Association of Pesticide Exposure With Neurologic Dysfunction and Disease

Freya Kamel; Jane A. Hoppin


Environ Health Perspect. 2004;112(9) 

In This Article

Neurotoxicity of Low-Level Exposure

Findings from studies of acute exposure to moderate levels of pesticides are inconsistent. Some studies of well-trained and -equipped pesticide applicators in the United States reported that exposure to OPs sustained during a single work shift (Maizlish et al. 1987) or assessed using a short-lived urinary biomarker (Dick et al. 2001) was associated with little neurotoxicity. However, several studies in developing countries, where exposures may have been higher, found that acute exposure to OPs was associated with increased symptom prevalence in commercial applicators (Misra et al. 1985) and farmworkers (London et al. 1998; Ohayo-Mitoko et al. 2000). Acute and chronic exposures are often correlated, sometimes making it difficult to separate their effects. The following discussion focuses on the effects of chronic exposure to moderate levels of pesticides, although in many studies acute exposure may also have occurred. Several types of neurologic end points are considered, including symptom prevalence, neurobehavioral performance, sensory and motor dysfunction, and direct measures of nerve function. Studies are summarized in Table 2 .

Studies of symptom prevalence are often based on variations of an established checklist (Lundberg et al. 1997) and evaluate a broad range of symptoms, including headache, dizziness, fatigue, insomnia, nausea, chest tightness, and difficulty breathing as well as symptoms suggesting cognitive (confusion, difficulty concentrating), motor (weakness, tremor), and sensory (numbness, tingling, visual disturbance) dysfunction. Pesticide exposure is associated with increases in prevalence of many symptoms, with little evidence for specificity. Most studies have focused on OPs; most of these found an association of exposure with increased symptom prevalence. Farmworkers (Gomes et al. 1998), greenhouse workers (Bazylewicz-Walczak et al. 1999), and factory workers (Bellin and Chow 1974) exposed to OPs reported increased symptom prevalence compared to unexposed workers. In particular, farmers and farmworkers who applied OPs had higher symptom prevalence than nonapplicators (London et al. 1998; Ohayo-Mitoko et al. 2000; Smit et al. 2003), as did commercial applicators (Misra et al. 1985; Steenland et al. 2000) and sheep dippers (Pilkington et al. 2001). Pesticides other than OPs also affect symptom prevalence: one study found that exposure to dichlorodiphenyltrichloroethane (DDT) was associated with increased symptom prevalence (van Wendel de Joode et al. 2001), as did one study of fumigants (Anger et al. 1986) although not another (Calvert et al. 1998). Additional studies have evaluated changes in mood and affect, using either self-report or validated scales. Workers exposed to OPs (Bazylewicz-Walczak et al. 1999; Steenland et al. 2000; Stokes et al. 1995) or DDT (van Wendel de Joode et al. 2001) reported higher levels of tension, anger, or depression on standard symptom questionnaires, and OP applicators showed elevated levels of anxiety on personality tests (Levin et al. 1976). Three studies found no association of OPs with symptom prevalence or affect (Ames et al. 1995; Fiedler et al. 1997; Korsak and Sato 1977).

Increased symptom prevalence was correlated with inhibition of erythrocyte AChE in four studies of OP exposure (Bellin and Chow 1974; Gomes et al. 1998; Leng and Lewalter 1999; Ohayo-Mitoko et al. 2000) and with inhibition of both erythrocyte AChE and plasma cholinesterase in two of these (Bellin and Chow 1974; Leng and Lewalter 1999). Another study found no relationship of symptom prevalence to inhibition of either erythrocyte or plasma cholinesterase (Lee et al. 2003). One study found that increased symptom prevalence was associated with self-reported pesticide exposure but not with depressed erythrocyte AChE activity (Ciesielski et al. 1994). Effects of OP exposure may not necessarily be caused by AChE inhibition (Pope 1999). Further, farmworkers have complex work histories and are likely to be exposed to pesticides other than OPs that may affect symptom prevalence without affecting AChE.

Neurobehavioral test batteries, including the World Health Organization Neurobehavioral Core Test Battery (Anger et al. 2000), the Neurobehavioral Evaluation System (Letz et al. 1996), and portions of other batteries, have been used to evaluate pesticide effects on cognitive and psychomotor function. Tests included in these batteries assess memory, attention, visuospatial processing, and other aspects of cognitive function; commonly used tests include symbol digit, digit span, visual retention, pattern memory, trail making, and others. Most studies indicate that pesticide exposure is associated with deficits in cognitive function. Sheep dippers (Stephens et al. 1995), nursery workers (Bazylewicz-Walczak et al. 1999), and other workers (Korsak and Sato 1977) exposed to OPs, malaria-control workers who sprayed DDT (van Wendel de Joode et al. 2001), vineyard workers exposed to fungicides (Baldi et al. 2001), fumigators exposed to sulfuryl fluoride but not those exposed to methyl bromide (Anger et al. 1986; Calvert et al. 1998), and farmers (Cole et al. 1997), farmworkers (Gomes et al. 1998; Kamel et al. 2003), and pesticide applicators (Farahat et al. 2003) exposed to multiple pesticides all performed worse on tests of cognitive function. There are some inconsistencies among these studies. Although most studies found deficits on one or more tests of cognitive function, different tests were affected in different studies, and a few studies found no relationship of OP exposure to any test (Ames et al. 1995; Daniell et al. 1992; Fiedler et al. 1997; Rodnitzky et al. 1975; Steenland et al. 2000).

