Pesticide toxicity
The large variety of chemical compounds that show pesticide properties means that there is a very wide range of toxicity in humans. It is believed that an oral dose of only several drops (100mg) of terbufos, an OP compound, is fatal to most adults, whereas another pesticide (amitrole) is nontoxic in humans even when several hundred grams are ingested. Even within a particular class of pesticide the lethal dose may vary considerably. Moreover, the metabolites of many pesticides (e.g. oxygen analogues of phosphorothionates) are much more toxic than the parent compounds. Commercially available preparations usually contain an active substance mixed with filler (solids) or dissolved in an organic solvent (liquids). Although certain pesticides are unlikely to cause acute toxicity, the vehicle in which they are formulated (toluene, xylenes, butan-1-ol, cyclohexanone, farbasol and solvent naphtha) may itself be toxic and, in some cases, can be the main causative agent for the symptoms observed. This needs to be taken into consideration in clinical treatment and in forensic toxicology both in terms of the analytical methods applied to samples and the interpretation of results.
The World Health Organization (WHO) has classified pesticides into five groups based on their hazard. In the WHO classification, hazard is considered as ‘the acute risk to health (that is, the risk of single or multiple exposures over a relatively short period of time) that might be encountered accidentally by any person handling the product in accordance with the directions for handling by the manufacturer or in accordance with the rules laid down for storage and transportation by competent international bodies’ (the WHO Recommended Classification of Pesticides by Hazard and Guide lines to Classification 2006). Classification is based on acute oral and dermal toxicity to the rat (LD50) and the estimated lethal doses related to a 70 kg person (see Table 4.6). It must be borne in mind that extrapolation of toxicity values from a test animal such as a rat to humans is a best estimate and may carry a high error factor for some substances. Realistic human lethal doses of pesticides can be estimated only on the basis of well-documented cases of poisoning.
The immense variety of chemical compounds with pesticidal properties means that the identification of an unknown substance is complex, particularly where no information is available about the likely identity of the pesticide. In clinical and forensic toxicology, unless specific information is available indicating that a particular pesticide should be targeted in analysis (e.g. a body found with a labelled pesticide product alongside), a broader screening procedure should be employed to identify active pesticide compo nents. Some colour tests can be very useful preliminary indicators of the class of compound and can confirm the constituents of a proprietary formulation. The ammonium molybdate test is used for phosphorus and phosphides in stomach contents and nonbiological materials (Flanagan et al. 1995). The furfuraldehyde test is used for carbamates in the same matrices (Flanagan et al. 1995). The phosphorus test can be used to detect OP compounds; although the limited sensitivity of the test means that it is not able to detect OP compounds in blood and the sodium dithionite test is used for diquat and paraquat (Tompsett 1970). While colour tests are useful to indicate preliminary classes, they are generally restricted to stomach contents and nonbiological materials and will only detect a rather restricted range of pesticides. Thin layer chromatography (TLC), GC and liquid chromatography (LC) can be used to screen for a far wider range of substances. TLC is a very useful screening and identification tech nique for pesticides in commercial preparations added to beverages or foodstuffs and in body fluids (stomach contents, urine) and tissues. Visualisation reagents are available which produce a variety of colours to facilitate differentiation. A large number of pesticides react with more than one reagent. For example, silver nitrate in solution or rhodamine in ethanol followed by an overspray of sodium hydroxide in ethanol can be used to visualise carbamates, chlorinated hydrocarbons, chlorinated phenoxy acids, OP compounds, pyrethroids, substituted ureas and triazines. Some reagents are consider ably more specific and can be used to confirm a particular pesticide class. OP compounds can be visualised using 4-(4-nitrobenzyl) pyridine and tetraethylene pentamine.
TLC has limitations in terms of sensitivity and restrictions in terms of the matrices to which it can be applied. The method of choice, regarded as the reference method for identification and confirmation of the presence of pesticide(s) in different materials, is GC-MS. Comprehensive libraries that contain reference spectra for many pesticides, their metabolites and decomposition products are available for use with GC-MS. The most important libraries are the Pfleger, Maurer & Weber Library and the National Institute of Standards and Technology (NIST) Library. GC and GC-MS methods need special sample preparation procedures for biological materials. A broad spectrum of pesticides in heterogeneous matrices (nonbiological and biological samples) may be submitted for toxicological analysis. Biological sample preparation may involve cleavage of conjugates, isolation from the matrix, clean-up steps and/or derivatisation of the pesticides and/or their metabolites. Pesticides can be isolated either by liquid–liquid extraction (LLE) or solid-phase extraction (SPE). LLE is still regarded as the more universal method for screening, whereas SPE is preferred for quantification of certain pesticides in blood samples or for extraction of a particular chemical class of pesticides, such as coumarin anticoagulants.
