Organophosphate-Induced Delayed Polyneuropathy

Toxicol Rev 2005; 24 (1): 37-49 REVIEW ARTICLE 1176-2551/05/0001-0037/$34.95/0 2005 Adis Data Information BV. All rights reserved. Organophosphate-Induced Delayed Polyneuropathy Marcello Lotti and Angelo Moretto Department of Environmental Medicine and Public Health, University of Padua, Padova, Italy Contents Abstract...37 1. Structure-Activity and Dose-Effect Relationships...38 2. Mechanisms...38 3. Clinical Aspects in Humans...39 4. Organophosphates Causing Organophosphate-Induced Delayed Polyneuropathy (OPIDP)... 40 4.1 Chlorpyrifos...40 4.2 Dichlorvos...41 4.3 Isofenphos...41 4.4 Methamidophos...42 4.5 Mipafox...42 4.6 Trichlorfon...43 4.7 Trichlornat...43 4.8 Mevinphos and Phosphamidon...43 5. Carbamates...43 6. Other Organophosphates Associated with OPIDP...44 7. Long-Term, Low-Level Exposures to Organophosphorus Esters and Neuropathy... 45 8. Conclusions...46 Abstract Organophosphate-induced delayed polyneuropathy (OPIDP) is a rare toxicity resulting from exposure to certain organophosphorus (OP) esters. It is characterised by distal degeneration of some axons of both the peripheral and central nervous systems occurring 1 4 weeks after single or short-term exposures. Cramping muscle pain in the lower limbs, distal numbness and paraesthesiae occur, followed by progressive weakness, depression of deep tendon reflexes in the lower limbs and, in severe cases, in the upper limbs. Signs include high-stepping gait associated with bilateral foot drop and, in severe cases, quadriplegia with foot and wrist drop as well as pyramidal signs. In time, there might be significant recovery of the peripheral nerve function but, depending on the degree of pyramidal involvement, spastic ataxia may be a permanent outcome of severe OPIDP. Human and experimental data indicate that recovery is usually complete in the young. At onset, the electrophysiological changes include reduced amplitude of the compound muscle potential, increased distal latencies and normal or slightly reduced nerve conduction velocities. The progression of the disease, usually over a few days, may lead to non-excitability of the nerve with electromyographical signs of denervation. Nerve biopsies have been performed in a few cases and showed axonal degeneration with secondary demyelination. Neuropathy target esterase (NTE) is thought to be the target of OPIDP initiation. The ratio of inhibitory powers for acetylcholinesterase and NTE represents the crucial guideline for the aetiological attribution of OP-induced peripheral neuropathy. In fact, pre-marketing toxicity testing in animals selects OP insecticides with cholinergic toxicity potential much higher than that to result in OPIDP. Therefore, OPIDP may develop only after very large exposures to insecticides, causing severe cholinergic toxicity. However, this was not the case with certain triaryl phosphates that were not used as insecticides but as hydraulic fluids, lubricants and plasticisers and do not result

38 Lotti & Moretto in cholinergic toxicity. Several thousand cases of OPIDP as a result of exposure to tri-ortho-cresyl phosphate have been reported, whereas the number of cases of OPIDP as a result of OP insecticide poisoning is much lower. In this article, we mainly discuss OP pesticide poisoning, particularly when caused by chlorpyrifos, dichlorvos, isofenphos, methamidophos, mipafox, trichlorfon, trichlornat, phosphamidon/mevinphos and by certain carbamates. We also discuss case reports where neuropathies were not convincingly attributed to fenthion, malathion, omethoate/dimethoate, parathion and merphos. Finally, several observational studies on long-term, low-level exposures to OPs that sometimes reported mild, inconsistent and unexplained changes of unclear significance in peripheral nerves are briefly discussed. Organophosphorus (OP) esters may cause several toxic effects Thus, the ratio of inhibitory powers for AChE and NTE reprein humans and animals. [1] A relatively rare toxicity is the distal sents the key dose-effects relationship of each OP and the crucial degeneration of some axons of both the peripheral and central guideline for the aetiological attribution of cases of peripheral nervous systems, known as organophosphate-induced delayed pol- neuropathy, when sorting out various other differential diagnoses. yneuropathy (OPIDP). Although several epidemics of this neurop- In fact, although OPIDP hazard is shared by several OPs, the risk athy occurred in the past, mainly due to accidental ingestion of of developing OPIDP is actually rather small for commercial tri-ortho-cresyl phosphate (TOCP), including the well known Gin- insecticides, because pre-marketing toxicity testing in animals ger Jack Paralysis, [2] preventive measures substantially reduced selects OPs with cholinergic toxicity potential much higher than the risk of OPIDP. [3,4] that causing OPIDP. [3,10] Therefore, OPIDP may develop only This article describes the clinical features of OPIDP in humans after very large exposures, such as those occurring in suicidal and critically discusses all reported cases, particularly those pubthis attempts, causing unambiguous cholinergic toxicity. However, lished since previous reviews. [1,5,6] In order that the clinical discuscause will not be the case for certain triaryl phosphates that do not sion is better appreciated, a short summary of current understandlubricants cholinergic toxicity and have been used as hydraulic fluids, ing of OPIDP mechanisms is included. or plasticisers. [11] All cases of human poisoning, so far convincingly described, 1. Structure-Activity and Dose-Effect Relationships have confirmed these chemical and biochemical characteristics of Structure-activity studies in animals, both in vitro and in vivo, OPs. indicate that only certain OPs cause OPIDP. Thus, phosphinates and other non-op esterase inhibitors such as carbamates and 2. Mechanisms sulfonyl halides do not cause neuropathy because of the nature of OPs react covalently as pseudo-substrates at the catalytic centre their interaction with neuropathy target esterase (NTE), a protein of a variety of serine hydrolases, including NTE and AChE. in the nervous system thought to be the molecular target of Contrary to true substrates, the rate of hydrolysis of the phospho- OPIDP. [7] rylated enzymes