Chapter 1: Toxicology of Insecticides—Introductory Considerations

Insecticides are a group of substances with heterogeneous toxicity, whose desired activity is the killing of unwanted insects. Closely allied are acaricides or miticides, terms used for substances that kill mites. Indeed, many acaricides are also insecticides. Many insecticides have mammalian toxicity that is related to their toxicity to the target organism, especially those insecticides that target the insect nervous system. Notable exceptions to this are insecticides that target systems present in insects but not mammals, such as the juvenile hormone analogues and chitin-synthesis inhibitors. Even those insecticides that target systems present in both insects and mammals may have target organism specificity conferred by physical or metabolic differences between insects and mammals.

To be weighed against their mammalian toxicity are the facts that insects and are important sources of agricultural loss of food and other crops such as cotton and can give rise to damage to buildings, where construction is of wood. A very important role of insecticides in public health is vector control. Many insects carry diseases such as malaria, yellow fever, Lyme disease, dengue and sleeping sickness, all of which cause considerable mortality and/or morbidity. Probably the most serious of these is malaria,¹  where the use of insecticides is a very important part of disease control.

As with many uses of chemicals, the key to a successful insecticide is selective toxicity against target insects and away from non-target insects and mammals.^2,3  An ideal insecticide will interfere with a biological system in the insect that has no counterpart in non-target species: this is the advantage of the juvenile hormone analogue and chitin-synthesis inhibiting insecticides. In the case of agricultural insecticides, the insecticide should be toxic to insects, but less toxic to plants, to humans and to other non-target organisms. In the case of insecticides targeting the insect neurological system, insecticides often exploit the relative accessibility of the insect nervous system to xenobiotics, when compared to humans, or different distribution of neurotransmission systems, together with the lack of a nervous system in plants.

In recent years, there has been some concern as to the possibility of deleterious effects from exposure to multiple pesticides, inter alia insecticide exposure, either as residues in food and water or at the workplace, home or environment, or indeed a combination of these pathways; this is further discussed below. Another recent development is the use of microbial insecticides: these are discussed in Chapter 10.

The insecticides can be classified in several ways, for example into those of natural or synthetic origin (see Table 1.1). This division, while perhaps useful for organic farmers, tell us little about toxicology. Insecticides of natural origin—e.g. pyrethrum, nicotine, rotenone (Derris), the ‘mectins—have little in common from the point of view of mammalian toxicology and, of course, the synthetic insecticides can be divided into many groups, depending on their mode of action and/or structure. Indeed some natural insecticides (e.g. pyrethrum) have more in common with synthetic insecticides (the synthetic pyrethroids) than they do with other natural insecticides.

Another way of classifying insecticides is by chemical structure (Table 1.2).

This classification is of much more use to toxicologists, partly because it provides some guide to effects in mammals. However, it is important to remember that this classification is not a complete guide to mammalian toxicology. The organophosphate esters are commonly anticholinesterase insecticides (exceptionally, pyrazophos is a fungicide), but other organic compounds of phosphorus have different actions both in target and non-target species e.g. glyphosate, which is a herbicide. Similarly, most carbamate insecticides are N-methyl carbamates: other carbamates have fungicidal and herbicidal action and are not cholinesterase inhibitors.

The most useful classification for toxicologists is probably by mode of action in the target species and a recent classification on that basis has been proposed by the Insecticide Resistance Action Committee, although it should be noted that the mode of action in target species is not always completely known.⁴  Nevertheless, in many cases where the mode of action in target insects is known, it may provide some guide to the toxicological action in mammals. Thus, many insecticides act on the insect nervous system and their effects in mammals are often related to this. Action on the insect nervous system may be on enzymes involved in neurotransmission (anticholinesterases—see Chapter 4) or directly on receptors (nicotine and the neonicotinoids [see Chapter 6] and fipronil [see Chapters 9 and 12]). Other neuronal structures may be targeted: the avermectins stimulate the release and binding of γ-aminobutyric acid (GABA), a neurotransmitter, at nerve endings (see Chapter 8). Organochlorines such as DDT have a more general action on sodium channels in neurons (see Chapter 3), and pyrethrins and synthetic pyrethroids also act on sodium channels by keeping them open (see Chapter 5). The reason for the utility of this classification is that the mammalian toxicology of these potentially neurotoxic insecticides is in large part related to the neurotoxicity in target organisms. With insecticides that target the insect nervous system, specificity towards insects and away from mammals may be achieved by exploiting the greater accessibility of the insect nervous system, metabolic differences between insects and mammals, or differences in distribution of receptors and other components of neurotransmission systems. Alternatively, specificity can be achieved by differences in binding affinity for receptors or by a combination of means.

