBC cholinesterase: 8 U/g Hb Plasma cholinesterase: 1.66 kU/L
Herbicide	12	82M	Glyphosate (Roundup®)	250mL	2 days in hospital – died of acute renal failure	Toxicology not undertaken due to haemodialysis and prolonged hospital stay	Pulmonary oedema, terminal bronchopneumonia, stomach contained 250 g of activated charcoal in a bolus, no gastric erosions noted. The brain demonstrated cerebral congestion and swelling but was otherwise normal. Histology showed focal bronchopneumonia, necrosis of proximal renal tubules and focal necroses in the liver.	Not performed.
	13	43M	Diquat/paraquat (unknown)	Unknown	3 days in hospital – died of respiratory failure	Diquat: 0.75 mg/L Paraquat: 4.5 mg/L	Thick brown mucopurulent sputum in trachea and bronchi, severe oropharyngeal and oesophageal ulceration, pulmonary congestion and oedema. 100 mL of brown watery fluid in stomach contents, patchy acute renal tubular necrosis. Histology showed scattered microscopic foci of pulmonary intra-alveolar haemorrhage, patchy hyaline membrane formation, scattered foci of alveolar neutrophilic infiltration and focal acute bronchitic changes. The liver demonstrated moderate centrilobular venous congestion but no necrosis.	Not performed.
	14	44M	Paraquat (unknown – paraquat concentrate)	>50mL	No timings documented – found dead at scene	Paraquat: 30 mg/L	Mucosal surfaces of the larynx, trachea and bronchi congested but no ulceration present. Pulmonary congestion present. No ulceration identified in the mucosa of the mouth, oesophagus or stomach. 50 mL of green-coloured watery mucoid fluid was contained in the stomach, as well as green discolouration of the duodenum and proximal small intestine. Histology showed pulmonary oedema, congestion and acute inflammatory infiltrates consistent with gastric aspiration.	Not performed.
	15	55M	Glyphosate (Roundup®)	Unknown	<6 h – found dead at scene	Glyphosate: NQ	Mucosal surfaces of the larynx, trachea and bronchi were normal. The lungs showed dependent congestion. The stomach contained approx. 50 mL of green fluid but no erosions present in the mucosa. Histology showed congestion and focal oedema of the lungs, no evidence of ulceration was identified within the gastrointestinal tract.	Not performed.
	16	68M	Diquat/paraquat (unknown)	Unknown	<24 h in hospital – died of multi-organ failure	Diquat: 1 mg/L Paraquat: 1 mg/L	The mucosa of the larynx, trachea and main bronchi was normal but reddened, the lungs were markedly congested and oedematous. The stomach contained 400 mL of black fluid. Histology showed pulmonary congestion and oedema with focal areas of intra-alveolar haemorrhage and early acute bronchopneumonia. The liver was congested and potential acute tubular necrosis of the kidneys. The brain showed early hypoxic cerebellar neuronal changes.	Not performed.
	17	57F	Glyphosate (Hortico®)	600mL	40 h in hospital – died of multi-organ failure	Glyphosate: NQ	Erythematous erosion was present overlying the epiglottis, the lungs markedly oedematous. Activated charcoal coated the tongue and there were multiple spotty erosions throughout the oesophagus, the mucosa elsewhere in the gastrointestinal tract erythematous and coated with activated charcoal. Histology showed erosive epiglottitis, acute submucosal haemorrhage of the oesophagus, congestion and oedema of the lungs, the liver showed moderate congestion and acute haemorrhages in several areas of the brain.	Not performed.
	18	38M	Diquat/paraquat (Sprayseed®)	<400mL	No timings documented – found dead at scene	Diquat: NQ Paraquat: NQ	Bloodstained fluid around nose and mouth, mucosal surface of the trachea and bronchi was erythematous, dark green fluid was present in the large airways, dependant congestion in the lungs, erosive oesophagitis, focal mucosal ulcerations and green staining of the stomach mucosa and 400 mL of dark green fluid in stomach contents. Histology showed pulmonary oedema and congestion, focal areas of acute superficial gastritis and congestion of the kidneys.	Not performed.
