Safety to humans

• In their Recommended Classification of Pesticides by Hazard, the World Health Organization (WHO, 2009) classified paraquat as “Moderately hazardous, class II”
• In their RED facts document (1997), the US Environmental Protection Agency (EPA) concluded that “paraquat is moderately toxic (Category II) by the oral route”. This category was confirmed in a US EPA risk assessment of 2006. EPA’s website states “Paraquat is highly toxic to humans”.
• The European Chemicals Bureau (2004) classified paraquat as “R25, Toxic if swallowed”. According to the current EU regulation 1272/2008, the classification is now Category 3 (H301, "Toxic if swallowed").

• In a US EPA risk assessment of 2006, US EPA classified paraquat in Category I.
• The European Chemicals Bureau (2004) classified paraquat as “R24, Toxic in contact with skin”. According to the current EU regulation 1272/2008, the classification is now Category 3 (H311, "Toxic in contact with skin").

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO considered paraquat “to be a mild skin irritant and a moderate eye irritant”.
• In a US EPA risk assessment of 2006, US EPA classified paraquat in Category I (eye) and Category III (skin).
• The European Chemicals Bureau (2004) classified paraquat as “R36/37/38, irritating to eyes, respiratory system and skin”. According to the current EU regulation 1272/2008, the classification is now Category 2 (H315, "causes skin irritation") for skin irritation, Category 2 (H319, "causes serious eye irritation") for eye irritation and Category 3 for respiratory irritation (STOT SE 3 - H335, "specific organ toxicity single exposure – may cause respiratory irritation").

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO considered paraquat “was not a skin sensitizer in the Magnusson and Kligman test”.
• In their RED facts document (1997) the US EPA concluded that paraquat was “not a skin sensitizer under the conditions of the maximization test of B Magnusson and AM Kligman”. This conclusion was confirmed in a US EPA risk assessment of 2006.
• In their review report for paraquat, the European Commission (EC, 2003) concluded that paraquat was “negative in Magnusson & Kligman protocol”. According to the criteria used in the current EU regulation 1272/2008, paraquat is not classified as a skin sensitizer.

• In their RED facts document, the US EPA (1997) concluded that “paraquat has been shown to be highly toxic by the inhalation route and has been placed in Toxicity Category I (the highest of four levels) for acute inhalation effects. However, the Agency has determined that particles used in agricultural practices (400 to 800μm) are well beyond the respirable range and therefore inhalation toxicity is not a toxicological endpoint of concern.”
• The European Chemicals Bureau (2004) classified paraquat as “R26, Very toxic by inhalation”. According to the current EU regulation 1272/2008, the classification is now Category 2 (H330, "Fatal if inhaled").
• See below for information about the ability to inhale paraquat into the lungs.

Paraquat dichloride is non-volatile and formulations containing paraquat are not applied through spray equipment which generates a significant proportion of respirable spray droplets at the spray nozzle. Paraquat is applied downwards towards the target plant with conventional hydraulic spray equipment for which the droplets produced are generally too large to be inhaled, being typically ~ 200-400μm diameter. Only a small fraction of the total spray droplet distribution leaving the spray nozzle is of respirable size with <0.2% less than 10μm. The inhalable fraction <30μm that is taken into the nose and the mouth during breathing is also too large to reach the alveolar region of the lung. This fraction could, however, be swallowed and is therefore a source of secondary oral ingestion. Mist blowers, which produce large numbers of very fine droplets (50μm and less), are not recommended for spraying paraquat.

• In their RED facts document, the US EPA (1997) concluded that for paraquat “particles used in agricultural practices (400 to 800μm) are well beyond the respirable range and therefore inhalation toxicity is not a toxicological endpoint of concern”.

