Applications of neurophysiological methods in occupational medicine A

neurophysiological review. Scand. j. environ. & health. (1975) 1-14. Key wonls: occupational diseases, neurophysiological methods, electroneuromyo graphy, electroencephalography, toxic neuropathy, toxic encephalopathy.

Electroencephalography (EEG) is the recording of the electrical activity of the brain. This low-voltage activity is amplified with a specific apparatus ( fig. 1) and is registered on paper. The EEG apparatus has at least 8 channels, but 16 channels are also widely used. According to the international 10-20 system 19 electrodes are placed on anatomically-fixed regions of the scalpI and one electrode is attached to each ear lobe. Resistance is lowered by rubbing the skin with alcohol and using an electrode paste under the electrodes, which are glued on or kept in place with a rubber net. Varying electrode combinations, i.e. I montages, are then recorded during the next 30 minutes, which gives an ample possibility Neurophysiological methods reveal certain types of lesions within the nervous system. Electroeneephalography shows involvement ,of the central nervous system, especially dysfunction of the more superficial structures of the brain, and electroneuromyography differentiates lesions at different levels of the peripheral nervous system and muscles. The myoneural junction can be studied with repetitive stimulation of the peripheral nerves.
Involvement of the nervous syst,em can be observed even in modern industrial plants, and, for example, in Finland occupational diseases with symptoms in the nervous system form the third largest group of oecupational diseases. (They follow occupational skin diseases and hearing loss caused by noise.) Nervous system symptoms are at times vague, and it may be hard to find specific signs in purely clinical examinations. Early neurological symptoms have not been adequately taken into account in setting threshold limit values (TLVs).
The classical symptom of lead poisoning, paresis of the radial nerve leading to drophand, praetically helongs tD history in modern industrialized countries. Sim,UarilYl grave neurological symptoms of carbon disulfide poisoning are not met any more. But some workers Btill complain of numbness and pain in their limbs 1 headache, and dizziness, These early nervous symptoms can be studied with careful neurological examination 1 but various neurophysiologioal methods should also be added in order to gain sensitivity and objective proof ,of lesions at an early stage. for examining different areas of the brain since several regions are ,always studied at the same time. Activation methods include photic stimulation, trains of short pulses of intensive light with certain frequencies; hyperventilation for 3 minutes; and sleep, either natural or induced with the help of sedatives, e.g., chloral hydrate.
Specifically trained technicians register the electroencephalograms (EEGs) under the supervision of ·a physician, Le., neurophysiologist. The neurophysiologist then interprets the EEGs according to well-documented principles (23, 33).
Different types of EEG abnormalities can be indentified. The most usual ones are slow wave abnormalities of varying degrees. They can be focal, Le., localized in a certain area only, or diffuse or generalized. Slow wave abnormalities are connected to dysfunctions of nerve cells and brain structures or to anatomic lesions. They reveal encephalopathy of different originstoxic, traumatic, 'metabolic, circulatoryor structural-alterations like malformations 2 or tumors or other expansive lesions. Increased intracranial pressure can also be revealed by slow waves.
Other EEG abnormalities include paroxysmal abnormalities, namely, sharp waves, spikes, and spike and wave discharges. Also these can be focal or generalized. Paroxysmal abnormalities are often seen in subjects with epileptic convulsions, but they can be found in the presence of other types of brain irritation too. For instance they are common in children with mental retardation or cerebral palsy, even without convulsions.
A good waking record ,is needed to gain general information on encephalopathy, but the amount of information can be increased by using different activation methods. Hyperventilation can activate slow wave abnormalities and is thus useful. In certain toxic encephalopathies it is important to look for possible irritative lesions; then photic stimulation and sleep are good auxiUary activation methods. Photic stimulation and hyperventilation should both be mcluded in every EEG exam-ination because ,of differentiai diagnostic purposes, Natural sleep during a part of the recording can also be encouraged, but for ambulant patients sleep induced by drugs can be limited to the diagnostic problems of epilepsy, An EEG mainly reflects the state of the brain at present. Therefore actively progressive diseases give severe EEG abnormalities, while slowly changing states like non progressive cerebral atrophy may le,ave the EEG intact. Deep lesions may well escape unnoticed because the activity diminishes in the distance between them and the scalp, An EEG examination is indicated in suspected cases of occupational diseases which involve symptoms of the eentral nervous system, such as headache, poor memory, dizziness, and loss of muscular strength or tactile sense of a cerebral type, for example, hemisyndromes. It is of coUrse also applioable in the differential d,iagnosis of neurological diseases like epilepsy, brain .tumors, and brain injuries and can at times give prognostic hints.