is extremely slow. In addition, a second reaction Among OPs, with the general structure R1R2P(O or S)X, that may occur, known as aging, which results in the cleavage of one of are capable of causing OPIDP, several factors determine their the bound alkyl groups of the phosphoryl residue (R 1 or R2) delayed neurotoxicity potential. Characteristics of the chemical leaving a negative charge at the residue attached to the active site. structure that increase this potential include the choice of In this way the enzyme becomes permanently inhibited. [12] phosphonates and phosphoroamidates rather than the homologous OPIDP is initiated by phosphorylation and subsequent aging of phosphates and the increase in carbon chain length of R1 and R2. at least 70% of NTE in peripheral nerves and this two-step process Characteristics that decrease this potential are bulky hydrophilic or occurs within hours of poisoning. [7] A similar threshold must also nitro-phenol groups at X and thioether linkages at X. [8] be reached after repeated exposures because even cumulative Studies with a variety of OPs on the sensitivity of the target doses higher than a single effective one may fail to cause OPIDP, enzymes showed that the comparative inhibitory power of OPs if the threshold level of inhibition is not reached. [13] The dose level against hen acetylcholinesterase (AChE) and NTE in vitro corre- and the interval between doses may or may not allow substantial lates with their comparative effects in vivo, i.e. death or delayed resynthesis of the enzyme and thus the reaching of the threshold. neuropathy. Moreover, comparison of the in vitro effects seen with This threshold was found to be much higher (>90%) in young than hen and human enzymes indicates that the hen animal model well in adult animals. [14] Organophosphinates, as well as other non-op predicts the development of OPIDP in humans. [9] inhibitors such as carbamates and sulfonyl fluorides, also cova-

Organophosphate-Induced Delayed Polyneuropathy 39 lently react with NTE, but cannot undergo the aging reaction. As a OPIDP is usually observed within a few days from the onset of consequence, these inhibitors do not cause OPIDP and, when symptoms and in the absence of further exposure no progression of given to experimental animals before a neuropathic OP, they the disease is observed. After repeated exposures to non-antichoprotect from OPIDP by occupying the NTE active site. Thus, the linesterase triaryl phosphates, the onset of symptoms and signs and inhibition of NTE catalytic activity is not the mechanism leading their full development is more variable and less definable. to axonal degeneration. [7,15] Hypotheses to explain the conse- The usual initial complaint is cramping muscle pain in the quences of these OP-NTE interactions include either a loss of a lower limbs followed by distal numbness and paraesthesiae. Pronon-esterase function of NTE essential for the axon, or a gain of a toxic function of phosphorylated/aged NTE. [16] NTE is also prelar and Achilles reflexes. Severe OPIDP is characterised by symp- gressive weakness then occurs, together with depression of patelsent in circulating lymphocytes and its inhibition was found to be a useful biomarker to herald the development of OPIDP. [17-19] toms and signs of neuropathy also appearing in the arms and Studies with the recombinant domain of NTE purified from forearms with depression of corresponding tendon reflexes. Physibacterial lysates suggest that membrane lipids are putative cellular cal examination reveals symmetrical predominantly motor poly- substrates of this enzyme, raising the possibility that NTE may be neuropathy, with wasting and flaccid weakness of distal limb involved in intra-neuronal membrane trafficking and lipid homeois muscles, especially in the legs. Objective evidence of sensory loss stasis. [20,21] However, since this activity is inhibited also by nonistic usually less severe or even absent. [18] Signs include a character- neuropathic compounds it is obviously not essential for the maintenance high-stepping gait associated with bilateral foot drop. Severe of the axon, and therefore not correlated with the initiation OPIDP may result in quadriplegia with foot and wrist drop as well of axonal degeneration. as pyramidal signs. The detailed clinical description of a case Studies have been performed with genetically modified mice in series of TOCP poisoning has been reported. [27] the attempt to clarify the physiological function(s) of NTE. How- In time, there is significant functional recovery in less severe ever, mice are known to be resistant to OPIDP, [22] and the data cases with most distal involvement and sparing of spinal cord obtained with these studies did not shed light on the mechanism of axons. Otherwise pyramidal and other signs of central neurologi- OPIDP. Nevertheless, they indicate the essential role of NTE in cal involvement may become more evident. The degree of pyramifetal development. Winrow et al. [23] showed that knock-out Nte / dal involvement determines the prognosis for functional recovery, mice died in utero, probably as a result of failed placental developand spastic ataxia may represent a permanent outcome of severe ment. Additionally, impairment of vasculogenesis in the yolk sacs OPIDP. [27,28] The very few cases of OPIDP reported in young and embryos of null mutant conceptuses suggests that NTE is also individuals indicate that they recover completely, even from serequired for normal blood vessel development. [24] Conditional inactivation of mouse NTE gene resulted in elimination of NTE vere lower- and upper-limb involvement. [29,30] protein in the central nervous system. [25] These animals displayed At onset, the electrophysiological changes include reduced neuronal degeneration and loss of endoplasmic reticulum in the amplitude of the compound muscle potential after supramaximal hippocampus, thalamus and cerebellum. stimulation of motor nerves, increased distal latencies, and normal In summary, although there is compelling evidence of the or slightly reduced nerve conduction velocities. The progression of involvement of NTE in OPIDP initiation, its role in axonal degeneration the disease, usually over a few days, may lead to non-excitability remains obscure. of the nerve. Electromyography reveals signs of