Specific studies to determine absorption of paraquat after oral administration (Syngenta, unpublished) revealed an absorption factor of 13%. Despite the low absorption rate, oral ingestion of paraquat containing products must be strictly avoided.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO considered paraquat to be “not well-absorbed”
• In their RED facts document (1997), the US EPA concluded that “paraquat was poorly absorbed after oral administration to rats, dogs and mice”
• In their review report for paraquat, the EC (2003) concluded that paraquat was poorly absorbed with “approximately 10% absorption”

Undamaged skin is an effective barrier to paraquat in concentrated or diluted spray-strength solution. Paraquat is water soluble and easily washed off. Prolonged and repeated contact of spray-strength paraquat with skin, through leaking spray equipment or poor personal hygiene can cause skin irritation and even damage in severe cases. These are visible warnings to the user that he or she is doing something wrong and they appear sufficiently early to avoid any further health risk.

• In their Reregistration Eligibility Document, RED (1997), the US EPA (1997) concluded that “0.3% of the applied 14C-paraquat dichloride was absorbed through intact skin”
• In their review report for paraquat the EC (2003) concluded that the dermal absorption of paraquat was “0.5% based on overall weight of evidence.”

Any absorbed paraquat is generally excreted unchanged with little or no metabolism.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO considered “paraquat is largely eliminated unchanged”.
• In their Reregistration Eligibility Decision Document (RED) (1997), the US EPA concluded that “paraquat was not metabolized by rats”.
• In their review report for paraquat the EC (2003) concluded that paraquat was subject to “minimal metabolism, representing <1% of recovery”.

Paraquat is not stored or accumulated in the body. In occupational circumstances in which very low amounts of the spray-strength product may be absorbed, it is rapidly and effectively eliminated via the urine. In laboratory studies it has been shown that low amounts of absorbed paraquat are rapidly excreted.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO noted that for paraquat “excretion of the radiolabel was rapid; about 90% was excreted within 72h”.
• In their Reregistration Eligibility Decision Document (RED) (1997), the US EPA concluded that for paraquat “most of the radioactivity was detected in faeces with 2-3 days after dosing and in urine, within 1 day after dosing”.
• In their review report for paraquat, the EC (2003) concluded that the rate of excretion of paraquat was “>90% in 72h”.

Paraquat has been evaluated for genotoxic potential in a range of in vitro and in vivo test systems. These studies provide a substantial body of data to indicate that paraquat does not have significant genotoxic potential. Paraquat has produced positive responses in some in vitro assays. These are considered not to be due to paraquat itself, but to the action of paraquat within cellular systems. Paraquat is not genotoxic in higher tier in vivo assays.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO concluded that “paraquat is unlikely to pose a genotoxic risk to humans”.
• In their RED facts document, the US EPA (1997) concluded that “paraquat shows no evidence of causing mutagenicity”.
• In their review report for paraquat the EC (2003) concluded that genotoxicity studies with paraquat were “negative in vivo. Some in vitro positives”.

• In their summary report for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “The weight-of-evidence indicates that paraquat is non-mutagenic and therefore is not considered to pose a significant genotoxic risk to humans.”

The chronic toxicity and potential carcinogenicity of paraquat have been examined in two-year studies in rats and mice. Effects on the lung were the principal effects noted in the chronic feeding studies, which is consistent with the findings from sub-chronic toxicity studies.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO concluded that “the weight of evidence suggested that paraquat was not carcinogenic in the rat. Paraquat was not considered to be tumorigenic in two studies in mice”.
• In their Reregistration Eligibility Decision Document (RED) (1997), the US EPA “classified paraquat as a Group E carcinogen (evidence of non-carcinogenicity for humans), based on a lack of evidence of carcinogenicity in acceptable studies with two animal species”.
• In their review report for paraquat the EC (2003) concluded that paraquat was “not carcinogenic”.
• In their summary report for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “Long-term feeding studies conducted in mice and rats revealed no evidence that paraquat was carcinogenic.”

No teratogenic potential has been demonstrated in developmental toxicity studies in rats and mice.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO concluded that “teratogenicity was not seen at any dose in any study in either rats or mice”.
• In their RED facts document, the US EPA (1997) noted that “the no-observed effect dose levels (NOEL) for maternal toxicity are at least or more conservative (protective) than the NOEL based on developmental toxicity”.
• In their review report for paraquat, the EC (2003) concluded that paraquat was only “embryotoxic at maternally toxic doses”.
• In their summary report for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “There was no evidence that paraquat had any teratogenic potential based on a range of studies conducted in mice, rats and rabbits.”