ELectToneuromyography
Electroneuromyography (ENMG) includes both electromyography (EMG) and the measurement of nerve conduction velocities. Diagnostic EMG is performed with concentric or bipolar needle electrodes inserted in appropriate muscles. The electrical activity of the muscles at rest and during a voluntary contraction is amplified with an electromyograph (fig, 2), The EMG activity is usually observed on the screen of an oscilloscope, and at the same time it can be heard throngh loud speakers. The human ear is good at separating the different sound frequencies in the spectrnm of the EMG, EMG activity can also be photographed either on moving paper Or with a polaroid camera. For later analysis it can also be recorded by tape recorders with a very large frequency spectrum, The evaluation of EMG activity requires much training and experience and can be performed only by specialized physicians, i.e., neurophysiologists (28).
The electromyograph is also used to measure nerve conduction velocities. It is then combined with an -electrical stimulator ( fig. 2), which triggers ihe electromyograph at the time of the stimulus. Maximal motor conduction velocities (MCV) of the peripheral nerves are fairly easy to measure (64). The nerve is stimulated with a skin electrode at two separate spots along the course of the nerve, and the motor responf::.'€ is picked up with skin electrodes and amplified with the ·electromyograph. The latencies from the stimulus to the response are measured, and the MCV is calculated by dividing the distance between the stimulation points by the difference of the latencies.
Different nerves conduct the impulses at different speeds. The MCVs of the arm nerves range from 50 to 65 m/sec, and those of the leg nerves from 40 to 55 ml sec. The MCV of the deep peroneal nerve is usually higher than that of the posterior tibial nerve.
The age of the subject and the temperature of the nerve affect the MCV. Most MCVs slow down by about 1 mlsec for every 10 years of 'age between 20 and 60 years. The lower the temperature of the nerve, the slower the nerve conduction velocity. The skin temperature at the stimulation point is easily measured, and to ensure reliable results it should be above 30°C -usually it is between 31.5 and 33.5°C.
Series of normal values of MCVs, taking age into account, have been publi-shed (64), but each laboratory should have its own. In research projects control series matched by age are ·of utmost importance.
If there is a question of disease or health, the controls should also be neurologically and medically healthy persons. If hazards of a certain occupational exposur-e are studied epidemiologically, the control subjects should probably not be prescreened neurologically, however.
More sensitivity can be added to nerve cunduction studies by measuring the amplitude ratios of responses elicited by stimulation at different portions of the nerve, as well as by comparing the duration of 4 the evoked muscle action potentials. Slower motor fibers can be ·studied separately by using paired stimuli based on the antidromic blocking technique (27,56).
Sensory conduction velocities (SCV) can be measured orthodromically or antidromically with skin or needle electrodes. One practical method involves stimulating the finger nerve with skin electrodes and picking up the tiny nerve potential with skin electrodes 'on two separate spots along the course of the nerve. A series of amplified nerve potentials is then averaged in an analog-digital computer ( fig. 2), after which latencies are measured and SCVs calculated.
Patients with myasthenia or myastheniform states exhibit abnormal fatigability because of disturbances in the myoneural junction. This abnormality can be shown electrophysiologically by observing the change l usually decline, in the amplitude of muscle action potentials elicited by repetitive stimulations of a motor nerve.
Muscle diseases appear in EMG as a decrease in the amplitude and duration of the motor unit potentials. Neurogenic diseases of the lower motor neuron type increase the duration and at times the amplitude of motor unit potentials and induce spontaneous activity in EMG in the form of fibrillations, fasciculations, or positive monophasic potentials. The number of recruitable motor units diminishes in neurogenic diseases, and thus during a maximal voluntary contraction only single motor units or a mixed pattern is achieved, instead of the normal interference pattern. If the neurogenic disease involves peripheral nerves, nerve conduction velocities are slowed. Conduction velocity of the slower motor fibers decreases earlier than the MCV. The neuropathy may be mainly sensory, however. Then only SCVs exhibit slowing. Thus with the help of ENMG it is possible to differentiate the levels and sites of peripheral neurogenic and myogenic diseases.
ENMG is indicated for patients with symptoms of disturbances in the peripheral nervous system, namely, flaccid paresis, muscular atrophy, muscular weakness, paresthesia and pain in the limbs, certain types of sensory loss, and ·abnormal fatigability. Even in cases with mild, mainly