denervation of the affected muscles with increased insertional activity, spontaneous 3. Clinical Aspects in Humans activity (fibrillation potentials and positive sharp waves), and reduced interference pattern. [17-19,31-37] During recovery, polyphasic motor unit potentials appear. These findings are consistent with Symptoms of OPIDP usually begin 2 3 weeks after single doses when, as in the case of insecticides, both cholinergic and those of distal degeneration of axons, as seen in the few cases intermediate syndrome signs have subsided. [26] The delay between where a biopsy of the sural nerve was performed, [17,36-40] and single or short-term exposure and the clinical onset of OPIDP parallel the clinical signs of peripheral neuropathy. Very few data depends on both the chemical and the dose. OPs with slow are available on spinal cord distribution of the lesions; they are pharmacokinetics or at relatively lower doses may cause OPIDP similar to that seen in the animals, and involve the distal pyramidal after a lag-time of 4 5 weeks, whereas higher doses of OPs may tract and the proximal columns of Goll (fasciculus gracili medul- reduce this period to as little as 10 days. Full clinical expression of lae spinalis). [27,41]

40 Lotti & Moretto Table I. Confirmed cases of organophosphate-induced delayed polyneuropathy (OPIDP) as a result of insecticide poisonings a Insecticide IUPAC name No. of cases (type of exposure) References Chlorpyrifos O,O-diethyl O-3,5,6-trichloro-2-pyridyl phosphothioate 1 (suicide attempt) 17 Dichlorvos 2,-dichlorovinyl dimethyl phosphate 5 (4 suicide attempts, 1 accidental) 31,32,34,35 Isofenphos O-ethyl O-2-isopropoxycarbonylphenyl 2 (suicide attempts) 18,45 isopropylphosphoroamidothioate Methamidophos O,S-dimethyl phosphoramidothioate 25 (9 suicide attempts, 1 accidental, 18,19,33,46,47 3 occupational, 12 not reported) Mipafox N,N-diisopropyl phosphorodiamidofluoridate 2 (occupational) 48 Phosphamidon/mevinphos 2-chloro-2-diethylcarbamoyl-1-methylvinyl dimethyl 1 (suicide attempt) 40 phosphate/2-methoxycarbonyl-1-methylvinyl dimethyl phosphate Trichlorfon Dimethyl 2,2,2-trichloro-1-hydroxyethylphosphonate 157 (29 suicide attempts, 24 accidental, 35-37,49-52 18 occupational and 100 not reported) Trichlornat O-ethyl O-2,4,5-trichlorophenyl ethylphosphonothioate 2 (suicide attempts) 38,39 a Cases of neuropathy likely caused by certain carbamates (carbaryl, carbofuran and metolcarb) are discussed in detail in section 5. A description of OPIDP cases caused by non-insecticidal organophosphorus esters, such as triaryl phosphates, is discussed in section 4. IUPAC = International Union of Pure and Applied Chemistry. 4. Organophosphates Causing and by the presence of intermediate syndrome. The characteristics Organophosphate-Induced Delayed of polyneuropathy were delayed onset, initial localisation in distal Polyneuropathy (OPIDP) lower limbs, electrophysiological data compatible with distal axonopathy, and late development of spasticity in severe cases. Evidence Several thousand cases of OPIDP caused by TOCP have been from animal studies, in which appropriate doses of the OP reported and the literature published until 1988 has been sum- were used, was also considered. marised. [42] TOCP and related triaryl phosphates cause no cholinafter In general, the onset of OPIDP occurred between 8 and 15 days ergic toxicity and have been used as plasticisers, lubricants and exposure in severe cases involving both the upper and lower hydraulic fluids. However, triaryl phosphates presently used as jetweeks. limbs, whereas in mild cases the onset was delayed for up to 6 engine lubricants contain very little, if any, TOCP. [4] Well known Some patients were followed up for months or years. Mild OPIDP epidemics include the Ginger Jack paralysis in the 1930s cases sometimes resulted in complete recovery, whereas severe and that in Morocco in the 1950s. [2,43] Almost all cases of neuropasion, cases showed spasticity after improvement of the peripheral lethy caused by TOCP were a consequence of accidental poisoning. revealing damage of the first motor neuron. Sensory involve- More recent reports include a 50-year follow-up of an epidemic ment occurred in some cases only and was generally less relevant probably as a result of TOCP poisoning in Italy [44] and a case of than the motor involvement. Estimated doses were all well above acute poisoning in a 4.5-year-old boy resulting in a mild neuropa- experimental oral doses that are lethal to 50% of animals tested thy that recovered in about 5 weeks. [30] (LD 50 ). Sural nerve biopsy was available in seven cases and showed distal axonal degeneration in five cases that had been In contrast with triaryl phosphates, cases of OPIDP caused by caused by chlorpyrifos, mevinphos/phosphamidon, trichlorfon and insecticides are fewer and those convincingly attributable to this trichlornat, [17,36-38,40] whereas it was apparently normal in the reclass of OPs are summarised in table I. Non-exclusive inclusion maining ones. [31,36] Case reports attributed to individual OP inseccriteria were evidence of exposure and reasonable identification of ticides are discussed in more detail in the following sections. the chemical, severity of the cholinergic syndrome, timing of onset, clinical characteristics of the polyneuropathy, and electrophysiological and morphological data. Except for cases related 4.1 Chlorpyrifos to chlorpyrifos, [17] isofenphos [18] and methamidophos, [18] the ana- In a case of a suicide attempt with chlorpyrifos, the estimated lytical confirmation of the causative agent in biological fluids was oral dose of 300 mg/kg was associated with a blood concentration not available. The severity of cholinergic overstimulation was of 245 µg/l of chlorpyrifos measured 3 days after ingestion. [17] assessed by type and duration of symptoms and signs, requirement The patient had severe cholinergic toxicity, requiring specific and of specific pharmacological treatment and of artificial ventilation, supportive treatment for 23 days. Lymphocytic NTE was still