	19	39M	Diquat/paraquat (unknown)	2 mouthfuls	31 h – died in hospital of multi-organ failure	Diquat: NQ Paraquat: NQ	Mucosal surfaces of the respiratory tract were congested, marked pulmonary oedema and congestion, no corrosive lesions identified in the gastrointestinal tract, the stomach contained a large amount of watery fluid containing charcoal, mild generalised cerebral oedema and diffuse congestion. Histology showed pulmonary oedema and congestion with small foci of early bronchopneumonia, early necrosis of the liver, acute tubular necrosis of the kidneys and vascular congestion of the brain.	Not performed.
	20	64M	Diquat/paraquat (Sprayseed®)	∼1L	No timings documented – found dead at scene	Diquat: NQ Paraquat: 80 mg/L	Externally, green dye was noted over the deceased's face and clothing. The mucosal surfaces of the respiratory tract were normal without evidence of gastric aspiration. The lungs showed patchy congestion. Green dye was also present within the oral cavity with dark green discolouration of the oesophageal, gastric and intestinal mucosa. The stomach contained 1100 mL of dark green fluid. Histology showed the lungs to be oedematous and congested but other organs appeared normal.	Not performed.
	21#	64F	Glyphosate (Yates Zero®)	Unknown	No timings documented – found dead at scene	Glyphosate: 1100 mg/L Alcohol (methylated spirits): 0.75%	White froth was exuding from the nose and the body smelled of glyphosate. Frothy white fluid was present in the trachea and main bronchi and the lungs were congestion and oedema. The mucosa of the oesophagus and stomach was discoloured green. The stomach also showed evidence of black corrosive gastritis and contained 750 mL of green fluid. Histology showed congested and oedema of the lungs, loss of cellular detail with streaming of nuclei in the oesophagus and the stomach showed accumulation of eosinophilic fluid within the lamina propria in association with focal haemorrhage.	Not performed.
	22	42M	Diquat/paraquat (Sprayseed®)	∼1L	6 h – died in hospital of multi-organ failure	Diquat: NQ Paraquat: NQ	Traces of charcoal were noted in the mouth and a general pesticide odour was noted of the internal organs. The mucosal surfaces of the larynx, trachea and bronchi were oedematous and congested. The lungs showed severe haemorrhagic oedema and congestion. The mucosal surface of the oral cavity was oedematous and the oesophagus epithelium showed diffuse sloughing. The stomach contained watery fluid containing charcoal and the small intestine contained greenish mucoid fluid and mucosal oedema. Histology showed congestion and oedema of the lungs and early acute tubular necrosis of the kidneys. The oesophagus, pharynx and stomach also showed oedema with associated patchy haemorrhage and/or necrosis.	Not performed.
	23	43M	Diquat/paraquat (unknown)	Unknown	14 days in hospital – died of multi-organ failure	Diquat: NQ Paraquat: NQ	The mucosal surfaces of the larynx, trachea and bronchi were normal. The lungs showed areas of haemorrhage. The stomach contained green liquid with no evidence of corrosive injury. Histology showed diffuse alveolar haemorrhage and oedema of the lungs, autolysis of other organs but otherwise normal.	Not performed.
	24	61M	Diquat/paraquat (Sprayseed®)	Unknown	∼3 h, found collapsed at home – died in hospital	Diquat: 8 mg/L Paraquat: 20 mg/L	The larynx contained copious green-blue mucoid secretions and the mucosal surface appeared congested, as did the mucosa of the trachea and bronchi. The lungs were oedematous and congested. There was green-blue discoloration of the tongue, oesophageal, gastric and small intestinal mucosa. The stomach contained mucoid secretions and turbid brown fluid. Histology showed oedema and congestion of the lungs and acute inflammatory changes within the oesophagus.	Not performed.
	25	87M	Glyphosate (unknown)	100mL	<24 h – died in hospital of multi-organ failure	Glyphosate: NQ	Autopsy not performed – cause of death could be ascertained from scene findings and clinical history. ENT consultant found circumferential ulceration in the hypopharynx and on the epiglottis.	Not performed.