Each step in GC-MS analysis for pesticides can be a source of artefacts or decomposition prod ucts as a result of light sensitivity, atmospheric oxidation, hydrolysis and heat. The major metabolites of many pesticides (e.g. carbamates and organophosphates) are sulfate and glucuronide conjugates. Cleavage of conjugates by enzymatic or acid hydrolysis is necessary before extraction. However, deconjugation of pesticides by acid hydrolysis drastically increases the formation of artefacts and can destroy analytes completely. Therefore, the gentle enzymatic method is recommended. The extraction of body fluids is further complicated because certain pesticides are decomposed readily by acids or alkalis. Moreover, the decomposition products of some subclasses of pesticides (e.g. substituted ureas) can react with the extraction solvent (e.g. ethanol or acetone) used for reconstitution of the dried residue after extraction. Many pesticides can undergo degradation during the chromatographic process. Therefore, the possibility of formation of artefacts during all these processes must always be taken into account and minimised wherever possible. Derivatisation of pesticides can often improve chromatographic separation. Moreover, in some cases, chromatography of the extract before and after derivatisation increases the identifying power of GC, and for GC-MS methods changes in the fragmentation pattern can yield additional information for characterisation. It is very important to use good-quality reference materials, although in some situations, such as when using TLC screening methods, commercial preparations can be an adequate substitute. As in virtually all analyses, to reduce false positives from artefactual sources, a blank solution should be subjected to the same procedure as the sample. It is also essential to check the viability of the reagents by analysing a reference compound. Pesticide standards are typically available prepared in sealed ampoules. Dilute solutions of many pesticides will not remain stable for more than a few months after opening the ampoules.
The quantity of the sample taken for extraction depends on its type and is restricted by its availability. Samples may be proprietary formulations (solid or liquid in an amount of several drops to several millilitres or grams), beverages or foodstuffs, soil samples (20 g), river or lake water (up to 500 mL) and body fluids – stomach contents, urine (5 mL), blood (1–2 mL) or tissues (5 g). High-performance liquid chromatography (HPLC) and combined liquid chromatography mass spectrometry (LC-MS) are also used as confirmation methods. A review by Alder et al. (2006) compared the residue analysis of 500 priority pesticides by GC-MS and LC-MS/MS. They concluded that LC-MS/MS using electro spray ionisation (ESI) offers better sensitivity than GC-MS for all pesticide classes considered with the exception of the organochlorine pesticides, and that LC-MS can overcome problems associated with thermal instability seen for many pesticides in heated GC injectors. It should be borne in mind that the study was based on comparison of literature data and that these data were taken from the environmental and food arenas, not necessarily toxicology. Nonetheless, it does indicate the potential for LC-MS/MS in pesticide analysis.
Although not available for as wide a variety of pesticides in human biological matrices as GC MS, LC-MS and LC-MS/MS methods are being developed. There are certainly numerous LC-MS methods available for analysis of most pesticide classes in foodstuffs and it should be possible to adapt these methods for toxicological purposes with suitable sample clean-up. As noted in Chapter 8, in clinical treatment, if pesticide poisoning is suspected a test for cholinesterase inhibitors should be carried out as this may help to identify the presence of OP and/or carbamate compounds. Initial screening procedures help identify individual pesticides but are not geared to quantifying the levels that are present. Measurements of pesticide concentration in biological samples from acutely poisoned patients can have an immediate bearing on treatment, particularly when active elimination procedures, such as diuresis or haemodialysis, are contemplated. In fatal cases of suicidal or homicidal poisoning, it is easier to detect and identify pesticides by examining suspect materials, such as liquids, food, clothing and soil, in which the concentrations are likely to be quite high. Thereafter, if there is strong evidence that the substance detected may have been responsible for the death, quantitative examination of postmortem tissues may provide conclusive proof of poisoning.