Organophosphate-Induced Delayed Polyneuropathy 41 repeated exposure to chlorpyrifos. If exposure occurred, it was low because only mild signs, possibly cholinergic in nature, were either observed by the authors or reported by the patients. The authors concluded that sensory neuropathy was caused by chlor- pyrifos. This conclusion has already been challenged [18] because of the lack of analytical confirmation of exposure, the mild cholinergic syndrome, if any, and the purely sensory involvement of the lower limbs. A man was exposed dermally and probably by inhalation to a 45% formulation of chlorpyrifos while spraying a nest of termites. [61] He had mild possibly cholinergic signs and was treated with atropine 0.5mg on three occasions and pralidoxime 2g. On day 2 he had paralysis of the left lower limb. Electrophysiological studies performed on day 3 showed prolonged sensory and distal motor latencies, absent F-waves in upper limbs, and asymmetric F-waves in lower limbs. Whole blood butyrylcholinesterase activities were normal on day 4. Muscle weakness was also seen in the upper left limb. Electrophysiological data obtained on day 31 were reported as consistent with a predominantly motor axonal periph- eral neuropathy, but details were not given. After 10 years, the clinical picture was only slightly improved. The author attributes this syndrome to chlorpyrifos exposure, but this report has been severely criticised. [62] In fact, acute toxicity was very mild, if any, it is unclear whether upper and lower limbs were similarly affected both on day 2 and 31, and the nerve involvement was asymmetri- cal. In addition, the early onset of neuropathy and the 10 years follow-up without recovery are against OPIDP and any toxic peripheral neuropathy. Chlorpyrifos has caused OPIDP in experi- mental animals. [63] 4.2 Dichlorvos Dichlorvos, besides being a pesticide on its own, is also the active metabolite of trichlorfon. It caused a severe cholinergic syndrome in all reported cases, [31,32,34,35] and intermediate syn- drome in two. [32,35] Dosage was estimated in one case corresponding to 1 g/kg. [31] In all these severe cases, the involvement of both lower and upper limbs was clinically and electrophysiologically consistent with OPIDP. In the follow-up of two cases, signs of spasticity were observed later in the course of the disease. [32,34] Dichlorvos has caused OPIDP in experimental animals. [64] 4.3 Isofenphos In a case of isofenphos/phoxim poisoning, the estimated dosage of isofenphos was 300 mg/kg, corresponding to a blood level of 0.1 µg/l, when measured 3 hours after poisoning. [65] The cholinergic syndrome was treated with high doses of atropine and pralidoxime for 12 days. OPIDP developed 3 weeks after the significantly inhibited on day 30 after poisoning. Sensorimotor neuropathy of the lower limbs developed about 6 weeks after ingestion and diagnosis was based on clinical, electrophysiological and morphological evidence. Martinez-Chuecos et al. [53] also reported a case of OPIDP after a severe chlorpyrifos poisoning, but no clinical and electrophysiological details were provided. Beside these cases, several atypical cases of peripheral neurop- athy have been attributed to chlorpyrifos. De Silva et al. [54] described isolated bilateral recurrent laryngeal paralysis in three subjects acutely poisoned by chlorpyrifos, parathion or methami- dophos, respectively. These patients had a severe cholinergic syndrome requiring assisted ventilation for 8 10 days, developed intermediate syndrome and, 14 26 days later, bilateral vocal cord paralysis, as seen by laryngoscopy. The striking selectivity for laryngeal nerves and the fact that parathion does not cause OPIDP may suggest other aetiologies for such laryngeal paralysis. Al- though the delay between extubation and onset of laryngeal symp- toms was up to 26 days, it is known that prolonged intubation may be the cause of similar paralyses. [55,56] No analytical confirmation of the poisoning was available in these cases. Bilateral vocal cord paralysis was also described in a 3-year-old child, 11 days after an accidental mild poisoning with chlorpyri- fos. [57] On day 19 he also showed areflexia and electrophysiologi- cal findings consistent with proximal polyneuropathy that recovered by day 52. The authors concluded that both the vocal cord paralysis and the peripheral neuropathy were due to chlorpyrifos exposure, although a traumatic cause for the vocal cord paralysis could not be excluded. This report was criticised because the electrophysiological pattern was not consistent with OPIDP (normal sensory and motor action potentials, lack of fibrillation potentials, proximal nerve damage), the acute poisoning was not severe and the recovery from neuropathy too quick. [58] Although neither the severity of poisoning nor the electrophysiology are consistent with OPIDP, such a rapid recovery would be expected in patients of a young age with OPIDP. A boy allegedly poisoned by chlorpyrifos had left vocal cord paralysis on day 5 after poisoning that recovered within 2 weeks. [59] Although the authors described this phenomenon as intermediate syndrome, this isolated and monolateral paralysis is not characteristic of any toxic disease. The patient also developed on day 9 right-sided wrist drop associated with motor and sensory action potential reduction of the right radial nerve that was interpreted as OPIDP. However, asymmetrical neuropathy, the involvement of the upper limb only, the very early onset, and the incomplete description of electrophysiology do not support the authors conclusions. Kaplan et al. [60] described seven cases of sensory neuropathy that developed within 4 weeks from the beginning of an ill-defined