Another group of insecticides target structures in insects that are not found in mammals, including the insect growth regulators (juvenile hormone analogues), chitin-synthesis inhibitors and ecdysone agonists. The toxicity of these compounds to mammals is generally low and, strangely, often referable to the mammalian haematological system (see Chapter 7).

Yet another way of classifying insecticides is by their use, in that this may decide how they are regulated (see below).

In the European Union (EU), including the UK, agricultural, horticultural and home garden insecticides are regulated under the Plant Protection Products Directive⁵  and subsequent regulations,⁶  whereas insecticides used in public hygiene and biocides are regulated under the Biocidal Products Directive.⁷  A major change in regulation of plant protection products is the replacement of a largely risk-based regulatory system under the Plant Protection Products Directive by a hazard- and risk-based system. There is little experience of such a system in either the EU or the rest of the world, and it remains to be seen how the new system will work.

Insecticides or acaricides used on animals to control ectoparasites are classified in the EU as veterinary medicines (this is not the case in the USA). Some of these veterinary medicines have a high toxicological profile, as those that are used in food-producing animals can give rise to residues in food in the same way that pesticides used in plant protection can. Furthermore, there has been some concern that ectoparasiticides, used for example on dogs and cats, may give rise to significant exposure of children through the hand-to-mouth pathway. Another consideration is that, because of the scale of their use, some veterinary medicines (e.g. sheep dips) may have a significant environmental impact. In the UK, mostly during the early 1990s, there was concern about the safety of organophosphorus (OP) sheep dips to farmers and farm workers.⁸  In the EU, Directive 65/65⁹  provided the initial basis for the regulation of veterinary pharmaceuticals, and this has been amended many times. Other legislation also impinges on veterinary drug regulation, notably Regulation EC 726/2004,¹⁰  which established the European Medicines Agency (EMEA) in London. Previously called the European Medicines Evaluation Agency, it has retained the acronym EMEA. Also established was the Committee for Medicinal Products for Veterinary Use (CVMP). A few insecticides are used in human medicine, notably malathion for headlouse treatment and trichlorfon (metrifonate), used in tropical medicine: in the EU, these are covered under similar legislation to veterinary drugs, but the relevant committee at the EMEA is the Committee for Proprietary Medicinal Products (CPMP). Another use for insecticides is vector control of insect-borne diseases, notably malaria. In the UK, endemic malaria in East Anglia and north Kent was largely eradicated by the time of the First World War, while in Italy malaria was eradicated after the Second World War and Spain was declared malaria-free in 1964. Insecticide vector control is therefore not a major issue in the EU and certainly not in the UK. However, in other countries where malaria is endemic, mosquito control is important in public health and the use of DDT in this role continues to provoke controversy.¹¹  Nor should it be forgotten that there are many other diseases spread by insects; with them, insecticides make a major contribution to public health. Lice and fleas (wingless insects) and ticks (mites) can all carry disease, some of which are very serious, such as typhus and bubonic plague.

Methods of risk assessment used by authorities that regulate plant protection products, biocides and veterinary medicines have until recently not differed much, but the relatively few insecticides used in human medicine are in general regulated on a risk-benefit basis. Logically, insecticides used in vector control should be regulated in the same way. The other compounds are subjected to risk assessment during the authorization process, on the general assumption that no individual benefit accrues from exposure. The regulatory authorities endeavour to ensure that exposure is below reference doses/health-based guidance values such as the acceptable daily intake (ADI). For insecticides with appreciable acute toxicity another reference dose, the acute reference dose (ARfD) is also elaborated, while for operators there is a third reference dose, the acceptable operator exposure level (AOEL). The ADI is normally calculated from the most sensitive study in the most sensitive animal species, additionally incorporating a safety factor of 100 or more.¹²  The ADI represents a mean maximum intake over time, but to avoid acute toxicity from insecticides, the concept of the ARfD was developed. The calculation of the ARfD also involves a safety factor, most often of 100, from those studies appropriate for assessing acute toxicity.¹³  In the risk assessment process for food exposure the calculated exposure of a ‘high-level’ consumer is compared with the reference doses.

There are a number of possible ways in which humans can be exposed to insecticides. Thus food consumers may be exposed to insecticides used on food crops or food-producing animals, and of course farmers and other applicators may be exposed. Insecticides used domestically in wood preservation, as household insecticides or on pet animals may be important sources of exposure of the general public. Moreover, insecticides may get into the water supply. It should also be remembered that the more acutely toxic insecticides have been used for suicide and murder.