Othe,' physical factors
It is possible that extensive exposure to noise causes changes in the brain and affects the EEG. Sudden aiarming noise has been reported to induce specific EEG changes, the period of which corresponded well with the time during which the tools. Damage of both the sensory and motor fibers has been reported for rock drillers and users of other pneumatic tools (15,41), as well as for lumberjacks (53). The vibration frequency of the power saw has proved to be especially dangerous to the peripheral nervous system (31). One field study revealed frequent abnormalities in the conduction velocity of the slower motor fibers (CVSF) of the ulnar nerve and also in the distal m-otor fibers of the median nerve ( fig. 3) among lumberjacks (54). Also Lukas and Kuzel (4i) observed elongation of the dis"al latency of the median nerve in users of vibrating tools. They interpreted this observation as a sign of the carpal tunnei syndrome. I myself have found the combination of lesion in the distal portion of the median nerve and partial damage in the ulnar nerve to be quite common in the clinical vibration syndrome. Similarily, slowing in the distal portion of the sensory fibers of the arm nerves has been shown (36). A specific etiological diagnosis based on neurophysiological studies alone is impossible in individual cases of suspected 'Occupational disease. Usually sensitive methods give a greater number of false positive results than methods with low sensitivity. However, when exposure to a toxic agen t can be documented and other evidenli causes for nervous system lesions can be excluded, positive neurophysiological findings give reasonable support for the diagnosis of occupational disease in the clinical work-up of patients with m'ild or nonspecific symptoms. The specific distribution of the lesions ean also indicate the role of a certain exposure.

OCCUPATIONAL SOURCES OF NEUROLOGICAL SYMPTOMS
Certain physical and chemical agents causing occupational diseases with neurological symptoms are discussed in this section with special reference to neuropbysiological findings.
Physical factors may induce neurological symptoms. For instance vibrating tools cause the vibration syndrome, which consists of arteriolopathy and neuropathy. Workers complain of cold and white fingers and of numbness and pain in the extremities exposed to local vibration. Wallerian degeneration has been shown in the peripheral nerves of animals exposed to vibration (3).
Several studies have appeared on the decreased vibration sensitivity of workers using vibrating tools (48). However, Lidstr6m (39) did not find permanent changes in the threshold of vibration 'Sensitivity of 30 workers occupationally exposed to vibrating tools. The temporary threshold shift (TTS) to vibration sensation immediately after vibration exposure was, however, significantly greater among the lumberjacks than am-ong the controls in Lidstrom's series (39).
ENMG examinaHans have revealed signs of neuropathy, i.e., damage in the peripheral nerves of workers who use vibrating subject was unable to take immediate, adequate steps to avoid an accident (52). Accidents have been reported to occur under such circumstances. Knowledge of EEG changes induced by long-term exposure to ,intensive noise are lacldng, however. An lbalian group has briefly reported that subjects with a mean exposure of 12 years to noise showed desynchronized, labile EEG traces (4).
Another physical factor which has been blamed for causing adverse effects in the nervous system is microwave radiation. Some studies describe electroencephalographic changes) namely, decreases in the alpha wave index and increases in the theta and delta wave percentages, as cited by Silverman (63). Well-controlled electrophysiological studies of persistent lesions in the nervous system after microwave exposure are, however, lacking.