Paraquat has been demonstrated not to interfere with reproduction in a three-generation reproduction study.

• In their evaluation of toxicity for the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2003), WHO concluded that “impaired fertility was not seen in these studies”.
• In their RED facts document, the US EPA (1997) noted that “there is no evidence that paraquat is associated with reproductive effects”.
• In their review report for paraquat, the EC (2003) concluded that reproductive toxicity studies with paraquat demonstrated “no specific effects on reproduction”.
• In their summary report for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “There was no evidence that paraquat caused reproductive toxicity in rats following dietary administration.”

The available data for paraquat have been reviewed and are considered not to show evidence of endocrine disruption in: a) Mammalian studies: including rat reproduction (multi-generation); rat and mouse developmental (teratology); one-year dog, lifetime rat and mouse; b) Environmental species: including a life-cycle study in Daphnia and reproduction studies in the bobwhite quail and mallard.

• In its report, “Towards the establishment of a priority list of substances for further evaluation of their role in endocrine disruption” for the European Commission DG ENV, BKH Consulting Engineers lists paraquat as a group III substance, which means there is no scientific basis for inclusion on the endocrine disruptor list.

The neurotoxic potential of paraquat has been extensively studied in laboratory animals. No clinical signs of neurotoxicity or consistent neuropathological changes have been reported following long-term dietary administration of paraquat to rodents or dogs in regulatory-compliant studies.

• In their report of a periodic re-evaluation of toxicity for the JMPR (2003) (paraquat), WHO concluded that “studies on the effects of paraquat on the central nervous system have used a variety of routes, including subcutaneous or intraperitoneal injection and direct injection into the central nervous system, and end points observed have been behavioral, morphological and neurochemical. Behavioral effects and loss of neurons in the substantia nigra were observed and, neurochemically, depletion of dopamine was reported in many, but not all, of these studies. However, the design of these studies renders the relevance of these data questionable for the risk assessment of dietary exposure to paraquat residues.” It was then concluded that “the available mechanistic and other animal studies did not support the hypothesis that paraquat residues in food are a risk factor for Parkinson’s disease in humans”
• In 2006 US EPA stated “Paraquat does not inhibit cholinesterase activity, produce cholinergic-like toxic signs, or affect morphology of the central or peripheral nervous systems. Furthermore, the molecular structure is not similar to those of compound classes known to have effects on the nervous system. Clinical signs of toxicity did not indicate neurotoxicity. As there is limited concern for neurotoxicity resulting from exposure to paraquat, neurotoxicity studies, including a developmental neurotoxicity study, are not required at this time.”
• Paraquat did not result in neurotoxic effects in modern US EPA guideline-compliant oral acute and sub-chronic (90 day) neurotoxicity studies reported in 2006.
• In their summary report for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “The standard toxicological database contains a range of unpublished acute, short-term repeat-dose, subchronic and chronic laboratory studies which do not report any evidence of neurotoxicity in rodents, rabbits or dogs, including any clinical signs (consistent with a proposed neurotoxicant) or neuropathy under the standard suite of pathology tests.”
• In the last decade animal studies have focussed on the entry of paraquat into the brain of mice and the potential to cause neuronal cell death in the area of the brain known as the substantia nigra. The main reason for this is the putative link between pesticides and Parkinson’s disease. Please refer to the section below: Paraquat does not cause Parkinson's disease.

Source:

US Environmental Protection Agency, Office of Prevention, Pesticides and toxic substances. 2006. Memorandum: Risk Assessment. Subject: Paraquat Dichloride: Human Health Risk Assessment for Proposed Uses on Ginger and Okra and Amended Uses on Soybeans, Wheat, Cotton, Cucurbits, Onions, and Tanier. PC Code: 061601, Petition Nos: 2F6433, 3E6764, 1E6223, 1E6332, 3E6763, and 1E6319, DP Barcode: D328653.