42 Lotti & Moretto suicidal ingestion and was characterised by flaccid quadriplegia out of 41 severe poisonings and one case out of 63 moderate followed, months later, by severe spasticity in both lower and poisonings were observed. All patients recovered within an unupper limbs. The estimated dose of phoxim (O,O-diethyl-cya- specified period of time, whereas one was still paralysed after 64 nobenzylideneamino oxyphosphonothioate) was 125 mg/kg, but it months. Because of the lack of details, the report is difficult to was not considered relevant to neuropathy development on the interpret. basis of negative experimental data. [65,66] Two female patients were possibly dermally exposed to a In a case attributed to isofenphos, [45] there was a severe cholin- formulation of methamidophos. [71] They had signs neither entirely ergic syndrome followed by a polyneuropathy, but few details consistent with a cholinergic syndrome nor associated with inhibiwere given. Some uncertainties arise in another case [67] that was tion of either plasma or erythrocyte cholinesterases, when meaattributed to isofenphos and to the dithiocarbamate fungicide sured within 24 hours of exposure. One patient complained on day maneb (manganese ethylenebis [dithiocarbamate]). Maneb does 9 of distal paraesthesiae in the legs and electrophysiological data not cause peripheral neuropathy in rodents, [68] the estimated dose suggested peripheral neurological damage, but no details were of isofenphos was quite low, and the cholinergic syndrome was given. There was a progression during the following 2 months very mild, requiring one atropine injection only. Isofenfos has showing asymmetrical, mainly sensory deficits that were associatcaused OPIDP in experimental animals. [65,69] ed with alteration of some psychometric tests. Full recovery occurred after 18 months. The other patient complained on day 5 of 4.4 Methamidophos reduced visual acuity of the left eye, muscle weakness and tendon hyperreflexia of the upper limbs. Electrophysiological data were Ten cases of OPIDP after suicidal ingestion or occupational normal on day 20, whereas sensorimotor deficits in the left lower exposures to methamidophos have been reported. [33] All patients limb and in the upper limbs appeared 2 months later. The clinical had similar clinical characteristics, and a typical one was described picture progressed over 5 months and recovered after 18 months, in detail. It showed severe cholinergic toxicity followed 2 weeks except for atrophy of hand muscles. The authors concluded that later by lower limbs motor neuropathy, with electrophysiological these patients had neuropathy caused by methamidophos but the data consistent with axonopathy. clinical picture is not consistent with OPIDP. In fact, the lack of a In one case of suicide, [18] with an estimated dosage of methamipresence severe cholinergic syndrome, the early onset of features, the dophos 600 mg/kg, the blood concentration was 27 µg/l and the of asymmetry, preferential localisation in the upper lymphocytic NTE was <10% of baseline, when measured 8 hours limbs, and progression up to 7 months, all speak strongly against after poisoning. Following a severe and prolonged cholinergic toxic neuropathies, including OPIDP. syndrome, the patient developed a purely motor neuropathy of the Miranda et al. [72,73] studied grip and pinch strength after acute lower limbs with a typical electrophysiological pattern of distal OP poisoning. They found that subjects severely poisoned by OPs axonopathy. known to cause OPIDP, mainly methamidophos, had a reduced In a case of suicide [19] with methamidophos, the cholinergic mean strength when measured 1 18 and 18 128 days after poisonsyndrome was severe, lymphocytic NTE was inhibited on day 3 ing. However, a not statistically significant reduction was also and a purely motor OPIDP developed 2 weeks after ingestion. observed in subjects severely poisoned by OPs not causing Electrophysiological data confirmed the axonal type of neuropaobserved OPIDP. After 24 32 months, a partial or complete recovery was thy. in the former and latter groups, respectively. The authors Another case [46] is not detailed but the preceding cholinergic report that some of the subjects did in fact develop clinical syndrome and the characteristic polyneuropathy was consistent neuropathy. Therefore, it is not clear how data from these subjects with OPIDP. influence the mean values of their group and if they can explain the Aygun et al. [47] reported 12 cases of OPIDP caused by difference with the mean of the group of subjects exposed to nonmethamidophos among 41 consecutive cases of OP insecticide neuropathic OPs. Methamidophos has caused OPIDP in experi- poisoning in Turkey. In six cases, electrophysiological data were mental animals. [74] available and consistent with OPIDP. Three of these showed both sensory and motor involvement, the others were only motor in 4.5 Mipafox nature. However, although the neurological examinations were consistent with OPIDP in all cases, several details are missing. Mipafox caused OPIDP in two laboratory workers involved in A series of OP poisonings in China, allegedly caused by the manufacturing of mipafox as a perspective pesticide. [48] Pamethamidophos, has been described. [70] Thirteen cases of OPIDP tients had signs of cholinergic toxicity before the development of a

Organophosphate-Induced Delayed Polyneuropathy 43 sensorimotor neuropathy on both upper and lower limbs consistent with OPIDP. Mipafox has caused OPIDP in experimental animals. [75] 4.6 Trichlorfon Hsiao et al. [79] reported a case of neuropathy after a severe poisoning with mevinphos but no clinical details were given. Mevinphos did not cause OPIDP in hens, when tested at about the LD50. [80] Several cases of OPIDP clearly caused by trichlorphon have 5. Carbamates been described. [35-37,49-52] A large number of trichlorfon-induced neuropathies have also occurred in the former Soviet Union, and Special consideration should be given to case reports of neurop- have been discussed. [50] When estimated, dosages of trichlorfon athy associated with carbamate ingestion, and caution in their were between 200 and 500 mg/kg, the cholinergic syndrome was aetiological attribution should be exercised. Whereas carbamates always severe lasting 4 7 days, the onset was between 1 3 weeks, may be NTE inhibitors, they are thought to be unable to cause electrophysiological and, when available, histopathological data neuropathy because carbamylated NTE does not undergo aging. were consistent with OPIDP. The involvement of upper limbs was Moreover, carbamates protect against OPIDP in experimental frequently observed. Follow-up in several cases showed improvehigh models. [7] However, some experiments showed that extremely ment of upper-limb neuropathy and concurrent development of inhibition of NTE caused by non-ageable inhibitors, includ- lower-limb spasticity. ing carbamates, correlated with the development of a mild neurop- A man was occupationally exposed to trichlorfon and 24 hours athy in hens. In an analogy with pharmacological models of druglater experienced vomiting and diarrhoea lasting for several receptor interactions, the