	26	56F	Diquat/paraquat (Sprayseed®)	1 mouthful	∼18 h – died in hospital of multi-organ failure	Diquat: NQ Paraquat: NQ	The lips showed blue-green staining which continued into the mouth without evidence of erosion. The trachea and main bronchi contained prominent mucous and oedema fluid. The lungs were very heavy due to pronounced pulmonary oedema. A small amount of black liquid was contained in the stomach, but no evidence of erosion. Histology showed prominent congestion with areas of haemorrhage, and early acute inflammatory infiltrates within the bronchi and occasional foci of early bronchopneumonia within the alveolar spaces. The oesophagus demonstrated inflammatory cell infiltrates and mucosal oedema.	Not performed.
Organochlorine	27	33M	Endosulfan (unknown)	Unknown	∼30–120 min – found collapsed at home, died at scene	Endosulfan: 0.1 mg/L	No specific autopsy findings were noted. Generalised pulmonary and cerebral congestion was noted, in conjunction with a strong odour of hydrocarbons in the stomach contents. Histology also showed pulmonary oedema and congestion, centrilobular venous congestion of the liver and generalised congestion in the brain without any evident hypoxic-ischaemic changes.	Not performed.
Organophosphate	28	80F	Malathion (unknown)	Unknown	No timings documented – found dead at scene	Malathion: 1.1 mg/L	Strong smell consistent with malathion, 150 mL of white material within stomach, oesophagus and mouth, macroscopic evidence of bilateral basal aspiration of gastric contents. Histology showed pulmonary congestion but was otherwise normal.	Not performed.
	29	51M	Malathion (unknown)	100mL	Unknown – autopsy report not available	Malathion: 5.5 mg/L	Not available.	Not available.
	30#	66F	Malathion (unknown)	Unknown	No timings documented – found dead at scene	Malathion: 0.5 mg/L Citalopram: 0.96 mg/L Note: evidence of ammonia and/or hypochlorite ingestion present at scene but could not be tested for	Odour of insecticide noted of body, pulmonary oedema and congestion, no corrosion identified in the gastrointestinal tract, 200 mL of watery mucoid brown fluid in stomach contents. Histology showed pulmonary oedema, congestion and microscopic foci of acute necrosis of the oesophageal epithelium with acute inflammation which was consistent with ingestion of either ammonia and/or hypochlorite.	Not performed.
	4#	77M	Methomyl (Lannate®) Dichlorvos (unknown)	Unknown	<24 h – collapsed and died at scene	Methomyl: 2 mg/L Dichlorvos: NQ	Strong-smelling green mucoid material was contained in the stomach, but no gastro-oesophageal erosions were present. Both lungs, the larynx and deep soft tissues of the neck were congested and oedematous. Mild swelling and cerebellar grooving were observed in the brain but were otherwise normal. Histology showed prominent vascular congestion in the lungs.	Vitreous humour: Sodium: 134 mmol/L Potassium: 20.1 mmol/L Urea: 9.7 mmol/L Creatinine: 63 µmol/L Glucose: < 0.6 mmol/L Betahydroxybutyrate: 0.15 mol/L Acetoacetate screen: Negative Cholinesterase studies not performed
	31	73M	Mevinphos (unknown)	Unknown	∼1–3 h – found dead at scene	Mevinphos: 80 mg/L	A small amount of white, creamy material was present in the oesophagus, and the stomach contained reddish-brown turbid fluid. Both lungs were congested and oedematous macroscopically and on histology. The liver also demonstrated mild centrilobular venous congestion.	Not performed.
	32	58M	Omethoate (unknown)	Unknown	∼2–6 h – found dead at scene	Omethoate: 40 mg/L	A chemical odour was noted in the air at the death scene. The glottic inlet and both lungs were noted to be congested and oedematous. Green-coloured gastric contents were present within the airways and the stomach contained green, watery fluid. Generalised congestion was noted of the brain but otherwise appeared normal. On histology, one of five lung sections showed evidence of acid digestion with traces of birefringent foreign material noted in the adjacent airway.	Not performed.