The available scientific evidence does not support the existence of a causal relationship between Parkinson’s disease and pesticides in general or paraquat in particular. Parkinson’s disease is overwhelmingly associated with old age and it is generally agreed that the development of Parkinson's disease involves both environmental and genetic factors (Obeso et. al., (2010). Missing pieces in the Parkinson’s disease puzzle. Nature Medicine, 16-6, 653-661). The relative contribution of these and other factors is still the subject of considerable scientific debate as evidenced by the significant number of scientific papers examining the association of pesticides, including paraquat, with the development of Parkinson's disease using studies in experimental animals and epidemiological investigations.

An analysis by independent senior medical and epidemiological scientists under the guidance of Professor Sir Colin Berry (Emeritus Professor at Queen Mary, University of London, and former member of the UK Medical Research Council) and Professor Pierluigi Nicotera (Founding Director of the German Center for Neurodegenerative Diseases, Bonn) found no causal link between exposure to paraquat and Parkinson’s disease (Berry, C., La Vecchia, C. & Nicotera, P., (2010). Paraquat and Parkinson’s Disease. Cell Death and Differentiation, 17, 1115–1125). This publication builds on the previous Consensus Statement on Parkinson’s Disease and the Environment in which 29 independent experts concluded that the evidence to date does not permit a conclusion to be drawn regarding the existence of an association between exposure to paraquat and the development of Parkinson’s disease (Bronstein et al., (2008). Consensus Statement Parkinson’s Disease and the Environment: Collaborative on Health and the Environment and Parkinson’s Action Network (CHE PAN) Conference June 26-28, 2007. Environmental Health Perspectives, 117:117-121).

Mandel et al. (2012) reviewed and evaluated two epidemiologic studies (Kamel et al, 2002 and Tanner et al., 2011) focused on pesticides, and in particular, paraquat as a cause of Parkinson's disease. Both studies are derived primarily from the US Agricultural Health Study (AHS). The review concluded that these and other epidemiological studies which they also reviewed were inadequately designed and often underpowered with very few exposed individuals. The studies were not population-based, failed to distinguish incident from prevalent cases (Tanner et al., 2011), relied on multiple comparisons, and may have reported results selectively. They also noted that the results across the studies are inconsistent. In summary, a conclusion regarding these relationships cannot be reached based on the current literature and further research with higher methodological standards is needed to reach a definitive conclusion.

Reference: Mandel JS, Adami HO, Cole P (2012). Paraquat and Parkinson's disease: an overview of the epidemiology and a review of two recent studies. Regul Toxicol Pharmacol. 2012 Mar;62(2):385-92. doi: 10.1016/j.yrtph.2011.10.004. Epub 2011 Oct 15.

Assessments have shown that there is no evidence of an increased incidence of Parkinson's disease, relative to that expected in the local and national population, either as the underlying cause of death or as a mentioned cause of death among workers involved in the manufacture of paraquat. Assessments have also shown there is no evidence of parkinsonism among survivors who ingested paraquat for the purpose of self-harm. Specifically:

• A cohort epidemiological assessment of workers involved in the manufacture of paraquat between 1961 and 1995 has evaluated whether there is any evidence, from analysis of death certificates, of increased incidence of Parkinson's disease as the underlying cause of death or as a mentioned cause of death among those with long-term exposure to paraquat.
• This work concluded that there was no evidence of an increased incidence of Parkinson's disease among paraquat production workers based on mentions of Parkinson's disease on the death certificates of workers who had died. A strength of this study is the likely higher exposure of workers engaged in paraquat production than many of the subjects in case-control studies classified as exposed to paraquat.

Reference: Tomenson J.A., Campbell C (2011). Mortality from Parkinson’s disease and other causes among a workforce manufacturing paraquat: a retrospective cohort study. BMJ Open 2011;2:e000283. doi:10. 1136/bmjopen-2011-000283.

• A medical toxicologist has conducted a health outcome assessment of survivors who intentionally ingested paraquat for the purpose of self-harm to evaluate whether there is any evidence of increased incidence of Parkinson's disease-like symptoms (i.e.  parkinsonism) among those with acute, high dose exposure to paraquat.
• This analysis found no connection between confirmed high-dose paraquat exposure in humans and the development of parkinsonism.