proposal was made that all NTE inhibi- days. [76] After 3 months, he developed a peripheral neuropathy. tors may have the potential to cause neuropathy. Thus, NTE Electrophysiology showed reduction of sensory and motor nerve inhibitors might have variable intrinsic activities to initiate neuconduction velocities and signs of denervation. Biopsy of the sural ropathy and non-ageable inhibitors, such as carbamates, might be nerve showed some axonal degeneration, together with reduction among the weakest. [15] of the number of nerve fibres that were replaced by macrophages. A case of neuropathy occurred in a patient after a suicidal Although the histopathological picture may be compatible with an attempt with carbaryl (1-naphthyl methylcarbamate). [81] Cholineraxonopathy, it is difficult to attribute this neuropathy to trichlorfon gic toxicity was severe and lasted for a few days, whereas signs because the delay in onset was excessively long, and the preceding and symptoms of neuropathy involving upper and lower limbs, cholinergic syndrome, if any, was too mild. Trichlorfon has caused began 3 weeks after ingestion and were consistent with distal OPIDP in experimental animals. [50] axonopathy, as confirmed by electrophysiology. A man was probably exposed to carbaryl for several months in 4.7 Trichlornat his house. [82] He was reported to experience a progressive glove- Two cases of OPIDP caused by trichlornat have been reportgiven. The authors concluded that since there are no other explana- and-stocking peripheral neuropathy but no further details were ed. [38,39] Both patients had severe cholinergic toxicity and 2 3 tions for the peripheral neuropathy, the cause must have been weeks later developed OPIDP on both upper and lower limbs. carbaryl exposure. No analytical evidence of exposure to carbaryl Electrophysiological data and the description of histopathological was given and only few, mild, possibly cholinergic, signs such as findings on sural nerve were consistent with distal axonopathy. rhinorrhoea, diarrhoea, lacrimation, ill-defined muscle fascicula- Trichlornat has caused OPIDP in experimental animals. [77] tion were at times reported. In addition, the clinical picture of a 4.8 Mevinphos and Phosphamidon neuropathy progressing after the end of exposure is also not consistent with toxic aetiologies An OPIDP case was reported [40] where the suspected causative Carbaryl was found to be a good NTE inhibitor in experimental agent was a mixture of mevinphos and phosphamidon. After animals and no development of neuropathy was reported. [83] severe cholinergic toxicity, the patient developed sensorimotor Following a suicide attempt with metolcarb (m-tolyl methylneuropathy followed by spasticity due to spinal atrophy, as seen carbamate), [84] the patient was comatose for 72 hours and on day 6 with magnetic resonance imaging. It is not clear whether the upper signs and symptoms of neuropathy begun to develop. Both electrolimbs were also involved. The estimated dose of mevinphos was physiology and histopathology revealed signs of axonopathy and 20g and that of phosphamidon was 50g. The latter has never been demyelination. Although loss of myelin and reduction of nerve reported to cause OPIDP in humans, whereas it has caused OPIDP conduction velocity are not characteristic, the whole clinical picin the hen. [78] ture suggests that neuropathy may have been caused by metolcarb.

44 Lotti & Moretto No experimental data are available on the neuropathic potential of amounts of O,O,S,-tri-methylphosphorothioate. [87,88] The cholinermetolcarb. gic syndrome required mechanical ventilation, and treatment with A man attempted suicide by ingesting 100mL carbofuran atropine and obidoxime for 3 days. The clinical course was (2,3-dihydro-2,2-dimethyl-7-benzofuranyl methylcarbamate), but characterised by several complications including arrhythmia, antino analytical confirmation of the diagnosis was available. [85] The cholinergic-induced ileus and renal failure. Two weeks after expocholinergic syndrome was treated with atropine and resolved sure, the patient developed diffuse interstitial lung fibrosis, per- within 72 hours. Paraesthesiae appeared 6 days after poisoning and haps linked to the presence of the tri-methylphosphorothioate, on day 10 muscle weakness, predominantly of the lower limbs, known to cause this toxicity in rats. [89] On day 10, the patient developed. On day 20, electrophysiological data showed reduced reported generalised weakness, especially on distal limb muscles. muscle action potentials with slightly reduced conduction velocithough Electrophysiological studies were not consistent with OPIDP al- ties of tibial and peroneal nerves, whereas distal latencies were there was histopathological evidence of some degenerating within normal values. Parameters of the sural nerve were normal. axons in the sural nerve. Recovery was complete after 3 months. Clinical but not electrophysiological improvement was observed Malathion neither caused OPIDP nor inhibited NTE in hens, [90] about 4 months later. The clinical picture is consistent with a and several hundred cases of human poisonings have never been peripheral neuropathy caused by carbofuran. However, no experidescribed associated with OPIDP. [91] Therefore, it seems more likely that the mental data are available on the potential of carbofuran to cause mild neurological signs derived from the patient s gen- neuropathy. eral status and prolonged intensive care. In summary, clinical evidence suggests that extremely high Rivett and Potgieter [92] described a case of severe poisoning doses of certain carbamates may cause delayed neuropathy in with malathion and brodifacoum, an anticoagulant rodenticide, in humans. Perhaps, the very effective antidotal and supportive treat- a man with a history of alcohol abuse. The cholinergic syndrome ment for cholinergic toxicity created the conditions for a very high required atropine treatment for 29 days and mechanical ventila- NTE inhibition, similar to that associated with carbamate neuropa- tion. The clinical course was complicated by pneumonia and thy in animals. However, high NTE inhibition in animals was septicaemia. Bilateral diaphragmatic weakness lasted 6 months. obtained by means of repeated administration. Allowing spontane- Transient bilateral sixth