	33	67M	Dimethoate (Rover®)	Unknown	No timings documented – found dead at scene	Dimethoate: 110 mg/L	A strong chemical odour was noted of the body, incised wounds present on both wrists. The lungs showed dependent congestion, with the mucosal surfaces of the respiratory tract appearing normal. 350 mL of odorous brown liquid was contained within the stomach. Histology was not performed as the cause of death could be ascertained from the toxicology results.	RBC cholinesterase: 6 U/g Hb Plasma cholinesterase: 0.57 kU/L
	34	80M	Malathion (unknown)	Unknown	∼2–4 h – died in hospital	Malathion: 1 mg/L*	The lungs showed severe pulmonary oedema and acute aspiration of gastric contents was evident with a large amount of clear-white non-purulent material in the air passages. The stomach contained a moderate amount of grey mucoid material. Histology was not contributory.	Results of cholinesterase studies not available
	35	78F	Malathion (unknown)	Unknown	No timings documented – found dead at scene	Malathion: 0.36 mg/L*	Clear/white frothy material present in pharynx and trachea, non-purulent mucoid material present in the distal bronchi, moderate pulmonary oedema, opaque and granular pale white-grey gastric contents and likely tablet residues. Histology showed pulmonary oedema and congestion and centrilobular necrosis of the liver.	RBC cholinesterase: 3 U/g Hb Plasma cholinesterase: 0.03 kU/L
Rodenticide	36	72M	Strychnine (unknown)	Unknown	Unknown – no autopsy report	Strychnine: 3.8 mg/L	Not available.	Not available.
	37	55M	Strychnine (unknown)	Unknown	14 day survival period in hospital – developed anoxic encephalopathy	Hospital admission blood sample not available, post-mortem toxicology not undertaken due to prolonged hospital stay	The stomach contained 200 mL of yellow-brown fluid. Pulmonary oedema, congestion and patchy changes consistent with adult respiratory distress syndrome on histology. Histology was otherwise non-contributory.	Not performed.
	38	69M	Strychnine (unknown)	Unknown	Unknown – no timings recorded on autopsy report	Strychnine: 3.2 mg/L	Externally, tiny spots of purple dye were noted on the upper and lower lip. A small amount of white froth was present in the trachea and bronchi and the mucosal surfaces were normal. The lungs were oedematous and congested. The stomach contained 300 g of partly digested food material including traces of purple dye. Histology showed pulmonary oedema and congestion, other organs within normal limits.	Not performed.
	39	18M	Strychnine (unknown)	1 teaspoon	8 day survival period in hospital – developed anoxic encephalopathy	Strychnine: 1.7 mg/L	Mucosal surfaces of the pharynx, larynx, trachea and main bronchi were normal. A small amount of bloodstained fluid was present in each pleural cavity. Gastric mucosa was normal, and the stomach contained a large quantity of bile-coloured fluid. Histology showed pulmonary congestion and haemorrhage. Extensive hypoxic changes were noted in the brain.	Not performed.
	40	46M	Strychnine (unknown)	Unknown	No timings documented – found dead at scene	Strychnine: 5.3 mg/L	Externally, post-mortem animal activity was noted on the face. Red fluid was present in the right main stem bronchus, the respiratory mucosa was normal. The lungs exhibited congestion. The stomach contained plentiful grey liquid but appeared otherwise normal. Histology showed red blood cells in the alveolar spaces of the lungs and other organs appeared normal.	Not performed.
	41	47M	Aluminium phosphide (Fumitoxin®)	NA (inhaled)	<2 h – died at scene	Aluminium: 0.75 mg/L	Phosphine gas level was greater than 3 parts per million when the body bag was first opened. Therefore, the autopsy was delayed until levels reached zero. Generalised pulmonary oedema and congestion but no aspiration of gastric material. The stomach was removed intact and unopened to confine any potential phosphine hazard, the contents of which contained 22000 mg of aluminium per kg. Histology was not contributory.	Not performed.