Insecticides
Chlorinated hydrocarbon insecticides are basically neurotoxins. Hoogendam et al. (25) studied the EEG of 122 workers occupationally exposed to chlorinated hydrocarbon 'insecticides and found an increase in abnormal EEGs (20.5 "/0 against 9 0 /0 among controls). Bilateral spike and wave discharges were noted in all cases that resulted in epileptiform convulsions. The EEGs returned to normal wi thin 2 to 6 months ,after -the termination of exposure. Later the sam'e research group published a 9-year follow-up Df the workers in an insecticide plant (26). During this period 17 workers, 5 0/0 of the total labor force, suffered from epileptic convulsions. Most of them had typical EEG abnormalities when recordings were tak·en within 2 weeks of the convulsions. All the most recent cases of convulsions occurred in connection with unusually high exposure. As a consequence the 'authors submit the workers to EEG studies twice yearly. A similar prevalence (22°/0) of abnormal EEGs among workers with exposure to chlorinated hydrocarbons has been found in Israel (43).
Organophosphorus pesticides at the exposure levels most common today do nort induce measurable decreases in the blood cholinesterase activity, and it is impossible 6 to measure their actual concentration in the blood or tissues. However, it has been shown electromyographically (14,32) that organophosphorus pesticides cause a functional abnormality in the neuromuscular junction similar to that noted in the overtreatment of myasthenic patients. Namely, after daily pesticide exposure part of the workers exposed in a formula'iion plant and in agriculture exhibited a reduction in the amplitude of ,the first evoked muscle action potential of a short train of stimuli applied after a short voluntary muscle contraction. This phenomenon has been considered mild in degree and only suggestive of a necessity to improve workers' protection.
Only a few reports have listed persistent peripheral neuropathy ,or psychological defects of central nervous system function in connechon with long-term occupational exposure to organophosphorus insecticides (45). The neurophysiological studies of workers in pesticide exposure mainly deal with acute eUects) and therefore studies on long-term low-level exposure are lacking. As an international workshop (16) concluded) more studies and new methodology are needed. Also Kimura (34) has stressed the need of 'more sensitive electrodiagnostic methods for detecting subclinical effects.

CaTbon monoxide
Acute carbon monoxide (CO) poisonig may lead to comas of various depths depending on the length and intensity of exposure to CO. During a CO-induced coma and shortly after it EEG abnormalities are common (20). They usually consist of slow waves prom'inently on the frontal areas. Revol et al. (49) described specific triphasic wave forms in some cases of com,atose CO poisoning. The wave form looked similar to that noted in hepatic comas, sometimes called »pseudoparoxysmal spike and wave)) or slow spike and wave. The authors thought that it was based on metabolic disturbances and considered it distinguishable from paroxysmal epileptiform discharges. Actually one EEG recording published by Geier et al. (20) had similar features. Epileptic convulsions do not belong to the picture of acute CO poisoning, and EEG abnormalities disappear with time after an 'acute episode.
The existence of true chronic CO poisoning is a matter of dispute. Chronic symptoms probably originate from repeated attacks of acute CO poisoning, which might occur in certain work places l e.g., in foundries or garages. Neurophysiological studies on patients from such places are scarce. One report claimed that a policeman exposed to CO in heavy traffic began to suffer from seizures of the psychomotor type and also showed focal spike discharges in his EEG ,and that these abnormalities were due to chronic CO poisoning (21). Both the man and his EEG became symptomless when CO exposure ceased. In short-time exposure experiments with human volunteers changes in the visual evoked response appeared at a carboxyhemoglobin level greater than 20 % while carboxyhemoglobin levels ·approaching 33 % did not alter gross spontaneus EEG activity (30).

ACTylamide
Whenacrylamide was introduced l several workers in the first plant using it developed numbness of the fingers and weakness and unsteadiness of the legs. Because of these symptoms Full-erton and Barnes (19) conducted electrophysiological and hi<stological studies on rats given repeated doses of acrylamide. Concomitantly with the appearance 'of m,ajor clinical symptoms the motor conduction velocity -of the sciat1.c nerve was significantly slowed. A similar slowing 'Of nerve conduction velocity, as well as a decreased excitability 'and decreased amplitude of the evoked muscle potentials, ha's been shown in rats and monkeys (38) and in the baboon (29). Fullerton (18) described electrophysiological findings in human acrylamide poisoning. A slowing of maximal motor conduction velocity was rare, but dispersion of the muscle action potential to nerve stimulation was more striking. She especially noted a disproportionate slowing -of conduction in the distal parts of the slower motor fibers. Japanese authors found electrophysiological abnormalities in the sensory conduction of workers handling acrylamide, mainly in the distal portions of the nerve, while disturbances in the motor nerves were slight and rare (69). They also observed a decrease in the amplitude of the muscle potential after stimulation of the motor nerve. Slight EMG abnormalities 'of a neurogenic type were reported for 7 of the 15 cases studied. Furthermore, slight EEG abnormalities occurred in three workers with complaints of dizziness and ataxic gait.