The effects of insecticide residues in food and water probably cause the greatest concern to the general public. In fact, reports of definitively diagnosed clinical poisoning by residues of pesticides in food seems to be comparatively rare, certainly in the EU and USA, and in comparison with deliberate and accidental poisonings and occupational intoxication; nevertheless, cases are reported from time to time (see below). The reasons for the paucity of reports of poisoning from residues are not clear: it may reflect a true rarity of poisoning or a lack of recognition of food-borne poisoning, which might, for example, be mistaken for bacterial food poisoning. Of course, occupational poisonings are far more easily identified, the proximity of cause and effect making diagnosis easier. Acute poisoning, where insecticides had been used in accordance with regulations, seems likely to be a very uncommon occurrence, if it occurs at all, at least in developed countries, simply because of the complexity of the regulatory process and the strictness of enforcement of those regulations. The problem of consumers who consume high levels of some foodstuffs is taken into account when maximum residue levels (MRLs) are pronounced to be toxicologically acceptable by comparison with the appropriate ADI. It is probable that residues of many multiples of the MRL would probably be necessary to produce acute poisoning. Thus it seems highly unlikely that extreme consumption, by itself, could give rise to insecticide poisoning from approved uses. In fact, analysis of reported consumer poisonings by insecticides shows that most reported instances occur from:

• spillage of insecticides on to food during storage or transport

• eating grain or seed potatoes treated with insecticides, where the food article was not intended for human consumption

• improper application of insecticides or failure to observe harvest intervals.^14–16 

The last may have been responsible for clinical poisoning with methamidophos reported in Taiwan by Wu and colleagues.¹⁷  Also, aldicarb, a carbamate anticholinesterase of high acute toxicity, has caused outbreaks of poisoning in the USA and Canada¹⁸  and in Ireland.¹⁹  The insecticides responsible have often been ones with low LD₅₀ (<20mg/kg bw).

For risk assessment purposes, exposure is calculated from field trials, where pesticide residues are measured, and from dietary surveys of food consumption. Also from field trial data an MRL is established to ensure proper usage of the insecticide by the farmer and this MRL must also be toxicologically acceptable. For older pesticides, there may be data from residue surveillance programs showing the frequency and magnitude of residues.

Exposure to insecticides from all sources and simultaneous exposure to mixtures of insecticides has always been a matter of concern. However, risk assessment of pesticides has, until recently, been route-specific and by individual insecticide. This started to change with the enactment in 1996 of the US Food Quality Protection Act.²⁰  This act introduced the terms aggregate risk assessment (assessment of pesticides from all sources of exposure) and cumulative risk assessment (risk assessment of more than one pesticide at a time). The first of these, aggregate risk assessment, is comparatively uncontroversial and simply a matter of data gathering, but the second presents many practical and theoretical problems. The term cumulative risk assessment has been criticised as inviting confusion with pharmacological cumulation, but the use of the term has become widespread and will be used here. Cumulative risk assessment is based upon simple and well-recognised toxicological principles that pesticides with a similar mechanism of toxicological action will act similarly, while pesticides with a different mechanism of toxicological action will act independently. Unfortunately, there are some problems with definitions of combined actions of chemicals, inter alia pesticides. The UK Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) produced a report on toxicity of mixtures of chemicals with special reference to pesticides²¹  and the definitions used in that report are given in Table 1.3.

In cumulative risk assessment the default assumptions are that pesticides with the same toxic action will produce, in combination, simple similar action and will exhibit dose additivity, whereas those with different toxic actions will show simple dissimilar action, which will produce effect additivity and no interaction will occur (no potentiation or antagonism). This means that before doing cumulative risk assessment, it is necessary to take a view on whether compounds possess a similar mechanism of action. That is to say, whether a group of compounds constitute a common mechanism group (CMG). This is easiest with groups such as anticholinesterase OPs when they are all known to act at a single macromolecule (acetylcholinesterase), but much more difficult where compounds have similar effects but possibly by multiple mechanisms, e.g. endocrine disruptors. Even with as homogeneous a group of compounds as the OPs, there are compounds that are atypical in target organism, for example pyrazophos or only have weak anticholinesterase activity (ethephon, tolclofos-methyl). Factors that need to be considered in defining CMGs have been reviewed.^22,23  Carrying out a cumulative risk assessment is easy in theory; it consists of adding the residues of the compounds together having allowed for potency and comparing the result with some figure to toxicological acceptability. In practice there are many problems and more than one way of doing cumulative risk assessment,²⁴  but these are by no means insurmountable. In the USA, much progress has been made, with various groups of compounds having been examined, e.g. OPs,²⁵  N-methyl carbamates²⁶  and some non-insecticidal pesticides. In the EU, the European Food Safety Authority (EFSA) organized a colloquium in 2006.²⁷  The EFSA panel on plant protection products and their residues (PPR panel) has proposed a tiered approach to cumulative risk assessment.²⁸  One concern addressed by the COT was concern about the ‘cocktail effect’: that is, the possibility that harmful effects that may be produced by exposure to food residues of many different pesticides and other chemicals, not necessarily toxicologically similar.²¹  Further, there is concern in Europe that exposure to chemicals from a variety of sources, even at low doses, may be deleterious. There is no clear evidence of the occurrence of such effects from pesticides and the importance of any such combined actions in humans is unclear. In fact it may simply be that the question of the existence of the ‘cocktail effect’ is unanswerable.