Deficits in psychomotor function could be caused by impairment of sensory input, motor output, or associative delays; tests used include reaction time, tapping, pursuit aiming, Santa Ana and other pegboard tests, and others. Most studies indicate that pesticide exposure is associated with deficits in psychomotor function. Farmworkers (Daniell et al. 1992; London et al. 1997), farmers (Fiedler et al. 1997) and termiticide applicators (Steenland et al. 2000) exposed to OPs, malaria-control workers who sprayed DDT (van Wendel de Joode et al. 2001), vineyard workers exposed to fungicides (Baldi et al. 2001), fumigators exposed to methyl bromide or sulfuryl fluoride (Anger et al. 1986; Calvert et al. 1998), and farmworkers with multiple exposures (Gomes et al. 1998; Kamel et al. 2003) all showed worse performance on tests of psychomotor function. Again, results for individual tests were not fully consistent within or among studies, and no change in psychomotor function was evident in two studies of OP exposure (Ames et al. 1995; Cole et al. 1997).

Neurobehavioral test batteries are often supplemented with tests of sensory or motor function. One frequently used test is vibration sensitivity, which evaluates peripheral somatosensory function. Most available evidence suggests this is not affected by moderate pesticide exposure. One study of farmers exposed to OPs found decreased sensitivity (Stokes et al. 1995), and another of farmers exposed to multiple pesticides found both decreased sensitivity and other signs of peripheral neuropathy (Cole et al. 1998). However, other studies of individuals exposed to OPs (Ames et al. 1995; London et al. 1998; Pilkington et al. 2001; Steenland et al. 2000), DDT (van Wendel de Joode et al. 2001), fumigants (Anger et al. 1986; Calvert et al. 1998), or multiple pesticides (Kamel et al. 2003) found no relationship of exposure to vibration sensitivity or other measures of somatosensory function.

Few studies have evaluated other aspects of sensory function. One study suggested that the sense of smell was not affected by OPs (Steenland et al. 2000); another study suggested a relationship with fumigants (Calvert et al. 1998). Visual contrast sensitivity was not affected by exposure to OPs (Steenland et al. 2000; van Wendel de Joode et al. 2001) or multiple pesticides (Kamel et al. 2003), but color vision was (Steenland et al. 2000). Retinal degeneration was associated with fungicide exposure in a case-control study of licensed pesticide applicators (Kamel et al. 2000). In general, these data are too limited to draw conclusions about the relationship to pesticide exposure to sensory function.

Similarly, few studies have considered motor function, and few inferences can be made about its relationship to pesticide exposure. Tremor was related to exposure to multiple pesticides in one study (Davignon et al. 1965) but not to OPs in two others (London et al. 1998; Steenland et al. 2000). Grip strength was not related to exposure to fumigants (Anger et al. 1986), DDT (van Wendel de Joode et al. 2001), or multiple pesticides (Kamel et al. 2003).

Balance is an integrated sensorimotor function. An early study found deficits in balance in apple farmers exposed to multiple pesticides (Davignon et al. 1965). In modern studies, balance is commonly evaluated by a test of postural sway; varying the conditions of the test may indicate whether impaired balance is related to deficits in visual, proprioceptive, or vestibular input. Three studies of individuals exposed to OPs (Steenland et al. 2000) or to multiple pesticides (Kamel et al. 2003; Sack et al. 1993) found that impaired postural sway was associated with exposure, but effects were small and another study found no relationship of OP exposure to postural sway (Ames et al. 1995). Effects were most evident when both visual and proprioceptive inputs were removed, suggesting that vestibular function may be affected (Kamel et al. 2003; Sack et al. 1993).

Studies that have evaluated peripheral nerve conduction have produced largely negative results. Several studies of OPs found little evidence of impaired nerve conduction (Ames et al. 1995; Engel et al. 1998; Steenland et al. 2000). One study of fumigators found deficits in nerve conduction (Calvert et al. 1998), but another did not (Anger et al. 1986). In contrast, fungicide exposure was related to impaired nerve conduction in a study of bulb farmers, which also found deficits in autonomic nerve function (Ruijten et al. 1994). One study found changes in electroencephalogram (EEG) associated with OP exposure (Korsak and Sato 1977).

Three studies have performed clinical neurologic examinations in a subset of individuals identified by field studies as having deficits related to OP exposure. Beach et al. (1996) studied sheep dippers with increased symptom prevalence (Stephens et al. 1995); Horowitz et al. (1999) studied apple farmers with decreased vibration sensitivity (Stokes et al. 1995); and Jamal et al. (2002) studied sheep dippers with peripheral neuropathy (Pilkington et al. 2001). In general, clinical examination confirmed the results of the field studies, although clinically recognizable neurologic abnormalities were minor and not present in all individuals identified by the field studies.

Individual response to pesticide exposure may be affected by polymorphisms in genes affecting pesticide metabolism. The best-known example is paraoxonase, an enzyme that hydrolyzes active metabolites of OPs (Costa et al. 2003). Animal studies suggest that changes in serum paraoxonase activity alter susceptibility to OP toxicity (Costa et al. 2003). In humans, paraoxonase polymorphisms affect the relationship of OP exposure to both erythrocyte AChE inhibition and symptom prevalence (Lee et al. 2003; Leng and Lewalter 1999; Mackness et al. 2003; Sozmen et al. 2002). Although Costa et al. (2003) have suggested that adequate evaluation of susceptibility requires measuring serum paraoxonase activity as well as genotype, recent population-based studies have suggested that the discrepancy between genotype and phenotype is relatively small and that nongenetic factors contribute relatively little to variation in serum activity (Ferre et al. 2003; Vincent-Viry et al. 2003).