^Only substances implicated in the cause of death are listed.

*Level detected likely to be significantly lower than the perimortem concentration due to post-mortem degradation

# Mixed-substance toxicity (includes one case of mixed-pesticide toxicity)

NQ: no quantitative data (‘present’).

Carbamates

Ingestion of carbamate pesticides was consistent across the study period (n = 11, 27%), with 1–4 deaths every five years. The majority of carbamate cases involved a product containing methomyl (n = 10, 91%), with only one case involving a carbaryl-containing product which occurred at the beginning of the study period. Cases due to carbamate toxicity exhibited a wide variety of non-specific pathological findings. The methomyl cases consistently showed pulmonary oedema with or without early acute bronchopneumonia and intra-alveolar haemorrhage. Gastrointestinal erosions were not frequently seen, only observed in two out of ten cases. Instead, findings were more representative of terminal multi-organ failure. The autopsy findings for the single case of carbaryl poisoning are summarised in Table 2.

Organophosphates

Ingestion of organophosphate pesticides was also consistent across the study period (n = 9, 22%) with 1–4 deaths every five years. These cases most often involved a product containing malathion (n = 5, 63%), with the three remaining cases involving dimethoate, omethoate and mevinphos, respectively. There was one additional case involving the ingestion of both dichlorvos and methomyl and is subsequently listed twice in Table 2 for both carbamate and organophosphate-related cases (Case 4). Malathion use was also consistent over the study period, with the latter three cases involving dimethoate, omethoate and mevinphos occurring in the middle point of the study period.

The four cases of malathion ingestion where autopsy results were available demonstrated somewhat inconsistent findings. Pulmonary oedema and congestion were present with aspiration of gastric contents in two cases. The liver showed early necrosis in one case, but this case was also associated with a high (but not lethal) paracetamol level. The autopsy findings for the single cases involving ingestion of dichlorvos, mevinphos, omethoate and dimethoate are summarised in Table 2.

Rodenticides

Use of rodenticides was confined to the earlier years of the study period, with no cases after 2011. Five of the six rodenticide cases involved the ingestion of strychnine, and the remaining case involved the inhalation of phosphine gas through the use of aluminium phosphide.

The four cases of strychnine ingestion where autopsy results were available demonstrated generalised pulmonary congestion and oedema, with patchy changes associated with adult respiratory distress syndrome in one case. Histology showed focal areas of collapse and haemorrhage in the lungs with red blood cells in the alveolar spaces. No notable brain changes were observed except for extensive hypoxic change in one case. The autopsy findings for the single case involving inhalation of aluminium phosphide are summarised in Table 2.

Organochlorines

Cases involving organochlorines were the least frequently observed (n = 1). The single organochlorine case utilised a product containing endosulfan and the autopsy findings were non-specific.

Biochemistry findings

Submission of blood samples for biochemical cholinesterase analyses was only undertaken for five cases; two involving malathion, two involving methomyl and one involving dimethoate. While malathion and dimethoate are organophosphate pesticides and methomyl a carbamate derivative, both classes of compounds elicit their toxicity through inhibition of cholinesterase. Unfortunately, the results of one case involving malathion were unavailable. The mean values for red blood cell and plasma cholinesterase levels for the remaining four cases were 5.75 U/g Hb (range 3–8 U/g Hb) and 0.67 kU/L (range 0.03–1.66 kU/L), respectively.