Quantitative analyses also have an important role in monitoring pesticide concentrations in soil, water supplies, rivers, lakes and foodstuffs; in some countries, legislative control of permis sible levels has been introduced. Their applica tion is likely to increase with the growing public concern about the release of pesticides into the environment. Many guidelines for quality assurance and method validation have been developed by scientific organisations, and several national and international proficiency-testing schemes are now in operation. Numerous analytical procedures for the quantitative analysis of pesticides in various media are described in the National Institute for Occupational Safety and Health (NIOSH) Pocket Guide to Chemical Hazards. Another recent publication contains original and sensitive GC MS and LC-MS methods using standardised sample-preparation procedures for the detection and quantification in human biological matrices of 61 pesticides of toxicological significance (Lacassie et al. 2001). As noted above, Alder et al. (2006) in their review of GC-MS versus LC-MS methods for pesticides conclude that LC-ESI MS/MS has considerable advantages over GC-MS for analysis of pesticides except for chlorinated hydrocarbons.
Organophosphorus compounds OP compounds are by far the most important class of pesticides, both in terms of worldwide usage and their toxicity to humans. They act by the irreversible inhibition of cholinesterases, which are responsible for hydrolysing, and thereby deactivating, the neurotransmitter acetylcholine (AcCh). Build-up of AcCh at the neural junction leaves the muscles, glands and nerves in a constant state of stimulation, which produces a wide range of acute symptoms. These include dizziness, confusion and blurred vision, excessive salivation and sweating, nausea and vomiting, and muscular weakness. Severe poisoning leads to coma, flaccid paralysis, breathing difficulties, cyanosis and cardiac arrhythmias. Atropine and pralidoxime are effective antidotes in severe cases. In acute clinical poisoning, diagnostic tests for depressed cholinesterase activity are most crucial. Detecting, identifying and quantifying the particular agent responsible has less bearing on immediate treatment, although some of the lipophilic diethyl phosphothiolates can be sequestered in the tissues for several days and patients who appear to have recovered may suffer a recurrence of toxic effects. Identification of the agent involved can alert clinicians to this possibility.
Two types of cholinesterases exist in the body. Acetylcholinesterase (AChE), which is also known as true cholinesterase, is found in red cells, nerve endings, lungs and brain tissues. Its main function is to hydrolyse AcCh at cholinergic nerve endings. The second type is usually known as pseudocholinesterase (ChE) and occurs in the plasma in addition to other body tissues. The exact physiological function of ChE is unknown, but it has the ability to hydrolyse a variety of esters in addition to cholinesterase. Depression of ChE can also be caused by non-pesticide chemicals, liver diseases and other factors (physiological, pharmacological or genetic). Measurement of red-cell AChE is therefore a more specific indicator of cholinesterase inhibition caused by OP or carbamate pesticides. Moreover, the repression of red-cell AChE activity can be demonstrated for up to 2–6 weeks after exposure, whereas that of plasma ChE returns to normal much more quickly. Neverthe less, in practice, plasma ChE activity is a useful indicator of exposure, since if normal values are found this effectively excludes acute poisoning by these substances. Some carbamate herbicides and fungicides, such as the dithiocarbamates, do not inhibit cholinesterases to any significant degree and are relatively nontoxic in humans. Postmortem specimens for AChE assay must be kept in cold storage and analysed as soon as possible to minimise the effects of spontaneous reactivation of the enzyme.
Carbamates In terms of toxicity, carbamate pesticides have a similar action to that of the OP compounds in causing a decrease in cholinesterase activity, but the binding to the active site of the cholinesterase enzyme is reversible. Conse quently, although the symptoms are practically identical to those of OP poisoning, they have a shorter duration. Carbamates can be divided into various subclasses, characterised by their different thermal stabilities. N-Methylcarbamates give thermal decomposition products, mainly substi tuted phenols. When analysed by GC-MS, these products give rise to mass spectra with abundant molecular ions. The compounds from other subclasses of carbamates are more thermally stable. GC-MS analysis of these more stable compounds results in mass spectra where the molecular ions are of low intensity but, together with diagnostic fragments, enable identification to be made. In LC-MS methods, the carbamates do not present a serious problem in terms of analysis. Positive-ion detection with a soft ionisation technique is the method of choice (Niessen 1999). Lacassie et al. (2001) have reviewed methods of analysis of various classes of pesticides for use in clinical and forensic toxicology, including LC-MS methods for carbamates.
