In times of drought and food scarcity, remote, impoverished, and rural communities may turn to unusual or even dangerous foodstuffs to survive. During 2007, a fatal outbreak of illness occurred among villagers in Bangladesh when they consumed Xanthium strumarium seedlings (commonly known as cocklebur) after destructive monsoon flooding eliminated most of the available local food supply.¹ Consumption of large amounts of these seedlings resulted in nausea, vomiting, loss of consciousness, hepatotoxicity, and in some cases, death among villagers.¹ Epidemics of lathyrism and neurolathyrism, a condition resulting in irreversible spastic paraparesis (extremity weakness) of the lower limbs resulting from ingestion of the grasspea (Lathyrus sativus), have occurred throughout history in times of drought and food shortages.^2,3 Konzo, a permanent spastic para- or tetraparesis, is reported to occur during dry seasons and is associated with consumption of improperly prepared cassava (Manihot esculenta).⁴

Seeds, wheat seed in particular, may be used to prepare foodstuffs such as homemade bread by certain populations even when food is not scarce. An outbreak of severe alkyl mercury poisoning occurred in Iraq during 1971–1972 when villagers prepared bread from seed treated with a mercurial fungicide. Previous similar outbreaks had occurred in 1956 and 1960 in Iraq.⁵ During the growing seasons of 1963–1965, an outbreak of methylmercury dicyandiamide poisoning occurred in Guatemala from wheat seed treated with this agent.⁶ Based on historical outbreaks, consumption of pesticide-treated seed or food products prepared directly from pesticide-treated seed can present an acute health risk.

During a recent international outbreak investigation, the local community was discovered to be chronically consuming wheat seed intended for planting. Staff from the Malawi Ministry of Health (MOH) and the Centers for Disease Control and Prevention jointly investigated an outbreak of neurological illness in a rural village that straddled the border of Malawi and Mozambique in 2009. The etiology of the outbreak was ultimately determined to be typhoid fever, a systemic illness caused by Salmonella enterica serovar Typhi. Similar neurological signs have been previously reported as a complication of typhoid fever.^7–9 During the outbreak investigation, the team discovered that villagers were consuming wheat seed provided by a non-governmental organization (NGO) as part of a program in which seed meant for planting is provided at little or no cost and a certain amount is later repaid after harvest. There had been a poor crop yield last season and there was a resultant food shortage, hence the community was directly consuming some of the seed. The seed had a pink-colored dusty coating that often indicates that a pesticide was applied. Community members would wash the seeds before eating them, believing that this removed the pesticides. To determine whether a public health threat existed from this practice, a sample of unwashed and washed wheat seed was collected from local villagers and shipped to a US Food and Drug Administration (FDA) laboratory for pesticide analysis.

The objectives of this investigation were as follows: (1) to determine which, if any, pesticides were present on both unwashed and washed seed samples, (2) determine whether rinsing in water affects pesticide concentration, and (3) determine whether pesticide concentrations, if present, represent a potential public health threat.

Of the 365 compounds tested, the following pesticides were detected and quantified: carboxin, pirimiphos methyl, permethrins, and carbaryl. Results are listed in Table 1. The remaining compounds were not detected.

Carboxin, pirimiphos methyl, permethrins, and carbaryl were found on both washed and unwashed samples of wheat seed intended for planting but being consumed directly by a community straddling the border of Malawi and Mozambique. Carboxin in particular had substantially elevated concentrations in both washed and unwashed seed samples. Washing did not remove all pesticides: consumption of washed seed samples may still expose the consumer to potentially harmful pesticides.

When compared with the other agents, carboxin had the highest concentrations on both washed and unwashed seed samples. Very little is known about the metabolism, biotransformation, and toxicity of this compound in humans. No reports of acute or chronic oral dosing studies or outbreaks of illness in humans were identified in the literature. One case report of a 7-year-old boy who ate several handfuls of carboxin-treated wheat seed was identified. This patient developed vomiting and headache within the first hour of ingestion, but symptoms rapidly improved after administration of an emetic.¹⁴ It is unknown if the patient’s symptoms were due to carboxin, another agent, or the seeds themselves (e.g., mechanical irritation). Rodent studies have identified liver and kidney changes as a result of chronic exposure to carboxin.¹⁵ No acute dietary risk assessments were performed, and chronic risks from food in the United States were determined to be below the US EPA’s level of concern; therefore, little data are available on this topic.¹⁵ At this time, there is insufficient information to determine whether a chronic health threat exists from exposure to foodstuffs with these amounts of carboxin.

Pirimiphos methyl is an organic phosphorous compound (OPC). OPCs are used as insecticides and irreversibly inhibit the enzyme acetylcholinesterase, which breaks down acetylcholine at neuronal synapses and the neuromuscular junction.¹⁶ OPC acutely poisoned patients may develop signs and symptoms of a peripheral neuropathy after poisoning (evidence of peripheral neuropathy was not consistently observed in the patients examined in this outbreak). Some occupational exposure studies (typically with inhalational and/or dermal exposures of OPCs) have also found associations between chronic OPC exposure and electroencephalographic abnormalities, sensory and motor peripheral neuropathies, neuromuscular dysfunction, neuropsychiatric and behavioral disturbances, parkinsonism, contact dermatitis, and even asthma.^16–20 Although testing results on both washed and unwashed seed samples did not exceed the tolerance value in aspirated fractions of grain used for comparison, they did slightly exceed the tolerance values for it two other grains (corn and sorghum). In addition, estimated daily exposures to this agent from consumed wheat seed did exceed the RfD. They did not, however, exceed the experimental dose listed by the EPA as the no observed effect level or NOEL (as demonstrated by transient plasma cholinesterase activity depression), measured as 0.25 mg/kg/day in animal studies.²¹ Therefore, acute toxicity at these exposure amounts is probably unlikely. Exposure to this agent in these doses probably does not pose an acute public health threat; however, the risks associated with chronic exposure to this agent in these amounts, including the development of peripheral neuropathy, are unclear.