nerve weakness and permanent sensory ous reactivation of AChE between doses, the limiting factor of motor neuropathy of lower limbs were also observed. The authors cholinergic toxicity was in this way circumvented. [15] concluded that malathion caused diaphragmatic paralysis. However, no other cases have ever been reported of OPIDP involving 6. Other Organophosphates Associated with OPIDP diaphragmatic innervation and the permanent peripheral neuropathy was likely associated with the history of alcohol abuse. Several other OP insecticides have been reported to cause Following a suicidal attempt with mecarbam [S-(N-ethoxy- OPIDP. However, these reports are not convincing for many carbonyl-n-methylcarbamoyl-methyl) O,O-diethylphosphoreasons, including lack of details, no evidence of OPIDP in experi- rodithioate] 2 g/kg, [93] the patient was comatose for 17 days but mental animals and more likely alternative aetiologies. neither the cholinergic syndrome nor the treatment were described. Aygun et al. [47] reported a case of OPIDP after mild fenthion No analytical confirmation of the diagnosis was available. When (O,O-dimethyl O-4-methylthio-m-tolyl phosphorothioate) poison- awakened, the patient developed muscle weakness predominantly ing. Neither analytical evidence for the causative agent nor a in the lower limbs, more evident on the left side, and a mild description of the clinical features of neuropathy were given. cerebellar syndrome with no significant sensory involvement. However, the electrophysiological data are consistent with Three months after poisoning, electrophysiological data were con- OPIDP. Although fenthion did not cause OPIDP and did not sistent with axonal degeneration of motor nerves of the lower inhibit NTE when administered to hens at about the LD50, [86] limbs. The description of sural nerve histopathology is difficult to doubts arise about the identity of the chemical involved in this interpret, because it shows both extensive segmental demyelincase. Martinez-Chuecos et al. [53] also reported on a case of severe ation and axonal sprouts. After 1 year there was partial recovery fenthion poisoning that resulted in OPIDP. The blood concentra- with evidence of involvement of the first motor neuron as shown tion of fenthion was 68.9 mg/l on day 11 after poisoning but no by brisk deep tendon reflexes in the lower limbs. Whereas the clinical and electrophysiological details were given. described clinical picture may be broadly consistent with OPIDP, A woman attempted suicide with a malathion [diethyl mecarbam did not cause OPIDP in hens when given a dose (dimethoxythiophosphorylthio) succinate] formulation containing corresponding to about the LD50 administered twice, 21 days relatively large amounts of isopropyl malathion and trace apart. [94]

Organophosphate-Induced Delayed Polyneuropathy 45 A case of polyneuropathy following a suicide attempt with an organophosphorus compound thought to be omethoate (O,O-di- methyl S-methylcarbamoylmethyl phosphorothioate) was de- scribed. [95] The patient had a severe cholinergic syndrome compli- cated by respiratory infection and required artificial ventilation for about 20 days. The onset of the neuropathy, about 3 weeks after exposure, was characterised by the initial involvement of left arm. Subsequently, both lower and upper limbs showed severe motor deficits. Electrophysiological data showed denervation in all distal muscles that were examined and severe reduction of motor con- duction velocity, both in upper and lower limbs. No analytical confirmation of the causative agent was available and some data such as those on nerve conduction were not consistent with OPIDP. Solé Violàn et al., [96] reporting on several cases of acute OP poisoning, quoted two cases involving dimethoate (O,O-dimethyl S-methylcarbamoylmethyl phosphorodithioate) and one involving omethoate that developed delayed polyneuropathy. Omethoate is the active metabolite of dimethoate and an insecticide on its own. However, no details were given on these cases. A case of reflex sympathetic dystrophy syndrome was attributed to dimethoate. [97] Ten weeks after the acute cholinergic syndrome, the patient devel- oped pain, swelling, stiffness and colour change in the hands. Electrophysiological studies of the upper extremities showed reduction of motor action potentials and conduction velocities. However, the long delay before the onset of features, the involve- ment of the upper limbs only, and the electrophysiological find- ings are not consistent with OPIDP. OPIDP has never been reported in several cases of omethoate poisoning in humans or from hen testing, and in a fatal poisoning no NTE inhibition was detected in nervous tissues. [98] A case of polyneuropathy was described after massive poisoning with a commercial formulation of parathion (O,O-diethyl O-4- nitrophenyl phosphorothioate). [99] This is in contrast with many other cases of parathion poisoning described in the literature that never resulted in OPIDP. Parathion and its active metabolite paraoxon do not cause OPIDP in hens. [77,100,101] However, the formulation was not analysed for the presence of ethyl bis-(4- nitrophenyl) phosphate, an impurity that inhibits NTE and might occur in commercial formulations. [102] In addition, the clinical picture was further complicated by methanol poisoning. Alonso et al. [103] also attributed to parathion a case of peripheral neuropathy. However, histopathology showed demyelination that was consistent with electrophysiological findings, thus suggesting a neuropathy other than OP-induced axonopathy. In addition no clear acute toxicity was described and the exposure to parathion was not demonstrated. A case of parathion poisoning where the estimated dose was high (about 400 mg/kg) and the cholinergic syndrome was severe, required high doses of atropine and pralidoxime, and artificial ventilation. [104] The patient was comatose for 23 days. On that day, deep tendon reflexes were absent without electrophysiological signs of denervation and with normal motor and sensory nerve conduction velocities. On day 28 electrophysiological signs of a mild distal sensorimotor polyneuropathy in the lower limbs both of axonal and demyelinating type were reported, but no details were given. The patient fully recovered by day 56. The quick recovery