Lead
Improvements in occupational hygiene have abolished the occurrence of lead palsy in modern industry, but several studies have shown that subclinical lead neuropathy stil! exists. A reduction in the MCV of the ulnar nerve has been observed in som·e patients with lead poisoning, but without clinical neurological symptoms (62).
Since it has been shown in animal experiments (17, 51) that segmental demyelination can be found in some nerve fibers only, more refined techniques than measuring the MCV are needed in occupational medicine. More sensitivity was gained by Cattonet aJ. (8), who showed a dimini,shed ratio of the muscle action potential amplitudes following stimulation of the lateral popliteal nerve at the knee and at the ankle. This dim'inished ratio represented temporal dispersion in the muscle action potential and thus a slowing in some fibers of the nerve in question. The same phe-nom~enon, slowing in a part of the nerve fibers, namely, in the slower fibers (CVSF) of the ulnar nerve has been demonstrated ( fig. 4) in patients with lead poisoning at the Institute of Occupational Health in Helsinki as weI! (56). We also found some slowing in the MCVs of the arm nerves when a group of patients with lead poisoning was compared to a control group. A mOre distinct abnormality was found in the motor distal latency of the median nerve ,of patients with lead poisoning. A discriminant function analysis proved tha t the CVSF of the ulnar nerve and the motor distal latency of the median nerve formed the best combination for separating the cases of lead poisoning from the controls. 7 8 activity have been found in young children with slightly elevated blood lead levels (40-60 pg/IOO ml) (2,13,47).
Dullness, irritability, headache, tremor, loss of memory, etc., have been described as symptoms of adult lead encephalopathy. Even recently lead 'encephalopathy was reported am·ong heavy drinkers who used lead-contaminated whiskey and in cases of inhaled lead fumes (1,74). However, EEG studies have not as yet gained interest among investigators of occupational lead poisoning.

Mercury
In the Minamata disease severe neurological symptoms and signs were common (12). However} in the Iraqi methylmercury catastrophe electrophysiological tests 7 to 8 months after the cessation of methylmercury intake did not reveal abnormalities in patients with mercury concentrations of 800 ng/mlor more in their blood (5). Th€ only positive finding wasmyastheniform failure in the myoneural junction in 2 of the 14 patients studied. The connection 'of this finding to methylmercury in take requires further study.
Preliminary results from our laboratory show a frequent occurrence of a prolonged latency -of muscle action potential after the stimulation of the facial nerve in fishermen with elevated blood mercury levels.
Neuropathological studies of r·ats by Cavanagh and Chen (9) have shown distinct differences in the pattern of neurotoxic degeneration caused by methylmercury as compared to other neurotoxic agents. The degener,ative lesions after methylmercury -intake were almost entirely eonfined to the primary sensory neurons of the dorsal root and trigeminal ganglia and to the granular cells of the cerebellar cortex. Even at an early stage the whole fiber was simultaneously undergoing fragmentation. The motor fibers seemed to be preserved intact. If the lesion caused by methylmercury is located almost exclusively in the sensory fibers in humans too, the lack of distinct findings in ENMG studies so far would be explained.
Only occasional reports of EEG studies in cases of m·etallic mercury exposure have been published. Some authors claim a Recently} our team disc-overed partial damage o.f the ulnar nerve in lead workers whose blood lead levels had never 'exceeded 70 pg/IOO ml during their entire period of 'exposure (mean exposure time: 4.6 years), ascertained by regular monitoring (59). Thus even blood lead levels previõ usly regarded as safe do not necessarily protect the nervous system from s·light damage.
Vasilescu (72) stressed the specific affection of lead to the radial nerve, but his study included subjects who had ingested illicitly distilled alcohol contaminated from leaded pipes. Thus the neurotoxic effect of other "impurities in the alcohol} and of the alcohol itself, cannot be separated from the effects of lead.
In childhood lead poisoning encephalopathic signs of ataxia, hyperirritability, and lethargy, ,as well -as seizures} are common. Psychological changes and the impairment of fine motor function as well as hyper-  5) and the deep peroneal nerves. Prolongation of the motor distal latency of the median and the ulnar nerves among the eS2 exposed workers proved that the distal portions of the nerves in the upper limbs were affected as well.
EEGs have been used to detect encephalop_ athy among workers exposed to es 2 < Styblova and Bolanova (67) found a high percentage of abnormal EEGs, namely! 37 % of 187 workers exposed to es 2 . The prevalence of abnormal EEGs was higher among workers suffering from a temporary or permanent disability as compared to workers who did not require a transfer from exposed work. A Finnish study also showed that 39 Ofo of 54 workers with longterm esz exposure had abnormal EEGs! while in the control group of 50 paper mill workers only 6 abnormal EEGs were found (60). The EEG is thus a suitable tool for demonstrating the toxic encephalopathy caused by es 2 .