Chlorinated hydrocarbons Chlorinated hydrocarbons are neurotoxins that also damage the liver and kidneys. Major clinical features of poisoning are headache, disorientation, paraesthesia and convulsions. Chlorinated hydrocarbons may be analysed intact using chromatography with a dual FID NPD (nitrogen–phosphorus detection) system, but greater sensitivity can be achieved using electron capture detection (ECD). The methods applied in clinical and forensic cases do not need to be highly sensitive, because most compounds that belong to this class are only slightly toxic and severe symptoms of poisoning are observed only after ingestion of large quantities (several grams). Moreover, the symptoms often result from the solvents in which the chlorinated hydrocarbons are formulated. A useful reference for the determination of chlorinated hydrocar bons in human serum using GC after SPE is Brock et al. (1996).
Pyrethrins and pyrethroids The term ‘pyrethrins’ is used collectively for the six insecticidal constituents present in extracts of the flowers of Pyrethrum cineraria folium and other species. Pyrethrins comprise esters of the natural stereoisomers of chrysanthemic acid (pyrethrin I, cinerin I and jasmolin I) and the corresponding esters of pyrethric acid (pyrethrin II, cinerin II and jasmolin II). Their low photo chemical stability has led to the manufacture of synthetic analogues (pyrethroids), which are highly toxic to insects. In recent years pyrethroids have been manufactured and used in large quantities.
Pyrethrins and pyrethroids have relatively low toxicity to humans, but exposure to these compounds by inhalation can cause localised reactions to the upper and lower respiratory tract, which leads to oral and laryngeal oedema, coughing, shortness of breath and chest pain. In acutely exposed sensitised patients a serious asthmatic-type reaction can be triggered that can prove fatal within a few minutes. GC-FID and GC-MS are appropriate detection systems for pyrethrins and they can be analysed either without derivatisation or after methylation (Bissacot and Vassilieff 1997; Fernández Gutierrez et al. 1998). Some pyrethroids such as cyfluthrine, cypermethrin and permethrin are halogen-containing and therefore GC-ECD provides a sensitive and selective method of detection for these substances.
Nitrophenols and nitro creosols Dinitrophenol, dinitro creosol and dinoseb stimulate oxidative metabolism in the mitochondria and cause profuse sweating, headache, tachycardia and fever. Dinitro creosol can be measured in blood specimens by colorimetry (Smith et al. 1978).
Chlorinated phenoxy acids are corrosive chemicals that damage the skin, eyes and respiratory and gastrointestinal tract. Ingestion of large doses causes vomiting, abdominal pain, diarrhoea, metabolic acidosis, pulmonary oedema and coma. Alkalinisation of the urine to increase the excretion of 2,4-dichlorophenoxyacetic acid (2,4-D) and other chlorophenoxy compounds has proved an effective therapy. Substituted phenoxy acids occur in commercial products as salts or esters. Conversion of salts by extraction and derivatisation to the corresponding methyl esters improves their chromatographic properties. The presence of isooctyl (2,2,4-trimethylpentyl) esters of chlorinated phenoxy acid herbicides can be indicated by using mass spectrometry.
Triazines Ingestion of about 100 g of atrazine can lead to coma, circulatory collapse, metabolic acidosis and gastric bleeding. This may be followed by renal failure, hepatic necrosis and a disseminated intravascular coagulopathy which may prove fatal. Haemodialysis is recommended for severe cases. Triazines contain several nitrogen atoms (e.g. atrazine, structure 12, Fig. 4.2), making GC-NPD a good choice for analysis. Most triazines, which are readily amenable to GC-MS, exhibit highly characteristic mass spectra of the parent compounds and yield the important degradation products, hydroxy- and des-alkyl triazines. By using LC-MS with atmospheric pressure chemical ionisation (APCI) and electrospray, and optimising the in-source parameters, the protonated triazine molecule can be seen without fragmentation (Niessen 1999).