Reference: Brent J., Schaeffer T.H (2011). Systematic Review of Parkinsonian Syndromes in Short- and Long-Term Survivors of Paraquat Poisoning. Journal of Occupational and Environmental Medicine, 53 (11): 1332-1336.

According to current scientific knowledge, Parkinson's disease does not develop naturally in any animals except humans. Rats and mice are widely used for modeling Parkinson's disease. Traditionally, animal models of Parkinson's disease rely on the use of specific toxins (most often MPTP), which selectively accumulate in the dopaminergic neurons, causing cellular dysfunction and death. A specific strain of mouse (C57BL/6), treated intraperitoneally with paraquat, has been used to model the cellular losses from the brain region (substantia nigra) seen in Parkinson's disease. The reliance on this model as evidence of the human condition is controversial and has been challenged in the scientific literature.

Several manuscripts have shown that paraquat, when administered systemically to C57BL/6 mice, can induce dopaminergic cell loss in the substantia nigra. However, the dose level at which this occurs is unclear and paradoxically does not appear to be accompanied by cell loss as evidenced by validated corresponding pathology markers.

A 2013 publication reported a series of studies in which paraquat was administered by intraperitoneal injection to C57BL/6J mice. Only 0.3% of the dose entered the brain and it was then slowly eliminated. In this study series paraquat did not cause pathological or neurochemical changes in the brain when given in doses up to and including 3 x 25 mg paraquat dichloride/kg. These studies did not consistently detect a significant dopaminergic cell loss, as reported by others, neither was there any evidence of neuronal cell death in the relevant region of the brain (SNpc), nor any effect on neurotransmitter levels in the striatum. Additional studies are required to resolve the differences between these results and those reported by others, including the in vivo studies by Peng et al. (2004).

Reference: Breckenridge C B et al., 2013 Pharmacokinetic, neurochemical, stereological and neuropathological studies on the potential effects of paraquat in the substantia nigra pars compacta and striatum of male C57BL/6J mice. NeuroToxicology, 37, 1–14

The results of this study have subsequently been confirmed in a comprehensive inter-laboratory investigation in which paraquat administered at maximum tolerated doses did not result in any neuropathology (Smeyne et al., 2016). This research employed three distinct methods of assessment, performed by three groups of investigators who were all ‘blinded’ to treatment, for a series of neuropathological indices evaluated in two ages and two sub-strains of male C57BL/6 mice, housed under different conditions in two laboratories. Paraquat, administered either once or twice weekly to 9- or 16-week old mice from two suppliers, had no effect on the number of dopaminergic neurons in the SNpc, as assessed by two groups, each blinded to treatment, using different stereological methods. In addition, blinded assessments conducted independently by a neuropathologist and a stereologist did not find any evidence of neuronal cell death (neuropathology) or neuroinflammation (neuropathology and stereology). Based on the scope and comprehensiveness of this study, the authors concluded that “… it is implausible that DA neurons could have died without detection in PQ-treated mice”.

Reference: Smeyne, R.J., Breckenridge, C.B., Beck, M., Jiao, Y, Butt, M., Wolf, J.C, Zadory, D., Minnema, D, Sturgess, N.C., Cook, A.R., and Botham, P., (2016). Assessment of the effects of MPTP and paraquat on dopaminergic neurons and microglia in the substantia nigra pars compacta based upon 2-D and 3-D stereological methods. PLOS ONE, DOI:10.1371/journal.pone.o164094, 2016 October 27

• In the Neurotoxicity supplement for paraquat, the Australian Agricultural and Veterinary Chemical Assessment Team Office of Chemical Safety, APVMA (2016) concluded that “the overwhelming weight-of-evidence, taking into consideration the available database of studies on experimental animals, human epidemiology studies and poisoning cases, is that paraquat does not induce neurotoxicity via the oral, dermal or intranasal routes; routes that are of relevance to human exposure to this pesticide. The OCS is therefore satisfied that exposure to paraquat, when used as a herbicide, does not pose a risk of neurotoxicity in humans and that lung damage remains the most sensitive and critical toxicological endpoint of concern in establishing health-based guidance values.”