HydToCa1'bon solvents
Bine and Zuidema (24) tested the toxicologic-al properties of several hydrocarbon solvents in rats and rabbits with oral administration, vapor inhalation, percutane.. ous administration, aspiration, and skin and eye irritation tests, Acute intoxication high prevalence of abnormal EEGs in chronic poisoning with metallic mercury or with signs of increased mercury absorption (37). A series of nine workers with metallic m,ercury vapor intoxication was reported in Florida (73). Their initial signs and symptoms were tremor) rigidity! irritability, memory disturbances) and) rarely) sensory loss. Five of the patients showed diffuse slowing in their EEGs. Slightly slower nerve conduction velocities on a group basis was noted in the distal portion of the median nerve both in motor and sensory fibers as well as in the MeV of the tibial nerve. Mild neurogenic changes were noted in motor unit potentials in the EMG of ali the patients, and for five of them the change was statistically significant when compared to the controls.
A power spectra analysis of the surface EMG of workers with inorganic mercury exposure revealed an increased band width (10). A retest after 4 to 6 months in a greatly reduced exposure shifted the EMG band width towards a more norm'al value.

Cal'bon disulfide
Several studies have shown a slOWing of the maximal motor conduction velocities of the peripheral nerves in subjects with carbon disulfide (CS,) poisoning (40, 42, 61) 71). EMG investigations have revealed neurogenic abnormalities which have partly been interpreted as signs of damage at the s-pinal cord level, especially in cases wi th normal nerve conduction velocities (61). Some disturbance aiso appeared in the myoneural junction (61), as judged from the fact that abnormal fatigability was demonstrated for several patients with manifest poisoning with the tetanic stimulation test. The nerves of the legs are more often involved in eS 2 neuropathy! as has been shown by Vasilescu (71).
Similar findings of neurotoxicity were found when workers with long-term eS 2 exposure were compared to paper mill workers (60). Both groups were strictly comparable with regard to age, type of work) social background, and other relevant factors, The eS 2 exposed workers, most of whom were actively working and displayed minor complaints only, had significantly slower MeVs in the leg was usually characteriz-ed by a depression of central nervous activity followed by incoordination and varying degrees of coma. Convulsions were occasionally seen. Electrophysiological tests were not applied in this study.
Clinical symptoms in long-term -exposure to organic solvents are often attributable to central nervous involvement. However, EEG studies of workers with solvent exposure are few. Chalupa et a1. (11) published a small series of acute poisonings from CO and industrial solvents; only ten subjects had been exposed to solvents, six of them to trichloroethylene. EEG abnormalities were frequent 'among the subjects, especially among those with solvent exposure. On the other hand, Mitchell and Parsons-Smith (44) stated that the EEG is usually unaffected. The EEGof a Japanese with a tr,ansverse lesion of the spinal cord after accidental exposure to trichloroethylene was reported to be normal also (50).
Benzene compounds induced EEG abnormalities in 40 out of 100 women employed at gluing plastic elements (65). Methylene chloride exposure also caused frequent EEG abnormalities among shoeworkers (22). Preliminary reports from our laboratory (55,57) have shown a high prevalence of EEG abnormalities among workers with occupational exposure to solvents or with chronic solvent poisoning. Among the 240 subjects studied, EEG abnormalities were morefrequ-ent among those with high exposure levels. Affections in the peripheral nervous system were not as common as central nerVDUS involvement, but also polyneuropathic signs in electroneurography increased in number as exposure became higher.
Short-term exposure to 500 to 1,000 ppm of methylene chloride induced changes in the visual evoked response from two of three volunteers (66), the findings were interpreted as signs of central nervous system depression.
Trichloroethylene or some of its breakdown products have caused cranial polyneuropathy, most marked in the fifth cranial nerve, in workers in industry (7,44,70). Similar case reports have appeared earlier in connection with general anesthesia of trichloroethylene and soda lime in a closed circuit as cited by Buxton and Hay-10 ward (7). These findings indicate that electromyography of facial muscles should be employed in the study of the neurotoxic effects of trichloroethylene.
Japanese authors have shown that n-hexane, especially as a component of glue, causes polyneuropathy. At lower exposure levels (up to 900 ppm) the polyneuropathy is predominantly sensory. With increasing concentrations the motor nerves are also involved and, especially among sniffers l motor polyneuropathy with muscular atrophy is severe. The sensory disturbances were few if any in the four cases reported in the Japanese literature (68). Some Italian bootmakers, working at home or in small factories, also showed motor polyneuropathy and abnormal EMG and MeV findings (6).
The voluntary} repeated inhalation of commercial lacquer thinner caused severe peripheral neuropathy in seven young male sniffers (46). Three of them were completely paralyzed} including bulbar involvement. Nerve conduction velocities were greatly decreased. The authors thought that methyl-amyl-ketone was the component of the thinner responsible for the severe symptoms.
The most severe polyneuropathic symptoms connected with hydrocarbon solvents seem to be caused by high concentrations either in accidental exposure or during sniffing.
Styrene is used in the manufacture of fiber glass products. It belongs to the organic solvent group and also produces neurotoxic symptoms. EEG abnormalities and an increased tendency to sleep have been noted in styrene exposure in Czechoslovakia (35). At the Institute of Occupational Health in Helsinki we have also demonstrated EEG abnormalities, the prevalence of which increased with increasing mandelic acid concentrations in the urine of the workers (58). Peripheral nervous lesions could not be connected to styrene exposure in our study, however.