Toxicology is a continually developing science, and a number of changes have occurred over the last few years. A major change in the regulation of plant protection products has been discussed above, in relation to hazard and risk. Two other developments are

• in hazard identification—developmental neurotoxicity

• in exposure assessment—probabilistic exposure assessment.

Developmental neurotoxicity is a major public health problem with neuroactive substances, notably ethanol,²⁹  but also some other substances.³⁰  Since many insecticides are neurotoxic, the possibility has had to be considered that they might be developmental neurotoxicants. A major meeting was held at Williamsburg, Virginia, USA in 1989 to consider how this problem might be approached.³¹  While many aspects of developmental neurotoxicity are undoubtedly tested in the conventional pesticide data package, there was concern that some aspects might be being missed. Consequently the United States Environmental Protection Agency (EPA) elaborated a guideline for a test of developmental neurotoxicity in the rat.³²  There is also an Organisation for Economic Co-operation and Development (OECD) guideline.³³  These guidelines describe a test that is not mechanism based but rather intended to detect as broad a range of developmental neurotoxicants as possible.

The test substance is normally administered orally to several groups of pregnant animals during gestation and lactation (day 6 of gestation to postnatal day 10) and offspring are selected for evaluation of neurotoxicity. The end points that are evaluated are gross neurological and behavioural abnormality, motor activity, auditory startle response and assessment of learning. At termination, the offspring brains are weighed and neuropathology is carried out (normally day 60 after birth). The EPA is evaluating these studies and accumulating a database.^34,35  Developmental neurotoxicity tests (DNTs) done for the EPA have been submitted to European regulatory authorities as part of packages when available even when nothing adverse occurred, but there is no general requirement to do DNTs for the EU (see Hass, 2003).³⁶ 

This is sometimes incorrectly called probabilistic risk assessment (no serious attempt has been made to apply probabilistic methodology to toxicology studies). Probabilistic exposure assessment uses distributions of data, e.g. food item intake, pesticide residue level and other factors, to calculate the distribution of pesticide residue intake.^37,38  There is evidence that it gives a better guide to pesticide intake than deterministic exposure calculations and its use becomes becomes essential when cumulative risk assessment is carried out, to avoid ‘compounded conservatism’.³⁹ 

It must be recognised that insects are important sources of agricultural loss and can damage property, for example where construction is of wood. Also, as discussed above, insects are vectors of disease in humans and other animals and failure to control disease vectors may render land unusable for agriculture or even uninhabitable. Weighed against the disadvantages of insecticides that accrue from their toxic effects on non-target species, including humans.

In addition to chemical names, insecticides have national common names (e.g. British Standards Institute [BSI], American National Standards Institute [ANSI]) as well as ISO (International Organization for Standardization) common names.^40,41  The list of ISO names is continually expanding and is updated regularly.⁴²  ISO names are allocated in English, French and Russian, and the English-language ISO name is not always the same as the BSI or ANSI common name. An example is iodofenphos, which is the BSI name, where the English-language ISO name is jodfenphos (curiously, the French language ISO name is iodofenphos!).

A few insecticides are used as drugs either in human or veterinary medicine. As such they have international non-proprietary names (INNs), as well as United States Adopted Names (USANs) and British Approved Names (BANs). INNs are bestowed in a number of languages (English, Latin, French, Russian, Spanish, Arabic, Chinese, and Latin) by the World Health Organization.⁴³  In some cases the INN differs from the ISO pesticide name: for example, trichlorfon, an insecticide, is the same as metrifonate, a pharmaceutical used in human medicine. Insecticides are commonly sold under trade names given by their manufacturers. It is customary to capitalize the initial letter of trade names but not of ISO or other approved names.

It is very important that in tests on pesticides the product used is identified precisely. Insecticides will have a technical specification, which gives the minimum content of the pure active ingredient and the identity and maximum content of impurities present at or greater than a certain level (normally 0.1% of the total content of the active substance). Pesticides are sold as a formulation which comprises the active substance and other substances. The formulation recipe consists of the nominal target content for the pure active substance, with acceptable tolerance limits, and the name(s) and quantity of all other components.⁴⁴  It should be noted that substances other than the active ingredient may contribute to acute toxicity: an example is the contamination of diazinon, an OP, with sulfotep, a more acutely toxic anticholinesterase.⁴⁵  Sometimes impurities will produce qualitatively different toxicity from the active substance; thus several trialkylphosphorothioates, contaminants of OP insecticides, produce lung injury.^46,47