Permethrins belong to a class of insecticides known as the type 1 pyrethroids. Pyrethroids are synthetic derivatives of naturally occurring substances with insecticidal activity found from extracts of the flower Chyrsanthemum cinerariaefolium. They have little mammalian toxicity because of their rapid metabolism and do not persist for long in the environment. In animals, large, acute doses of type 1 pyrethroid toxicity can cause extensive tremors, twitching, increased metabolic rate, and hyperthermia (fever). However, in general, the type 1 pyrethroids are unlikely to cause systemic toxicity in humans unless large amounts are ingested (e.g., suicide attempt).²² No reports of chronic oral dosing studies with permethrin in humans were identified in the literature from which to compare findings from this outbreak. Although testing results for permethrins were above existing tolerance values, and the estimated daily exposure was similar to the RfD, the NOEL for this agent (based on animal data) is much higher (5 mg/kg/day).²³ Based on what is known about permethrins, the estimated exposure amounts from this study, and the fact that any actual dose (implies absorption) is less than the estimated exposure, we feel that permethrins (in this situation), most likely do not pose a significant acute or chronic public health threat.

Carbaryl is a carbamate insecticide that inhibits acteylcholinesterase, the enzyme responsible for metabolizing the neurotransmitter acetylcholine after normal neuronal depolarization. It does this reversibly, in contrast to the similar acting OPCs, which, over time, will bind irreversibly. Peripheral neuropathy may occur due to chronic oral administration of carbamates.²⁴ However, chronic human ingestional exposures to carbamates and their associated toxicity patterns are incompletely described and defined. One human study that orally dosed volunteers with carbaryl at 0.06–0.13 mg/kg daily for 6 weeks found no deleterious changes. In this case, the maximum daily dose for a 70-kg adult would have been 9.1 mg; the estimated daily exposure amount in this study was 3.27 mg.²⁵ Long-term follow-up of acutely or chronically carbamate exposed individuals in general has been incomplete, but a variety of neurobehavioral effects have been reported such as fatigue, lethargy, mood changes, poor coordination, and others.^26,27 Chronic carbamate exposure in these doses would not be expected to pose a significant acute or chronic health threat based on currently available data.

Anecdotal reports from Uganda and other countries suggest that the practice of eating seed donated or otherwise provided from NGOs for planting may be widespread in Africa and other parts of the world, especially during periods of food insecurity. However, there is an absence of data characterizing the extent of this practice and associated health risks from long-term exposure. Washing appeared to reduce the amount of three pesticides (carboxin, permethrins, and carbaryl) in our study but did not remove all of any of the pesticides. Of particular note was that washing did not appear to reduce the concentration of carboxin to an amount considered even close to an exposure without any health risk (below the RfD) in our estimated exposure scenarios. These limited results suggest that washing alone is not effective in removing all risk of pesticide exposure, but may reduce the risk of exposure for some pesticides. The reason for the small increase (4.6%) in pirimiphos methyl concentration in the washed seed sample when compared with the unwashed sample is unclear, but may be because of normal measurement variability associated with testing concentrations in such small amounts. Although this was a very limited study (two samples), it does raise concern, that this practice may be associated with the potential for a public health threat from certain pesticides such as carbaryl and primiphos methyl, especially in the chronic exposure setting. The extent and severity of any such threat is dependent on type and concentration of pesticide used, washing practices, duration, and type (consistent, intermittent, etc…) of exposure.

Several suggestions were communicated to the Malawi MOH for consideration. They included consideration toward enacting a public health information campaign to advise villagers that washing does not remove pesticides from the seeds and advising villagers to discontinue the practice of eating either washed or unwashed seeds intended primarily for planting due to unknown health risks. Finally, there is a known health risk associated with exposure to OPCs. In the estimated exposure scenarios discussed previously, there is potential to easily exceed the pirimiphos methyl RfD guidance value. Implementation of surveillance for syndromes consistent with cholinergic poisoning should be considered.

There is very little information on clinical effects resulting from chronic ingestion of these agents. The amounts ingested as outlined in Table 1 are estimated exposure amounts and not true doses. These agents would have to be absorbed into the body after ingestion, and each has a different bioavailability value. Therefore, the actual dose from the exposure would likely be less than what has been estimated. It is important to note that in the United States, if no tolerance value exists for a chemical for wheat, then any amount of agent present would not be tolerated and the foodstuff would be removed from the market. However, this practice may not apply to all countries; final decisions regarding acceptability or tolerance values are up to the appropriate national or local public health authority.

Populations directly consuming seed meant for planting are likely at risk for unintended exposure to pesticides. Washing may reduce, but not completely eliminate, the concentration of some, but not all pesticides. Adverse effects on health will be dependent on factors such as type and amount of pesticide used, amount of seed consumed‘ and domestic food handling practices. The risk of long-term effects from chronic exposure to pesticides such as carboxin and pirimiphos methyl in the amounts described is unclear and of some concern, while the risk of significant long-term effects from chronic exposure to permethrin and carbaryl in these amounts is probably small.

NGOs and relief agencies need to be aware that donated or otherwise provided (e.g., seed provided at little or no cost as part of subsidization program) may be directly consumed, which not only undermines the intent of providing seeds to improve long-term food production but could also adversely impact health. Consumption of pesticide-treated seeds may further undermine the health of already challenged populations, facing the larger health threat of food insecurity. The extent and potential public health impact from this practice is unknown and needs further study to ensure that this food safety issue is not adding to existing food security threats.