Quaternary ammonium compounds Ingestion of concentrated paraquat formulations causes burning of the mouth, oesophagus and stomach, and after massive absorption patients die of multiple organ failure. Absorption of smaller amounts can lead to renal damage followed by a progressive pulmonary fibrosis that causes death from respiratory failure, in some cases after 2 to 3 weeks of ingestion. Treatments to reduce absorption or increase elimination have not been effective. A strongly positive urine test with the dithionite test (see below) in a sample collected more than 4 hours after ingestion indicates a poor prognosis. Measurement of the plasma paraquat concentration is a more accurate prognostic guide. Diquat is also an irri tant poison that causes vomiting, diarrhoea and epigastric pain. In severe cases, liver and renal failure, convulsions and coma may ensue, but diquat ingestion does not lead to progressive pulmonary fibrosis.
Paraquat and diquat are not extractable by conventional LLE. The diene or monoene prod ucts of reduction of paraquat and diquat by sodium borohydride can be extracted by diethyl ether from alkaline solution for chromatography. Very limited data are available for the mass spec tral characterisation of these compounds using electron impact ionisation. Colorimetric determination of paraquat and diquat after reduction with sodium dithionite under alkaline conditions is probably the most widely used technique. Both of the bipyridylium reduction products have absorbance maxima at 396 and 379 nm. Using an ion-pairing extraction technique, a lower limit of measurement of 50 lg/L can be achieved (Jarvie and Stewart1979). Radioimmunoassay and fluorescence polarisation immunoassay methods for the determination of paraquat in serum are very sensitive and require only small sample volumes, but they are not widely available. Paraquat can also be determined in serum by HPLC-UV. Diquat may be analysed in biological specimens by most of the procedures described for paraquat. Specific HPLC procedures for paraquat and/or diquat have also been described (Ameno et al. 1995; Arys et al. 2000; Ito et al. 2005) and a capillary electrophoresis–MS method has recently been developed for the analysis of paraquat and diquat in serum (Vinner et al. 2001). An LC-MS/MS method is available for the analysis of paraquat and diquat in whole blood and urine following SPE clean-up (Lee et al. 2004).
Phosphides Hydrogen phosphide (IUPAC name phosphane; commonly known as phosphine) is widely used as an insecticide and rodenticide (agricultural fumigant) and is usually generated by the action of water on metallic phosphides (aluminium, magnesium or zinc). Inhaled phosphine is readily absorbed by the lungs. Following the ingestion of metallic phosphides, phosphine is generated in the stomach and the gas acts on the gastrointestinal system and CNS. In severe cases abdominal pain, vomiting, convulsions and coma develop rapidly and death usually ensues within 2 hours. The ammonium molybdate test and commercially available detector tubes (Guale et al. 1994) may be used as qualitative and quantitative procedures for stomach contents and nonbiological materials. Phosphine can also be determined in biological samples by using GC and NPD detection (Chan et al. 1983).
Coumarin anticoagulants Accidental and intentional ingestion of 4 hydroxycoumarin rodenticides (Fig. 1) can lead to serious poisoning manifested by bleeding in multiple organ sites. Treatment consists of supplements of vitamin K (mild cases) and, for serious cases, infusions of fresh frozen plasma or purified clotting factors until the prothrombin time returns to the normal range. Warfarin and the super warfarin anticoagulant rodenticides (brodifacoum, bromadiolone, coumatetralyl and difenacoum; Fig. 1) can be analysed either intact or after derivatisation, by either GC or GC-MS methods, these being the most sensitive and selective. Five of the 4 hydroxycoumarin anticoagulants (brodifacoum, bromadiolone, coumatetralyl, difenacoum and warfarin) can also be resolved and determined in serum by HPLC with fluorimetric detection (Felice et al. 1991).
Figure 1 Chemical structures of the 4-hydroxycoumarin anticoagulant rodenticides.
Organic and inorganic metallic compounds A wide range of organic and inorganic metallic compounds are found in agricultural use. Inorganic and organometallic compounds are used as acaricides (organotin), herbicides (organ arsenic), fungicides (dithiocarbamate compounds of nickel and dithiocarbamate complexes with manganese and zinc, organic and inorganic compounds of copper and mercury) and rodenticides (magnesium, aluminium and zinc phosphides, and thallium sulfate). For some compounds, exposure to the organic form results in more serious toxicity and the features of poisoning may be quite different from those of the inorganic compound. Metallic compounds and their associated clinical symptoms are discussed below under metals and anions together with the methods used for their analysis.