Discussion

It has been estimated that pesticide self-poisonings account for up to 20% of all suicides globally.⁷ In Western Pacific low- and middle-income countries, cases comprise up to 48.3% of all suicides,⁷ a situation likely driven by poverty and availability. It is noted, however, that suicide methods vary markedly among communities with caustic agents being favoured over pesticides in countries such as Surinam.¹¹ However, in Australia, intentional pesticide self-poisoning only accounted for 0.2–0.5% of all suicides between 2012 and 2021.¹² The comparatively high proportion of pesticide suicides in low- and middle-income countries is largely attributed to widespread engagement in agricultural activities and the ready availability of highly hazardous pesticides. Over the last 20 years, national bans of these agents have resulted in significant reductions in pesticide suicides in several countries,^{6,8–10,13,14} particularly in China which previously contributed to more than half of the world's pesticide suicides.¹⁵

While pesticide suicides are not nearly as prevalent in Australia (and SA specifically) compared to other countries and communities, it is important to recognise that while low in number, they have continued to occur over time.¹⁶ The prevalence and lethality of pesticide ingestion in countries such as China is attributed to a lack of appropriate training and tailored treatment protocols.¹⁷ The absence of fatal unintentional pesticide ingestion cases within the current study highlights the effectiveness of stringent pesticide storage guidelines and public awareness of pesticide toxicity in reducing cases of fatal inadvertent exposure.

The demographic profile of pesticide suicides in the current study differed greatly from countries such as Sri Lanka, where most cases involve young males.^18,19 A more recent analysis showed that globally, pesticide poisonings were more prevalent among young females.²⁰ However, a high number of countries with known high rates of pesticide suicide were not included in this study as they had not provided data to the World Health Organisation.²⁰ While the majority of decedents in the current study were also male (76%), more than half of the cohort were elderly (51%). Overall, 90% (n = 37) of decedents were over the age of 40 years.

There are some similarities in the circumstances surrounding the ingestion of pesticides between countries. Studies have shown that those who commit suicide by ingesting pesticides often make the decision to utilise the method at short notice without significant premeditation in response to an acute emotional crisis.^17,21–23 A study of individuals who intentionally ingested pesticides in Sri Lanka highlighted that very few patients purchased pesticides prior to the suicide attempt,²² demonstrating the impulsive nature of the act in most cases. This scenario was observed in seven cases in the current study, where the decedent ingested a pesticide product shortly after an argument or dispute with a partner or family member. The choice of pesticide ingestion rather than another method of suicide was largely attributed to easy availability, utilising substances that were already in the home, perhaps in the absence of a feasible or tolerable alternative.²² Certainly, the availability and accessibility of a method of suicide has been shown to be a determinant of the profile of suicide methods in different communities, particularly those involving non-pharmaceutical substances.^24,25

Historical studies of acute pesticide exposures show that the most frequently implicated compounds were organophosphates and organochlorines in Sri Lanka and Malaysia and bipyridyls in Thailand.^2,26 In the current study, the most frequently implicated category of pesticide was herbicides (n = 15), specifically products containing paraquat (± diquat) which contributed to 67% of herbicide cases and to 24% of all cases overall (n = 10). The second most frequently implicated pesticide category was carbamates, where methomyl-containing products contributed to 91% of the carbamate pesticide cases and to 24% of all cases (n = 10). Collectively, paraquat- and methomyl-containing compounds contributed to almost half (48.8%) of all pesticide-related deaths in SA over the 20-year study period.