CONCLUSIONS
The most common lesion of the peripheral nervous sy-stem caused by an occupational agent is polyneuropathy. In many instances the neuropathy is more striking in the distal portions of the nerves, as in the case of lead-, acrylamide-, CS 2 -and vibrationinduced neuropathy. Only alkyl mercury neuropathy distinctly involves the whole length of the sensory axon at an early stage. In most cases neuropathy at current exposure levels is mild or subclinical; this is displayed well by the fact that more sensitive and more refined techniques are being developed in order to detect the effects of neurotoxic substances.
Myopathy caused by industrial poisonings has not been reported, but disturbances of myoneural juction are found in insecticide exposure and in CS 2 poisoning.
Encephalopathy is no doubt a potential danger, especially in exposure to various solvents~CS 2 , styrene or other hydro-carbons~and perhaps in chronic CO exposure. A combination of EEG abnormalities and signs of peripheral neuropathy is suggestive of effects of a neurotoxic substance.
EEG abnormalities with clinical significance are found not only after workdays but also during periods without exposure to toxic chemicals. Within a period of a few months without exposure, EEG abnormalities tend to disappear in most eases, if the toxic effect has not been long-acting and if encephalopathy has not been severe. Normal EEGs in connection with symptoms of the psycho-organic syndrome or dysfunction of the autonomous nervous system are found in a fair number of people, especially when encephalopathy is slowiy progressive and mainly involves the deep brain structures. Thus ·a normal EEG does not rule out toxic encephalopathy.
Neurophysiological methods in occupational ·medicine are useful in the .following aspects: 1) In the study of neurotoxicity of known or new chemicals or the neurological involvement caused by physical factors. Strictly controlled epidemiologic studies of humans or well-planned animal experiments are the most informative. However, the knowledge derived from animal experiments cannot be -directly applied to man, since certain speciesspecific variations in effects are common.
2) In the setting of safety norms. Ner-VOllS involvement must always be consid-ere-d unaJG.ceptable, and, when such signs are present, exposure levels should be lowered.
3) In the early diagnosis in cases of suspected occupational diseases. Furthermore, neurophysiological examination may aid in deciding whether or not the patient qualifies for workmen's compensation. 4) In making the decision whether or not a worker may remain exposed. It is important to lower the exposure level, or to stop harmful exposure periodically or totally, early enough to prevent permanent lesions. At an early stage neurotoxic effects are often reversible. 5) In detecting especially vulnerable workers. Persons with nervous system diseases should he excluded from work with exposure to neurotoxic chemicals.