does no support a diagnosis of OPIDP. Aygun et al. [47] described a case of OPIDP after parathion poisoning, but no analytical evidence of the causative agent and no clinical details were given. A slowly progressive peripheral neuropathy in two brothers was attributed to repeated occupational exposure to several OPs, including parathion and chlorpyrifos. [105] However, the absence of an acute cholinergic syndrome, the progression of the disease and the predominant upper limb involvement do not support the diag- nosis of OPIDP. A case of Guillain-Barrè syndrome was described [106] and at- tributed to accidental exposure to merphos (tributyl phosphotrithioite). Merphos cannot be considered the causative agent because the neuropathy developed 4 days after exposure, and OPIDP is an axonopathy and not a demyelinating disease. Himuro et al. [107] reported on a case of a man who was poisoned by the nerve agent sarin (O-isopropyl methylphosphonofluoridate) during the terrorist attack on the Tokyo subway in 1995. He had severe poisoning, was comatose and maintained on assisted venti- lation and total parenteral nutrition for 15 months until he died of pneumonia. At autopsy, the sural, and less so, the sciatic nerves, showed reduction of mainly large myelinated fibres and denervated Schwann cell subunits. In the spinal cord there was total loss of myelinated fibres, except in the posterior columns. In the brain, severe hypoxic-ischaemic lesions were observed. No electrophysi- ological data were available. Although, the pathology is sugges- tive and sarin has caused OPIDP in hens, [108] the information provided by this case is not sufficient to confirm that sarin caused OPIDP in humans. 7. Long-Term, Low-Level Exposures to Organophosphorus Esters and Neuropathy Several observational studies on individuals with varying long- term, low-level exposures to OPs including occupational expo- sures such as those occurring in farmers who dip sheep, [109,110] and exposures of military during the first Gulf War [111,112] have been performed with the aim of detecting mild peripheral neuropathy or

46 Lotti & Moretto Table II. Long-term, low-level exposures to organophosphorus esters and peripheral nerve studies: concerned insecticides Insecticide IUPAC name Reference Azinphos-methyl S-(3,4-dihydro-4-oxobenzotriazin (d)-(1,2,3)-triazin-3-ylmethyl 119,120 Bromophos O-4-bromo-2,5-dichlorophenyl O,O-dimethylphosphorothioate 121 Chlorfenvinphos 2-chloro-1-(2,4 dichlorophenyl)-vinyl diethyl phosphate 122 Chlorpyrifos O,O-diethyl O-3,5,6-trichloro-2-pyridyl phosphothioate 119-123 Diazinon O,O-diethyl O-2-isopropyl-6-methylpyrimidin-4-yl phosphorothioate 109,119-121,124 Dicrotophos cis-3-(dimethoxyphosphinyloxy)-n,n-dimethyl crotonamide 125 Dimethoate O,O-dimethyl S-methylcarbamoylmethyl phosphorodithioate 124,126 Dimethylvinphos (Z)-2-chloro-1-(2,4,5-trichlorophenyl) vinyl dimethyl phosphate 125 EPN O-ethyl O-4-nitrophenyl phenylphosphonothioate 124 Fenthion O,O-dimethyl O-4-methylthio-m-tolyl phosphorothioate 127 Leptophos O-4-bromo-2,5-dichlorophenyl O-methyl phenylphosphonothioate 124 Malathion Diethyl (dimethoxythiophosphorylthio) succinate 121,124,126,128 Methamidophos O,S-dimethyl phosphoramidothioate 124 Methyl parathion O,O-dimethyl O-4-nitrophenyl phosphorothioate 122 Mevinphos 2-methoxycarbonyl-1-methylvinyl dimethyl phosphate 129 Monocrotophos Dimethyl (E)-1-methyl-2-(methylcarbamoyl) vinyl phosphate 123,126 Parathion O,O-diethyl O-4-nitrophenyl phosphorothioate 122,123 Phentoate S-α-etoxycarbonylbenzyl O,O-dimethyl phosphorodithioate 124 Phosphamidon 2-chloro-2-diethylcarbamoyl-1-methylvinyl dimethyl phosphate 119,120 Phosmet O,O-dimethyl S-phtalimidomethyl phosphorodithioate 119,120 Thimet O,O-diethyl S-(ethylthio)methyl phosphorodithioate 128 Trichlorphon Dimethyl 2,2,2-trichloro-1-hydroxyethylphosphonate 128 IUPAC = International Union of Pure and Applied Chemistry. changes in peripheral nerve functions. Most of these studies have of workers exposed at levels of DEF (S,S,S-tributylphosphobeen recently reviewed. [113] In general, these studies did not accu- rodithioate) causing lymphocytic NTE inhibition, [130] in a group of rately assess exposure and sometimes even the identification of the spray operators exposed to OPs whose identity was not reportcompounds was missing. [114-118] Reported changes in peripheral ed, [131,132] in workers exposed to azinphos-methyl, and possibly nerves were usually mild, inconsistent and unexplained. Some- phosmet and methyl parathion, [133] and a recent, carefully conducttimes they were reported to be reversible, sometimes irreversible, ed prospective cohort study in workers exposed to chlorpyrifos because the features were observed a long time after cessation of where exposure was assessed by urinary excretion of the exposure. The significance of these changes, if real, remains chlorpyrifos metabolite 3,5,6-trichloro-2-pyridinol and by mild unclear. In addition, the lack of histopathology, of follow-up data, effects on plasma butyrylcholinesterase. [134] and of an experimental model for peripheral nerve changes that seem different from classic OPIDP are also of concern. Electrophysiological results were usually examined together on a 8. Conclusions group basis and correlation with clinical data was almost always OPIDP is a very rare toxicity. The reasons include the developmissing. Finally, a most important aspect was missing from the ment of insecticides that cause OPIDP at doses higher than those authors evaluation of these results, which is that these pesticides causing severe cholinergic toxicity after either single or repeated are far better inhibitors of AChE than NTE. Pesticides cause exposures, and the removal of neuropathic triarylphosphates from peripheral neuropathy at exposure doses that inevitably cause commercial mixtures of non-anticholinesterase OPs. Nevertheless, cholinergic toxicity. Insecticides involved in the studies are listed some cases may still occur after massive exposures to certain in table II. insecticides, such as those occurring after a suicidal attempt. Nevertheless, in some of these studies, no effects on peripheral Overwhelming clinical and experimental evidence indicates nerve function have been found. These include studies in a group that not all OPs cause OPIDP. Consequently, whenever cases of