Most cases of pesticide ingestion in the current study occurred at the home address of the decedent (66%), with only a few cases ingesting the pesticide elsewhere (17%). The majority of decedents died at the same location as where the incident occurred (61%) which is reflective of the acute nature of pesticide toxicity when ingested. In 2010, the compounds responsible for the largest proportion of hospital admissions and deaths (94% and 98% respectively) in Sri Lanka were organophosphates, paraquat, MCPA, propanil and glyphosate.¹⁸ Those with the highest case fatality were paraquat, dimethoate and fenthion, with fatality rates of 42%, 20.6% and 14.8%, respectively.¹⁸ The case fatality rate of aluminium phosphide has also been documented in cases from India.²⁷ While it appears that the toxicity and lethality of paraquat is well documented in the literature, cases involving fatal ingestion of methomyl-containing compounds have not been widely reported until now. In the current study, several decedents notified a relative of their actions immediately following the incident, further emphasising the importance of rapid medical attention in cases of suspected pesticide ingestion to the survival of patients. The details surrounding the incident, particularly the type and dose of pesticide ingested is integral to ensuring treatment protocols are effective.

Safety data for pesticide products provide detailed information regarding the dose-dependent clinical symptoms, and by extension, an indication of the post-mortem findings that may be observed at autopsy. For example, paraquat causes severe cellular toxicity and has been associated with oropharyngeal burns and multi-organ failure comprising renal and hepatic necrosis, myocarditis and pulmonary haemorrhage with high doses.²⁸ These findings were observed in most paraquat cases in the current study, where oropharyngeal burns were also observed to varying degrees in addition to variable progressions of multi-organ failure. However, post-mortem findings for other pesticide products were largely inconsistent, where their interpretation was further complicated by the ingestion of other substances, prolonged survival period, medical intervention and limited information on the circumstances of the incident (e.g., dose, timing etc).

Several pesticides highlighted in the current study have been banned in many countries. However, some are still used in Australia including paraquat, chlorpyrifos, dimethoate, malathion and dichlorvos.²⁹ Given that paraquat contributed to 24% of all cases of the current study, it may be pertinent to reconsider the availability and accessibility of these compounds to the general public in an effort to reduce the incidence of fatal pesticide suicides. However, in countries such as Australia where pesticide suicides are rare, it may be difficult to determine the effect of prevention measures, if implemented.

Supplemental Material

sj-docx-1-msl-10.1177_00258024231197914 - Supplemental material for Features of fatal pesticide ingestion in South Australia

Supplemental material, sj-docx-1-msl-10.1177_00258024231197914 for Features of fatal pesticide ingestion in South Australia by Lilli Stephenson, Corinna Van Den Heuvel, Melissa Humphries, Christine Nash and Roger W. Byard in Medicine, Science and the Law

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Lilli Stephenson

Roger W. Byard

Supplemental material

Supplemental material for this article is available online.

References

Karunarathne

Gunnell

Konradsen

, et al.

How many premature deaths from pesticide suicide have occurred since the agricultural Green Revolution?

Clin Toxicol (Phila) 2020; 58: 227–232.

Jeyaratnam

Lun

. Survey of acute pesticide poisoning among agricultural workers in four Asian countries. Bull World Health Organ 1987; 65: 521–527.

Jeyaratnam

. Acute pesticide poisoning: a major global health problem. World Health Stat Q 1990; 43: 139–144.

Wesseling

McConnell

Partanen

, et al. Agricultural pesticide use in developing countries: health effects and research needs. Int J Health Serv 1997; 27: 273–308.

Eddleston

Sheriff

Hawton

. Deliberate self harm in Sri Lanka: an overlooked tragedy in the developing world. Br Med J 1998; 317: 133–135.

Gunnell

Fernando

Hewagama

, et al. The impact of pesticide regulations on suicide in Sri Lanka. Int J Epidemiol 2007; 36: 1235–1242.

Mew

Padmanathan

Konradsen

, et al. The global burden of fatal self-poisoning with pesticides 2006–15: systematic review. J Affect Disord 2017; 219: 93–104.

Kim

Min

Lee

. Suicide overall and suicide by pesticide rates among South Korean workers: a 15-year population-based study. Int J Environ Res Public Health 2019; 16: 4866.

Knipe

Gunnell

Eddleston

. Preventing deaths from pesticide self-poisoning-learning from Sri Lanka's success. Lancet Glob Health 2017; 5: e651–e652.

10.

Chowdhury

Dewan

Verma

, et al. Bans of WHO Class I Pesticides in Bangladesh-suicide prevention without hampering agricultural output. Int J Epidemiol 2018; 47: 175–184.

11.

Byard

. Caustic ingestion-a forensic overview. J Forensic Sci 2015; 60: 812–815.

12.

Australian Bureau of Statistics. Causes of Death, Australia, 2021, https://www.abs.gov.au/statistics/health/causes-death/causes-death-australia/2021#data-downloads (2022, accessed 29 December 2022).

13.

Myung

Lee

G-H

Won

H-H

, et al. Paraquat prohibition and change in the suicide rate and methods in South Korea. PLoS One 2015; 10: e0128980.

14.

Cha

Chang

Gunnell

, et al. Impact of paraquat regulation on suicide in South Korea. Int J Epidemiol 2016; 45: 470–479.

15.

Page

Liu

Gunnell

, et al. Suicide by pesticide poisoning remains a priority for suicide prevention in China: analysis of national mortality trends 2006-2013. J Affect Disord 2017; 208: 418–423.

16.

Pannell

Gilbert

Gardiner

, et al. Death due to malathion poisoning. J Clin Forensic Med 2001; 8: 156–159.

17.

Phillips

Yang

Zhang

, et al. Risk factors for suicide in China: a national case-control psychological autopsy study. Lancet 2002; 360: 1728–1736.

18.

Dawson

Eddleston

Senarathna

, et al. Acute human lethal toxicity of agricultural pesticides: a prospective cohort study. PLoS Med 2010; 7: e1000357.

19.

Van der Hoek

Konradsen

Athukorala

, et al. Pesticide poisoning: a major health problem in Sri Lanka. Soc Sci Med 1998; 46: 495–504.

20.

Kõlves

de Leo

. Suicide methods in children and adolescents. Eur Child Adolesc Psychiatry 2017; 26: 155–164.

21.

Gunnell

Eddleston

. Suicide by intentional ingestion of pesticides: a continuing tragedy in developing countries. Int J Epidemiol 2003; 32: 902–909.

22.

Eddleston

Karunaratne

Weerakoon

, et al. Choice of poison for intentional self-poisoning in rural Sri Lanka. Clin Toxicol 2006; 44: 283–286.

23.

Eddleston

Phillips

. Self poisoning with pesticides. Br Med J 2004; 328: 42–44.

24.

Austin

Winskog

van den Heuvel

, et al. Recent trends in suicides utilizing helium. J Forensic Sci 2011; 56: 649–651.

25.

Stephenson

Wills

van den Heuvel

, et al. Increasing use of sodium nitrite in suicides-an emerging trend. Forensic Sci Med Pathol 2022; 18: 311–318.

26.

Jeyaratnam

de Alwis Seneviratne

Copplestone

. Survey of pesticide poisoning in Sri Lanka. Bull World Health Organ 1982; 60: 615–619.

27.

Chugh

Dushyant

, et al. Incidence & outcome of aluminium phosphide poisoning in a hospital study. Indian J Med Res 1991; 94: 232–235.

28.

AIRR Apparent Pty Ltd. Safety Data Sheet - Apparent Paraquat 250 Herbicide, https://apparentag.com.au/documents/msds/66103_APPARENT_PARAQUAT_250_HERBICIDE_MSDS_1.pdf (2021, accessed 13 April March 2023).

29.

Davies

. The dirty dozen: 12 pesticides that are banned elsewhere but still used in Australia, https://www.theguardian.com/australia-news/ng-interactive/2022/sep/27/12-pesticides-banned-elsewhere-but-still-used-legal-in-australia-what-is-paraquat-atrazine-fiprinol-pesticide-chemicals-food-crops-animals (2022, accessed 3 January 2023).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB