Document oNkJyvE7rBK10RzydpKZD7Mo

JOM Volume 36, Number 10, October 1994 1079 A Review of Recent Research on Health Effects of Human Occupational Exposure to Organic Solvents A Critical Review Edward L. Baker, MD, MPH In 1985, two important international conferences were held at which the relevant literature on solvent-related effects on the nervous system was re- viewed.'.* Health effects on other or- The health impact of workplace solvent exposure remains an issue of substantial interest and concern to occupational health professionals. As a result of research performed in the 1970s and 1980s.policies and programs were developed throughout the world to control excessiveexposureto solvents. To an extent, these programs have been responsible for reduction of the occurrence of solvent-associated encephalopathy and other health effects. In this review of research perJormed since 1985,particular attention is given to issues of reversibility of neurotoxicityfollowing exposure cessation. Furthermore, health effectsinvolving other organ systems, particularly reproductive, renal, and hepatic disorders, are discussed. Future research directions are discussed. Finally, thepractical implicationsof these recent researchfindings are described with afocus on the management ofprevention programs at the work site. gan systems received limited attention at those conferences. Since 1985, numerous studies have been performed in an attempt to clarify the issues further.) This review attempts to summarize these studies. In each section, a brief synopsis of the most important recent studies is provided, followed by an assessment of the significance of the aggregate findings in a research area. The review evaluatesthe need for further research and describes thL\ pract;d implica- tions of the research for control of solvent neurotoxicity in the work- place. / ,/- Dr Baker is Director, Public Health Practice Program Office, Centers for Disease Control and Prevention, Atlanta, Georgia. This work was presented at the Paint-Health and Environmental Risk Management International Symposium, Singapore, June 2, 1992. Address correspondence to: Edward L. Baker, MD, MPH, Director, Public Health Practice Program Office, Centers for Disease Control and Prevention, Atlanta, GA 30333. 0096-1736/94/3610-1079$03.00/0 Copyright 0 by American College of Occupational and Environmental Medicine The Nervous System The Central Nervous System Neurobehavioral Efects. Crosssectional studies. In a report prepared for a 1985 World Health Organization working group meeting, Roberta White and I summarized the existing literature on neurobehavioral testing in solvent-exposed population^.^ In the studies we reviewed, differences between exposed and unexposed were most commonly seen on tests of psychomotor function and short-term memory. Studies published since then show a similar pattern (Table 1). In particular, performance on tests of memory function, particularly shortterm memory abiiity, was poorer in solvent-exposed workers. In one pop- ! !I li 1: il f / I 1080 Review of Exposureto Solvents Baker TABLE 1 Recent Neurobehavioral Studies Study Exposed Group (N) Hanninen et a1 (1991)` Moen et a1 (1990)' Morrow et a1(1990)" Bleecker et al (1991)5 Twins (28) Seamen (85) Clinic patients (32) Paint factory workers (187) Bolla et al (1990)' Parkinson (1990)92 Maizlish et a1 (1987)" Electronics plant workers (140) Spray painters (124) Maizlish (1985)94 van Vliet et a1(1989)" Fidler et a1(1987)16 Cherry (1985)93 Solventexposed (379) Painters (101) Painters (44) Tolueneexposed(52) Baker et al (1988)13 Spurgeon et a1 (1992)6 Painters (186) Painters (90) Solventexwsed (1441 ~ +, Exposurerelatedeffect seen; -, exposurerelated effect not seen. Visual Performance Motor Deficit Memory Deficit Behavioral Symptoma ulation studied by Bleecker and colleagues,5 deficits in performance on memory tests were observed in groups without evidence of behavioral symptoms of the type seen with solvent toxicity. Because each study used different neurobehavioral tests, comparison of results between studies on the actual tests is difficult. However, some similarities can be observed (Table 2). Performance on two tests of short-term memory function (ie, associate learning tasks and tests of memory span) has been reduced consistently in solvent-exposed workers. Psychomotor function, as measured by the digitsymbol substitution test, was reduced in exposed workers in two recent stud- ies, and in other studies before 1985. The block design test, evaluating vis- uospatial ability, also showed differ- ences between exposed and unexposed in two recent studies. Only a few studies attempted to develop exposure models and to compare exposure estimates with test performance. Such comparisons involve, at times, the use of statistical techniques to make the exposure-response comparison while controlling for the confounding effects of age, etc. Spurgeon et a16 have recently completed studies of two solvent-exposed populations: brush painters (n = 90) and a heterogeneous group of solventexposed workers (n = 144). They observed decrements on performance on tests of psychomotor function (ie, symbol-digit substitution) and short- term memory (ie. paired associate learning) that were correlated with in- dices of exposure derived from work history. Those with greater than 10 years of exposure showed performance decrements particuIarly on memory tests. The results obtained from the two separate populations were strikingly similar. TABLE 2 Tests on Which Solvent-Exposed Workers Performed Poorly Relative to Unexposed Workers Study Teat Function Tested Hanninen et a1 (1991)' Associate learning Digit span Block design Short-term memory Short-term memory Visuospatial ability Moen et al (1990)' Memory span Raven's matrices Short-term memory Visual abstraction Morrow et a1(1990)'' Associate learning Visual reproduction Block design Pegboard Short-term memory Memory Visuospatial ability Psychomotordexterity Bleecker et a1 (1991)5 Digit symbol Serial digit learning Reaction time Trails Attention Short-term memory Psychomotor speed Attention Maizlishet at (1987)'* Memory span Short-term memory Baker et al (1988)13 Symbol digit Reaction time Pattern memory Attention speed Short-term memory JOM Volume 36, Number 10, October 1994 1081 In the Hanninen et a1' Finnish report of pairs of monozygotic twins with discordant solvent exposure histones. the investigators divided their exposed population into moderate and low exposure groups. Those with moderate exposure showed greater performance decrements when compared with their twin than did those with low solvent exposure. Moen et ala used an exposure index that combined intensity of exposure and duration of exposure in the 85 seamen whom they studied. This index was used in a multivariate regression model relating exposure to effect while controlling for relevant confounding factors. Performance on two tests. evaluating visual abstraction ability and memory span, showed statistically significant negative correlations with the exposure index. In studying 187 workers in two paint manufacturing plants. Bleecker and colleague^^.^ modelled exposure using personal breathing zone sampling results obtained over a 13- to 15-year period. The results were com- bined to form a lifetime-weighted av- erage exposure estimate that was used in regression analyses. Test performance was negatively correlated with exposure on four tests (digit-symbol substitution, serial digit learning. reaction time. and trail making [A and B]) after controlling for relevant confounding factors. Ng et all" observed increased neurobehavioral symptoms and reduced test performance in Chinese printing and paint workers. Increased reports of fatigue. irritability. depression. poor memory. and sleep disturbances were noted. Reduced psychomotor and memory test performance were also observed. van Vliet et a]'' developed two ex- posure indices in studying 379 exposed Dutch workers: a "painters index" based on the work practices of painters in the study and a "general exposure index" for all exposed study participants. This study did not use neurobehavioral testing; a symptom questionnaire was used to assess effects of exposure. The painters index and the general exposure index were correlated with a composite score for "prenarcotic symptoms." Maizlish et all2developed an exposure index based on airborne exposure measurements at the time of testing, duration of employment, and relative toxicity of quantified solvents. Test performance decrements were not correlated with this exposure index. In a study of 186 construction painters reported by Baker et correlations between performance on various neurobehavioral tests and solvent exposure indices were observed. Disruption in mood was correlated with indices of exposure to solvents. Impaired performance on certain tests of psychomotor function also was observed to correlate with exposure intensity. Symptoms of forgetfulness, disorientation, dysphoria. lassitude. and difficulty falling asleepwere noted with increased frequency by those more heavily exposed. In 1988. Gade et ai" performed a reanalysisof neurobehavioral test data in 20 workers previously diagnosed as having solvent encephalopathy and compared their results with 20 matched control subjects. Extensive neurobehavioral testing had been performed between 1980and 1983;using these data. statistical analyses were performed and led the authors to conclude that the workers previously diagnosed as having solvent-associated encephalopathy had no evidence of significant neurobehavioral impairment. In this report. Gade et al" used statistical techniques to adjust for the presumed confounding effects of age. education. and verbal intelligence. These statistical techniques deserve particular attention because the conclusions rest heavily on the validity of the approach taken. Controlling for the effects of age and education is common and usually appropriate in studies that use neurobehavioral tests as an outcome measure because age and education are correlated with performance on many neurobehavioral tests. However, in this study. because age and education were factors on which matching was based, the exposed and control groups were virtually identical with respect to age and education; further statistical adjustment for these two factors would seem to be unnecessary. Adjusting for verbal intelligence differences was justified by the authors (and by others performing similar research) to address possible "premorbid" noncomparability of intellectual functioning. This adjustment was accomplished using the results on four tests (an adult reading test and the information. similarities. and vo- cabulary subtests of the Wechsler Adult Intelligence Scale). The as- sumptions underlying this procedure are that test performance on these four tests is an accurate and unbiased indicator of premorbid function in all areas of neurobehavioral function evaluated in the study and, most im- port. ntly, that test performance is not adversely effected by chronic solvent exposure. Because these assumptions have been tested inadequately in other studies and because the data of Gade et all4 show decrements in neurobehavioral performance in tests of memory span and visuomotor speed before adjustment for verbal intelligence test performance, the significance of this study reanalysis is unclear. Mikkelsen et in an extensive analysis of 85 painters and 85 bricklayers, noted changes in neurobehavioral function in relationship to solvent exposure. Cerebral atrophy also was noted to be increased in relation to the history of solvent exposure. The authors observed little risk of organic brain damage in persons with fewer than 13 years of exposure to the equivalent of a time-weighted average of 40 ppm of white spirit. In summary. these 16 studies, involving more than 1000 exposed workers in 6 different countries. on 3 continents, yield results that are reasonably consistent among themselves and with studies published before 1985. In these studies. solventexposed workers demonstrate decrements in short-term memory and psychomotor function. This decrement was not related to various potential confounding factors. There was. in most studies. a "dose-response relationship," such that an increased level of cumulative exposure was associated with performance decre- ments. The types of neurobehavioral deficits are similar to those seen in heavily exposed workers and in per- 1082 sons with pronounced neurobehavioral dysfunction due to solvent abuse. Some studies have focused exclusively on symptoms related to solvent exposure. .A study of industrial spray painters by Fidler et all' showed multiple correlations between indices of exposure drawn from work history questionnaires and reports of various neurotoxic symptoms in currently employed painters (eg, dizziness. fatigue, and acute lightheadedness at work). Two other similar studies by Maizlish et all' and Bolla et al' failed to find substantial correlations between cumulative exposure indices and current symptoms. In these two studies. workers were employed in manufacturing facilities where measured exposure levels were rather low (with the exception of plant 4 in the Maizlish study) in relation to existing permissible exposure limits. Chronic symptoms consistent with prior reports of solvent toxicity were reported in a study of 32 clinic patients reported by Morrow et al." From these symptom studies, there is little convincing evidence that persistent chronic symptoms are observed in currently employed workers if they work where exposures are controlled below the permissible exposure limits. Studies reporting symptoms in exposed workers are apparently either documenting persistent symptoms developed from high exposures in the Review of Exposure to Solvents Baker past or among workers (eg, construction painters) who are employed in situations with inadequate current exposure controls. Although epidemiologic studies. particularly cross-sectional ones, have limits in assessing causality. criteria have been developed by which studies can be evaluated in this regard." Three important criteria are (1) consistency of findings. ( 2 ) biological plausibility, and ( 3 ) presence of a dose-response relationship. The studies reviewed have demonstrated consistent results, as mentioned above. Although biologic models for solvent neurotoxicity are limited. the findings in these studies are consistent with a diffuse toxic encephalopathy that preferentially effects certain brain structures.l9 Finally. dose-response relationships were observed in most studies. Therefore. these cross-sectional studies, viewed in the aggregate, support the view that chronic solvent exposure causes impairment of neurobehavioral function. Follow-up studies using neurobehavioral methods. Five follow-up studies(Table 3 ) have been performed since 1985. These studies are extremely important in evaluating the long-term significance of solventassociated changes in neurobehavioral function. Orbaek and Lindgren" reported on follow-up studies of 32 men diagnosed as having chronic toxic encephalopathy 2 1 to 88 months previ- ously. Symptoms improved considerably in these workers who had been removed from exposure at the time of diagnosis. Neurobehavioral testing showed deterioration in test performance on a paired associate learning task and in simple reaction time. Most neurobehavioral test results were unchanged from initial testing at time of diagnosis. Gregersen" reported on follow-up of a cohort of solvent-exposedworkers tested 5 and 10 years after initial examination. Exposure continued for most of the group during the interval but airborne concentrations of solvents were reduced considerably over the follow-up period. Both exposed and unexposed workers were followed over the 10-year interval. Symptoms. particularly complaints of memory disorders and impairment in concentration, improved over the period. In some workers for whom exposure ceased, symptoms persisted with a frequency greater than in the control SOUP. In a separate study,?' 21 painters diagnosed as having solvent-induced chronic toxic encephalopathy were evaluated 5 years after original diagnosis. All were removed from exposure to solvents. Symptoms improved considerably. Neurobehavioral testing showed impairment in memory, attention, and abstraction ability. Social follow-up revealed that those who were able to return to work in another TABLE 3 Follow-Up Studies Study Orbaek et al (1988)'' Gregersen (1988)" Gregersen et al (1987)22 Edling et a1(1990)23 Morrow et a1(1991)~' Subjects (N) Chronic toxic encephalopathy (32) Solvent-exposed(59, 53) Chronic toxic encephalopathy (21) Symptomatic only (65) Chronic toxic encephalopathy (46) Toxic encephalopathy Follow-up Interval, y 2-7 5.5 and 10.6 5 6.8 (mean) 6.7 (mean) Results Symptoms: better Tests: some worse, others unchanged Symptoms: better Symptoms: better Tests: unchanged Symptoms: better Symptoms: unchanged Tests: some improved JOM Volume 36, Number 10, October 1994 1083 job were younger, had a history of less exposure, and were less intellectually impaired than those who failed to obtain new jobs. Edling et al'3 reported a follow-up study of two groups: 65 workers with symptoms but no impairment of psy- chometric test performance impairment, and 46 workers with mild chronic toxic encephalopathy (ie. symptoms and test performance abnormalities). Those with symptoms alone (type 1 solvent toxicity. according to the 1985 US solvent workshop classification' experienced improvement in certain symptoms (depression, concentration difiiculties. and lack of initiative), when the results of the group are taken as a whole. Three of the 65 showed persistence of symptoms and deterioration of test performance leading to their being diagnosed as having mild chronic toxic encephalopathy on the second examination. Those 46 workers with mild chronic toxic encephalopathy on the initial examination generally showed persistence of symptoms and test performance decrements on follow-up testing. However. 12 of the 46 showed sufficient test performance improvement to be reclassified as type 1 solvent toxicity. Morrow et a1" reported some improvement in neuropsychologic test performance in persons with mild toxic encephalopathy following solvent exposure. Predictors of improvement in this group included lack of history of a "peak" exposure to solvents and absence of significant psychological distress. In summary, these follow-up studies present a reasonably consistent picture. Persons showing symptoms only after solvent exposure show im- provement after removal from solvent exposure or after significant exposure reduction. Persistent deficits in neurobehavioral function are observed in persons with more severe initial impairment even if additional exposure is totally eliminated. Some of those with persistent impairment suffer sub- stantial limitations in functional ability and are persistently disabled. Other Central Nervous System E$ fects. At least five case-referentstudies of selected neurologic conditions have been performed in recent years to evaluate the association of solvent exposure with these disorders (Table 4). Another paper reported the results of a meta-analysis of 1 1 studies of Alzheimer's disease in which solvent exposure was evaluated. Two studies examined groups receiving disability compensation for neuropsychiatric disorders. van Vliet et all ' observed increasing likelihood of receiving disability compensation among Dutch construction workers as exposure increased; an exposure index was constructed using job history information. Brackbill et all' examined US Social Security records for persons receiving disability compensation for presenile dementia, alcoholism, affective psychoses, neurosis. personality disorders. and other diseases of the brain (International Classification of Diseases Code 347). They found an odds ratio for the painting occupation of 1.42 (95% confidence interval [CI]: 1.04. 1.94). These two studies are similar in design and results to that of Axelson et al." one of the first studies to suggest a relationship between solvent exposure and neuropsychiatric disorders. Four recent studies investigated more discrete neurologic disorders. OFlynn et al" examined death certificates of 557 men who died from presenile dementia and found no evidence of increased rates of employment in solvent-exposed jobs in comparison with a matched control group (relative risk [RR]. 1.14). Littorin et al" compared occupational histories of 104 cases of idiopathic focal epilepsy with matched referents and found that risk of disease increased with exposure level. Eightythree multiple sclerosis patients were compared with referents by Flodin et aF9and observed to have higher rates of solvent exposure (RR. 2.34). Hawkes et a130investigated patterns of employment in the leather industry of patients with motoneuron disease; the results are difficult to interpret for methodologic reasons. The case-referent studies are quite varied in approach and results; in most instances. further research is needed to determine the significance of these findings. However, there appears to be some consistency of results between the three studies of persons receiving disability compensation. Together with the cross-sectional studies discussed above. these studies of disability pensioners raise concern that the disorders noted in cross-sectional studies of currently employed workers may lead to significant disability. This concern is best addressed scientifically through follow-up studies in which serial evaluations are performed on persons believed to have solvent-related encephalopathy. One hundred eighty-seven paint manufacturing workers were tested by Schwartz et a13' using the University of Pennsylvania Smell Identification Test. Nonsmoking solvent-exposed workers were found to have decre- TABLE 4 Case-Referent Studies of Neuropsychiatric Disorders Study Cases van Vliet et al (1989)" Brackbill et al (1990)25 O'Flynn et al (1987)2' Littorin et al (1988)28 Flodin et a1 (1988)29 Various neuropsychiatric disorders Various neuropsychiatric disorders Presenile dementia Focal epilepsy Multiple sclerosis Increaseddisease risk with increased exposure. Exposure Measure Job history questionnaire Job history Job title Job title Job history Odds Ratio 1.47 1.14 (NS) 2.3 Review of Exposureto Solvents Baker exposure. Each person experienced ties in the cerebrospinal fluid (CSF) the initial "panic attack" after work- of 19 patients with chronic toxic en- place solvent exposure. Subsequent cephalopathy (CTE). Elevated levels attacks occurred both at work and in of protein, albumin, and IgG were nonwork situations usually after ex- observed; lowered concentrations of posure to an olfactory stimulus. The certain amino acids were also found. authors describe the use of lactate in- Taurine levels were negatively cor- fusion as part of the diagnostic proc- related with a solvent exposure in- ess. dex. Prior to 1985, several studies re- Barregard et a14' performed a simi- ported electroencephalogram (EEG) lar study of 23 patients with CTE and abnormalities in workers exposed to found a slight increase in CSF protein styrene and in patients with "chronic and IgG levels. The difference be- solvent poisoning." Orbaek et a138per- tween patients and a matched control formed a study of 32 workers with group was not statistically significant. chronic toxic encephalopathy em- Prior Swedish and Finnish studies dis- ploying EEG power spectrum analy- cussed by the authors had failed to sis: 24 of the 32 were retested 17 to 75 show CSF changes in similar popula- months later. Initial testing showed a tions. doubling of the EEG power in all four Computed tomography scans of 62 recording channels: follow-up testing CTE patients showed no evidence of showed reduction in power but not to structural brain changes when com- the level of the control group. The pared with control subjects in a study authors interpret the study as showing by Orbaek et al.43This study failed to evidence of diffuse organic brain in- confirm one prior "positive" study volvement. and was consistent with two other Electrophysiologic tests measuring "negative" reports. the cortical response to auditory, vis- Diminished cerebral blood flow ual, and somatosensory have been (7% lower than unexposed referents) used widely to diagnose certain neu- was noted by Hagstadius et al," when rologic disorders (eg, multiple scle- examining 28 patients with CTE. Re- rosis) and with increasing frequency peated testing 24 to 84 months later in workers exposed to toxic sub- showed persistent but less marked re- stances. These techniques have been ductions in flow. The authors found recommended by some as a way to these results consistent with prior evaluate central nervous system func- studies of cerebral blood flow in sol- tion because results are not influenced vent-exposed workers and a group of by the effort of the person tested. house painters. Hazemann et alj9 studied 13 per- In summary, these varied studies of sons diagnosed as having psycho- other central nervous system effects organic syndrome using peripheral show variable relationships with sol- and somatosensory conduction stud- vent exposure. Solvent-exposed work- ies. Peripheral conduction velocities ers may show evidence of loss of smell, Ledin et a135performed follow-up and amplitudes were normal. In con- as noted by Schwartz et al;45this study studies of seven patients with chronic trast, central conduction velocity was is methodologically sound and the toxic encephalopathy using dynamic slowed; the degree of slowing of con- findings are consistent with prior re- I posturography. Testing in 1986 (re- duction was greater than that seen in search. The observation of color vi- ! ported by Moller et a136)showed ab- a group of alcoholic persons tested sion loss in solvent-exposed workers. normalities of posture consisting of using similar methods. Three subjects although of interest, requires further increased body sway and impaired who were both solvent-exposed and study to evaluate the results found by ability to cancel the vestibula-ocular alcoholic showed greater conduction Mergler and Blain.46 Several studies slowing than any other group. Mas- point to a mild effect of solvent- sioui et ala published a report on the exposure on neurologic systems (pre- same population showing normal au- dominantly the cerebellum) that con- ditory and visual evoked responses, in trol balance and posture. One follow- addition to the results noted in the up investigation revealed persistent Hazemann paper. dysfunction 3 years after removal Moen et a14' observed abnormali- from exposure. The biologic s i g n 6 JOM Volume 36, Number 10, October 1994 1085 cance of these mild changes in test performance is unclear. CSFconcentrations of albumin and IgG may be elevated in solventinduced chronic toxic encephalopathy patients. Such patients may also exhibit diffusely reduced cerebral blood flow. Structural changes. as measured by computed tomography. do not appear to be present in solvent-induced chronic toxic encephalopathy. Electrophysiologic studies. Recent peripheral nervous system conduction studies revealed inconsistent findings: one was "positive," the other "negative." Earlier published research of this type has tended to show conduction abnormalities consistent with mild peripheral nervous system dysfunction. None of these studies has shown severe neuropathy of the type seen after exposure to n-hexane and other structurally similar hexacarbon solvents. EEG studies in solventexposed workers have tended to show diffusechanges consistent with diffuse cortical dysfunction. 1990 Conference on Organic Solvents and the Nervous System. In May 1990, the Commission of the European Communities and the Danish Ministry of the Environment convened an international conference on organic solvents and the nervous system. The conference was designed to review and discuss recent knowledge accumulated since a prior conference on the same subject held in Copenhagen in 1985.' The conference report provides essentially a summary of the presentations given and the opinions expressed by the participants. AIthough the conference was not designed to obtain consensus on particular issues, some summary views are expressed in the report. Review of epidemiologic studies showed that solvent-exposed workers had ( 1 ) an increased risk of disability pensions for neuropsychiatric disorders: (2) excess neuropsychiatric symptoms. if currently employed: and (3) in most studies. lower performance in neurobehavioral tests. if currently exposed. The clinical manifestations were believed to range from an affective syndrome. characterized by mooddisturbances alone, to a severe chronic toxic encephalopathy. characterized by short-term memory impairment, fatigue, attention deficits, and other forms of cognitive change. This encephalopathy may be irreversible. but usually not progressive, upon cessation of exposure. In summary, there was general agreement that the syndrome of chronic toxic encephalopathy, a term preferred over "dementia," does occur in workers with excess exposure to solvents. The condition may be associated with significant deficits in neurobehavioral function and, in some, may be sufficiently severe to result in permanent disability. The disorder does not seem to be progressive after cessation of exposure. Exposed persons identified at an early stage of dysfunction (ie, symptoms in the absenceof neurobehavioral dysfunction) appear to recover total function without sequelae. In addition to reviewing human research, the conference report discussed recent animal studies. These studies are less numerous and provide an incomplete view of the effects of solvents on exposed animals. Effects have been observed on brain morphology, brain biochemistry, and behavior. The relevance of these changes to the effects seen in humans is unclear. Peripheral Nervous System Certain solvents are recognized to cause a symmetrical distal sensonmotor peripheral neuropathy. Outbreaks caused by exposure to solvents such as n-hexane and methyl n-butyl ketone have been well documented" and the basic pathophysiology well delineated. Exposure to these solvents is limited to a relatively small segment of the work force. Many more are exposed to other solvents either singly or in mixtures; peripheral nervous system function in such groups has received limited attention. Peripheral nervous system testing, by Orbaek et al.38of 32 chronic toxic encephalopathy patients (the same group described above) showed slowing of motor conduction in the me- dian nerve that was more pronounced in follow-up testing 22 to 72 months later. Slowing in peroneal conduction velocity was observed only at the follow-up examination. Sensory conduction studies showed substantially reduced amplitude in median and sural nerves with a prolongation of the action potential in comparison with a control group; sensory conduction velocity in the median nerve was also slowed in the follow-up examination. Psychophysical tests measuring temperature sensation showed differences between the patient group and control subjects consistent with sensory decrement. The decrement in sensory function persisted in followup testing. The authors interpret these findingsas indicating that exposure to solvent mixtures can cause a mild sensory neuropathy that is partially reversible after cessation of exposure. Maizlish et al," in a study of printers and spray painters, observed reduced two-point discrimination ability in the lower extremities but did not find reduced vibration sensation. The methods used in this study were the routine physical examination. a technique much less reproducible than the techniques used by others in three later studies. A s part of a study using nerve conduction testing, Orbaek et a13*observed slight changes in temperature sensitivity among 32 men with solvent-induced chronic toxic encephalopathy. These changes were observed in both the upper and lower extremities and remained unchanged when a portion of the group was retested 22 to 7 2 months after initial testing. Demers et a14' observed similar deficits in vibration sensitivity among a group of commercial painters. Reduced vibration perception in all four extremities was noted in comparison with an unexposed control group. Bove et used standardized tests of vibration and temperature perception to evaluate peripheral nervous system function in solvent-exposed construction painters. Threshold for perception of vibration in the lower extremities was raised in proportion to two indices of exposure to solvents over the previous year. Temperature 1086 Review of Exposure to Solvents Baker sensitivity was also effected in a similar manner. In the study of Bleecker et als (summarized above in the section on neurobehavioral studies). vibration sensation in the upper and lower extremities was evaluated using. a standardized approach similar to that of Bove et al." Elevated thresholds for vibration sensation were noted in the lower extremities: the degree of change was correlated with a lifetime exposure index. In summary, workers exposed to solvent mixtures containing substances other than n-hexane and methyl n-butyl ketone have shown evidence of subclinicalchanges in sensory thresholds. These workers did not show evidence of clinical signs or symptoms. As discussed below, studies using nerve conduction testing have shown mild changes consistent with disruption of sensory nerve function (see Orbaek et al,38for example). Taken as a whole, these recent studies suggest that mild subclinical disruption in peripheral nervous system function does occur in workers exposed to solvent mixtures. Neuropsychiatric Disorders In contrast to the many studies of cognitive changes in solvent-exposed workers, few investigators have focused on personality changes. Morrow et als' observed a pattern of personality disturbances in 22 clinic patients with a history of solvent exposure. Using the Minnesota Multiphasic Personality Inventory, they observed increased rates of anxiety, depression, complaints of difficulty concentrating, and feelings of unreality and disturbances in thinking. The authors note striking similarities between these observations and those seen in posttraumatic stress disorder experienced by combat veterans. Ryan et al" report an unusual syndrome. cacosmia, in a group of solvent-exposed workers seen in an occupational medicine clinic. Cacosmia is an olfactory hypersensitivity in which affected persons respond to neutral or mildly unpleasant odors (eg, hairspray, gasoline, or perfumes) with headaches, dizziness, and feelings of nausea. Decrements on neurobehavioral testing were observed in these workers. McCrank and Rabhen" report four cases of progressive supranuclear palsy associated with solvent use. This condition is characterized by paralysis of external ocular movements, dysarthria. dysphasia, dystonic rigidity of the trunk and neck. and pseudodementia. The exposure histories of these four cases were vaned and difficulty to evaluate. Sleep apnea, initially associated with solvent exposure in a 1983 case report. was found in 7 of 15 patients with CTE.53Another group of 8 workers exposed to trichloroethane had a higher number of sleep apnea events (mean = Z/hour) than did a control group (mean = ?/hour). These and other reports of various neuropsychiatric syndromes associated with solvent exposure are difficult to evaluate. Because solvent exposure is common. chance associations with these conditions of unknown etiology undoubtedly occur: these reports, even when presented as a series of cases, could r e p resent chance associations. Heightened interest and attentiveness by the investigators to the possible association could have led to certain biases. Alternatively, as in the case of the original description of solvent encephalopathy by a Swedish occupational physician, these case series could represent genuine steps in our understanding of the varied manifestations of solvent neurotoxicity. Further study may resolve the dilemma. Other Systems Reproductive Effects Spontaneous Abortion. In two separate studies of pregnancy outcomes in solvent-exposed women, increased rates of spontaneous abortion were noted. Lindbohm et als4 used three Finnish data bases to identify 120 women who experienced spontaneous abortion and 336 control subjects with normal pregnancy outcomes. The odds ratio (OR) for solvent exposure was 2.2 (95% CI, 1.2 to 4.1) overall: women exposed to aliphatic hydrocarbons had a higher odds ratio for exposure (OR, 3.9: 95% CI, 1.1 to 14.2), as did women employed as graphics workers (OR, 5.2; 95% CI, 1.O to 84.7). Lipscomb et alS5found a similar association in studying women employed in the electronics industry in California. First-trimester solvent exposure was associated with spontaneous abortion (OR, 3.34: 95% CI, 1.42 to 7.81). Those with regular or daily exposure showed a higher exposure risk (OR, 4.44: 95% CI, 1.86 to 10.58). Congenital Malformations. McDonald et aIs6 observed increased rates of urinary tract defect in offspring of women exposed to solvents. Toluene was the most common specific substance to which affected women were exposed. In the same study, the investigators failed to find an association between adverse reproductive outcomes and employment in various job categories without solvent exposure. Tikkanen and Heinonen" observed slightly increased rates of cardiovascular anomalies, particularly ventricular septal defect. in women exposed to solvents during the first trimester of pregnancy. The relative risk of 1.3 was not statistically significant for all cardiovascular defects and marginally significant (RR, 1.5) for ventricular septal defect. Other Effects on Offspring after Maternal Exposure. Eskenazi et aIs8performed detailed neurobehavioral evaluations on the children of solventexposed mothers at 3 to 4 years of age. No differences in neurodevelopment were noted when performance was compared with a matched control group. Maternal Efects. Increased rates of preeclampsia in solvent-exposed women were noted by Eskenazi et along with increased rates of pregnancy-associated hypertension. Relative risks, adjusted for confounding factors, were 3.9 and 3.0, respectively. The authors related this observation to reports of solvent-associated renal disorders (discussed below). JOM Volume 36, Number 10, October 1994 1087 Efrects of Paternal Exposure. Two investigations of paternal solvent exposure and reproductive outcome have appeared recently. Daniell and Vaughanm reported low birth weight in the offspring of male body-shop workers (RR. 1.6) and painters (RR. 1.4). The excess risk was noted only in full-term infants. Brender and Suarez6` observed increased odds of paternal solvent exposure in cases of anencephaly (OR, 2.53). Painters showed the highest odds ratio at 3.43. Genotoxicitv. Kelsey et aI6' observed that acute exposure to solvents in construction painters was associated with increased rates of sister chromatid exchange (SCE), a measure of genotoxicity. The combination of a history of solvent exposure and of cigarette smoking was associated with an additive increase in SCE rate. Salamanco-Gomez et aP3 found similar changes in 35 children who were chronic abusers of solvents by inhalation. A ninefold increase in the rate of chromosome breaks was observed. Rates of SCE in exposed children were approximately 3 times that seen in unexposed comparison children. Summary. In reviewing selected recent articles regardingreproductive effects of solvent exposure, an association appears to exist between maternal solvent exposure and increased risk of spontaneous abortion. Both studies summanzed above noted moderate levels of risk (OR. 2 to 3): one study demonstrated increased risk with increased exposure levels. Effects on offspring after either maternal or paternal exposure are more difficult to interpret because each study summarized above deals with a different outcome and because little prior research exists to support the specific observations. Further study is needed. The observation that solvent-exposed mothers experience increased rates of preeclampsia is rather intriguing, particularly in view of other evidence of solvent-associated renal disease. This finding deserves particular attention in subsequent research. In view of its biological significance, further studies of genotoxicity in solvent-exposed workers should be given priority. Cancer Most recent studies and one review article that deal with the association between occupational solvent exposure and cancer development focus on malignanciesof the lymphatic and hematopoietic tissues. Olsson and Brandt"` observed an odds ratio of 3.3 for solvent exposure in a group of non-Hodgluns lymphoma patients. Cases that were localized to supradiaphragmatic sites showed a higher association with solvent exposure (OR, 6.5). Increasing duration of solvent exposure was associated with increased risk of lymphoma. In a study of a cohort of dry-cleaning workers, Blair et a16` observed in- creased standardized mortality rates for esophageal and cervical cancer that were statistically significant. Increased rates were also observed for other cancers including lymphosarcoma, reticulosarcoma, and Hodgkin's disease that were not statistically significant. Although the numbers of deaths were small. the greatest risk for cancers of the lymphatic and hematopoietic systems (standardized mortality ratio, 4.0) was seen in workers with heaviest solvent exposure. Brandt et alb6reviewed a variety of studies and observed that both Hodgkin's disease and non-Hodghn's Iymphoma have been associated frequently with solvent exposure: the level of association was moderate (OR. 3 to 7). Brandt also reviewed numerous studies that reported the well-established association between acute leukemia and benzene exposure, along with the limited studies relating myeloma and solvent exposure. Hernberg et aI6'observed increased rates of liver cancer in Finnish women exposed to solvents. In a well-designed case-referent study. the investigators noted an odds ratio for exposure of 7.8. Men included in the same study did not show this association. Brandt and colleagues6s studied lymphoma cells from non-Hodgkm's lymphoma patients with solvent exposure and found consistent chromosomal aberrations. The specific changes, involving a translocation on chromosome 14, were not seen in patients without solvent exposure. Renal Disease Increased urine concentrations of erythrocytes. leukocytes,and albumin were noted by Hashimoto et aP9 in a group of 2 15 solvent-exposed workers. These findings are largely consistent with the observations of Hotz et aL70who found increased urine concentrations of erythrocytes, albumin, and N-acetylglucosaminidase in 148 solvent-exposed workers. Other studies published before 1985 have reported similar findings. In the Hashimoto study, these changes in urine were not associated with reduction in glomerular filtration rate, as noted by serum blood urea nitrogen and creat- inine concentrations. Hashimoto et aF9 suggest that the cellular changes may derive from effects on the lower urinary tract. One recent study by Hamngton et a1" investigated the association of solvent exposure with the occurrence of glomerulonephritis and renal cancer. Solvent exposure occurred no more frequently in 50 biopsy-proven cases of glomerulonephritis than in referent cases. No association with solvent exposure was noted for renal cancer. The authors compare these results with two other "positive" and one other "negative" study of solvent exposure and glomerulonephritis. No conclusions on the overall issue are drawn by the authors. Followingthe summary of a case of glomerulonephritis in a 29-year-old man employed as an automobile cleaner (exposed to Stoddard solvent). Daniell and colleagues-' reviewed the relevant literature. They found that despite acknowledged methodologic limitations. seven of the nine casecontrol studies published before 1988 showed statistically significant associations between solvent exposure and glomerulonephritis. The risk vaned between 2.8 and 8.9: three studies found that the prevalence of glomerulonephritis increased with the intensity and direction of solvent exposure. Daniell et d7*also reviewed other studies showing various abnormalities in renal function among solvent- 1088 Review of Exposure to Solvents Baker exposed populations. Navarte et a173 ception of one outbreak due to metabolism and measurement of presented one case report of chronic dimethylformamide and one large serum bile acid concentration-have tubulointerstitial nephritis in a sol- cross-sectional study. recent papers been used in three studies. DossingEo vent-exposed worker: the patient also have focused on case reports after found significant changes in anti- had a history of ulcerative colitis. heavy exposure and studies using pyrine metabolism among jet-fuel- Based on their review of prior animal more "sensitive" indications of he- exposed workers. Bile acid concentra- and human research. the authors con- patic function. tions were elevated in solvent-exposed clude that the case was probably Redlich et a P reported on the re- workers studied by Franco et aLR' caused by solvent exposure. Sesso et sults of an investigation of 58 workers Their results were similar to an earlier al-4performed a case-control study of exposed to dimethylformamide at a study of styrene-exposed workers by 17 patients with biopsy-proven cres- fabric coating factory. The study fol- Edling et al.M2 cent-c glomerulonephritis and 34 lowed the occurrence of a case ofacute In summary, exposure at high levels matched hospital control subjects. toxic hepatitis in a 40-year-old man to recognized hepatoxic solvents con- The relative risk for solvent exposure employed at the plant. Exposure to tinues to result in cases of acute hep- was 5.0 (95% CI. 1.14 to 22.00). dimethylformarnide occurred through atitis, which in some cases may result Summary The literature regarding solvent-induced renal disorders has continued to grow in recent years. Consistent results have been obtained in several studies regarding changes in urine sediment and protein concentrations: solvent exposure appears to cause increased levels of erythrocytes and leukocytesin urine. The exposure level necessary to cause this effect is uncertain. as is the long-term significance of these findings. The association between glomerulonephritis and solvent exposure continues to be difficult to assess despite several recent methodologically sound studies. On balance, the majority of the investigationsshow an association (three- to fourfold risk) between glomerulonephritis and solvent exposure. In some cases, an exposureresponse relationship was seen. When it occurs, glomerulonephritis associated with solvent exposure tends to be of a particular type-rapidly progressive glomerulonephritis. A plausible biologic mechanism has been postulated: antiglomerular basement membrane antibody related to release of antigen after solvent exposure. Other types of renal disorders, such as disease of the renal tubules, may be associated with excessive solvent exposure. Further research is needed to explore those associations. Liver inhalation and skin contact in the index case and other employees: appropriate ventilation and skin protection had not been used. Biochemical tests of liver function showed that 36 of the 58 (62%) workers had elevated blood concentrations of aspartate amino transferase or alanine amino transferase, indicating hepatic dysfunction. Improved work practices and removal of workers most severely affected resulted in improvement in most: some workers showed persistent elevations of liver enzyme levels. Harrison et a17"reported two cases of fulminant hepatic failure in workers exposed to nitropropane, a known hepatotoxin; one worker died. Hodgson et a1" reported on a series of four patients with fatty liver disease and 1.1,1 -trichloroethane exposure. This solvent has been used widely, in part because it has been believed to have little potential for hepatotoxicity; the authors of the case series feel that 1.1.1 -trichloroethane may have a greater potential for causing liver dysfunction than previously recognized. Cordes et ai7' reported on two cases of toxic hepatitis after heavy exposure to methylene chloride and to solvent mixtures. Chen et a179reported on an extensive study of two paint manufacturing facilities and 12 spray-painting facilitiesinvolving a total of 180workers. Increased levels of y-glutamyltransferase activity was observed to in fatal fulminant hepatic failure. Most well-designed epidemiologic studies of workers exposed to other. more commonly used solvents have found little evidence of hepatic dysfunction under adequately controlled conditions. Pulmonary Since 1985, two reports of solvents in relationship to pulmonary dysfunction. specifically disorders of the airways, have been published. Hawing et aP3 noted that asthmatics demonstrate a decline in forced expiratory volume in 1 second and symptoms of discomfort after inhalation challenge with a mixture of organic solvents. Lerman and Kipenx4reported a case of reactive airways dysfunction syndrome in a worker after acute exposure to a mixture of solvents; after the acute exposure episode. she continued to experience chest tightness when exposed to a variety of irritant substances. In summary, airways disorders (exacerbation of asthma or development of reactive airways dysfunction syndrome) have been as- sociated with solvent exposure. Cardiovascular Cardiovascular disorders in relationship to solvent exposure were recently reviewed by Wilcosky and Simonson." The role of carbon disulfide in the etiology of cardiovascular Although liver disease after expo- correlate with severity of solvent ex- disease is well recognized. Methylene sure to certain solvents such as carbon posure. after controlling for levels of chloride is metabolized to carbon tetrachloride is well recognized in the alcohol consumption and other fac- monoxide, leading to elevated blood medical literature. relatively few re- tors. concentrations of carboxyhemoglo- cent reports of solvent-induced liver More sensitive measures of liver bin; significant acute exposures can disease have appeared. With the ex- function-assessment of antipyrine aggravateexisting coronary artery dis- JOM Volume 36, Number 10, October 1994 1089 ease, in some cases possibly leading to fatal myocardial infarction. Mortality studies of methylene chloride-exposed workers have provided some evidence of increased cardiovascular disease rates. Other solvents have been associated with cardiac arrhythmias (eg. 1,1,1-trichloroethane and fluorocarbons). Fatal myocardial degeneration was reported in a man who chronically abused solvents for more than 5 yearss6 In summary, certain solvents have been associated with specific cardiovascular disorders: carbon disulfide and methylene chloride with coronary artery disease; I , 1,1 -trichloroethane and certain fluorocarbons with cardiac arrhythmias. Skin Dermatitis is well recognized as an effect of chronic cutaneous exposure to various solvents. Yakes et a]'' reported increased rates of dermatitis in solvent-exposed newspaper pressroom workers. Eighteen of 2 12 (8.5%) exposed workers had evidence of eczematous dermatitis on examination. Hematopoietic System Increased red blood cell volume and decreased red blood cell counts in car repair painters and mechanics were reported by Beving et a]." The significance of these findings is unclear. Related Issues Solvent Abuse Although the exposures are dramatically higher in solvent abusers than in occupationally exposed persons. some relevant information can be obtained from studies of solvent abuse. Morton89recently reviewed the literature on solvent abuse in children and adolescents. From that review and other literature, several points deserve attention. First. solvents have an inherent potential for acute intoxication and abuse. In the workplace. this potential may result in workers acting in such a way that exposure is increased so that the intoxicating effects. which some experience as a positive sensation, can be experienced. Sec- ond, chronic abuse has resulted in encephalopathy associated with cerebral atrophy, cerebellar degeneration. and peripheral neuropathy. As norcd above, these syndromes are associated with excessive occupational exposure. Finally, those who abuse solvents. particularly as adolescents, have been noted to abuse alcohol and other drugs, either simultaneously or in later life. Ethanol Consumption Certain solvent-exposed occupational groups (eg, construction painters) may consume alcohol to a greater extent than other occupational groups (eg, other construction workers) that are similar socioeconomically. Fidler et aIw observed significantly greater occasions of moderate and heaw drinking in construction painters than in drywall tapers who belonged to the same labor union, had similar levels of education, and who were of similar socioeconomic status. The reasons for this increased level of consumption are unclear. Some speculate that selection factors may operate: those who are moderate or heaw drinkers may choose an occupation (eg. painting) in which such behavior is common. Others relate drinking patterns to the circumstances (ie. social isolation) or the exposures (eg, solvent fumes) of the job by hypothesizing that these job factors may contribute to ethanol consumption. There are interesting similaritiesbetween the health effects of various industrial solvents and ethanol. Ethanol is in fact used as a solvent in certain industrial processes. Both industrial solvents and ethanol cause acute and chronic central nenous system disorders that have similar features. The syndromes of acute solvent and ethanol intoxication are both described as "being high." Chronic toxic encephalopathy due to solvents is characterized by neurobehavioral deficits similar to those seen after chronic ethanol intake (eg. psychomotor and short-term memory deficits).Abuse of both solvents and ethanol is well recognized. In addition to neurotoxicity, both may cause liver dysfunction. Exposure to solvents may poten- tiate the acute effect of ethanol consumption, possibly through competition for certain metabolic pathways. Chronic consumption of ethanol may potentiate the toxic effects of solvents (eg, somatosensory evoked potentials9'). Our understanding of the health effects of solvent exposure may be improved by comparative examination of the much more extensive literature on the chronic effects of ethanol consumption. Furthermore, prevention of solvent-induced disorders may benefit from examination of strategies that relate to ethanol consumption. Prevention Strategies As in all occupational disorders, the most basic strategy for prevention of exposure-induced disorders is to control exposure. In the case of solvent toxicity. substitution of non-solventcontaining products for those containing solvents has occurred. Particularly, substitution of water-based paints for solvent-based paints has occurred with attendant health benefits. However, in using water-based paints. the exposure hazards should also be carefully considered. If substitution of a less toxic product does not occur. exposure control through appropriate industrial hygiene measures should be accomplished. In selecting people to work in a solvent-exposed position. medical evaluations should be performed to identify those who have conditions that could be aggravated by solvent exposure (eg, asthma. peripheral neuropathy. skin disorders). This approach could be camed further in an attempt to identify persons with cognitive deficits on neurobehavioral testing: however. this has not occurred. If it were attempted. significant legal and ethical concerns would undoubtedly arise. Although no reports exist of excluding persons with a history of alcoholism or current heaw alcohol con- sumption from solvent exposure. a medical justification may exist for such action. Because heavy alcohol consumption may potentiate both the acute and chronic neurotoxic effects of solvents, persons with such a his- 1090 Review of Exposure to Solvents Baker tory could be assumed to be at increased risk. Again, significant legal and ethical considerations would apply to policies that exclude persons with heavy alcohol consumption from solvent-exposed jobs. Periodic biologic monitoring of solvent-exposed workers has a theoretic basis but has not occurred widely. Because some solvents are metabolized to substances that can be measured in blood, urine. or exhaled breath, biologic monitoring can be performed. The half-life of these metabolites is short and therefore sample collection must be timed carefully to relate directly to exposure periods. As a result, biologic monitoring is often impractical under normal working conditions. Use of neurobehavioral tests or other tests of organ system function (eg, peripheral nervous system or liver) for periodic testing of exposed workers has not occurred commonly in solvent-exposed populations. Both theoretic and practical considerations limit such use in most settings. Summary Prevention of solvent-induced disorders relies heavily on strategies designed to control exposure. The most important of these include substitution of less toxic products, use of ventilation systems, and use of personal protective equipment to reduce skin and respiratory absorption. References World Health Organization. Chronic E$ fects of Organic Solvents on [he Central Nervous System and Diagnostic Criteria. Copenhagen and Oslo: World Health Organization and Nordic Council of Ministers: 1985. Cranmer JM, Golberg L. eds. Neurobehavioral effects of solvents. Neurotoxicol 1986:7:1-95. Gerr F. Letz RE. Health effects of occupational and environmental exposure to solvents. In: Rom WN, ed. Environmental and Occupational Medicine. 2nd ed. Boston: Little. Brown: 1992. Baker EL. White W. The use of neuropsychologicaltesting in the evaluation of neurotoxic effects oforganic solvents. In: Chronic Eflects of Organic Solvents on the Central Nervous System and Diagnostic Criteria. Copenhagen and Oslo: World Health Organization and Nordic Council of Ministers: 1985:219-242. 5. Bleecker ML, Bolla KI. Agnew J, Schwartz BS. Ford DP. Dose-related subclinical neurobehavioral effects of chronic exposure to low levels of organic solvents. Am J Ind Med. 1991:19: 71 5-728. 6. Spurgeon A. Gray CN. Sims J. et al. Neurobehavioral effects of long-term occupational exposure to organic solventstwo comparable studies. Am J Ind Med. 1992. in press. 7. Hanninen H. Antti-Poika M, Juntunen J, Koskenvuo M. Exposure to organic solvents and neuropsychological dysfunction: a study on monozygotic twins. Br J Ind `Wed. 199 I ;48:18-25. 8. Moen BE, Kyvik KR. Engelsen BA. Riise T. Cerebrospinal fluid proteins and free amino acids in patients with solvent induced chronic toxic encephalopathy and healthy controls. Br J Ind .Wed. 1990:47:277-280. 9. Bolla KI. Schwartz BS, Agnew J. Ford PD. Bleecker ML. Subclinical neuropsychiatric effects of chronic low-level solvent exposure in US paint manufacturers. J Occup Med. 199032:671-677. IO. Ng TP. Ong SG. Lam WK, Jones GM. Neurobehavioral effects of industrial mixed solvent exposure in Chinese printing and paint workers. .Veurotoxicol Teraiol. 1990;l2:661-664. I I . van Vliet C. Swaen GM, Volovics A, et al. Exposure-outcome relationships between organic solvent exposure and neuropsychiatric disorders: results from a Dutch case-control study. Am JIndMed. 1989;16:707-718. 12. Malllish NA. Fine U. Albers JW, Whitehead L. Langolf GD. A neurological evaluation of workers exposed to mixtures of organic solvents. Er J Ind Med. 1987:44:14-25. 13. Baker EL. Letz RE. Eisen EA. et al. Neurobehavioral effects of solvents in construction painters. J Occup Med. 1988:30:1 16-123. 14. Gade A, Mortensen EL, Bruhn P. `Chronic painter's syndrome.' A reanalysis of psychological test data in a group of diagnosed cases, based on comparisons with matched controls. Acta i V e t d Seand. I988:77:293-306. 15. Mikkelsen S . Jorgensen M, Browne E. Gyldensted C. Mixed solvent exposure and organic brain damage. A study of painters. .4cra .Veurol Scand Suppl. 1988: 118:1-143. 16. Fidler AT. Baker EL. Letz RE. Estimation of long term exposure to mixed solvents from questionnaire data: a tool for epidemiological investigations. Br J Ind ,Wed. 1987;44:133-141. 17. Morrow LA, Ryan CM, Hodgson MJ. Robin N. Alterations in cognitive and psychological functioning after organic solvent exposure. J Occup Med. 1990; 32:444-450. 18. Monson RR. Occupational Epidemiol- ogy. Boca Raton, Fla: CRC Press; 1985. 19. White RF. Feldman RG. Neuropsycho- logical assessment of toxic encephalop athy. Am J Ind .Wed. 1987:11:395-398. 20. Orbaek P, Lindgren M. Prospective clinical and psychometric investigation of patients with chronic toxic encephalop athy induced by solvents. Scand J Work Environ Health. 1988:14:37-44. 21. Gregersen P. Neurotoxic effects of or- ganic solvents in exposed workers: two controlled follow-up studies after 5.5 and 10.6 years. Am J Ind Med. 1988:14: 68 1-701. 22. Gregersen P, Klausen H, Elsnab CU. Chronic toxic encephalopathy in solvent-exposed painters in Denmark 1976-1 980: clinical cases and social consequences after a 5-year follow-up. .4m J IndMed. 1987:11:399-417. 23. Edling C, Ekberg K. Ahlborg G. et al. Long-term follow up of workers exposed to solvents. Br J Ind Med. 199047: 75-82. 24. Morrow LA, Ryan CM, Hodgson MJ. Robin N. Risk factors associated with persistence of neuropsychologtcal deficits in persons with organic solvent exposure. J Nen? Ment Dis. 1991;179: 540-545. 25. Brackbill RM, Maizlish N. Fischbach T. Risk of neuropsychiatric disability among painters in the United States. Seand J Work Environ Health. 1990:16: 182-188. 26. Axelson 0,Hane M, Hogstedt C. A case- referent study on neuropsychiatric disorders among workers exposed to solvents. Scand J Work Environ Health. 1976:2:14-20. 27. O'Rynn RR, Monkman SM, Waldron HA. Organic solvents and presenile dementia: a case-referent study using death certificates. Br J Ind Med. 1987:44: 259-262. 28. Littorin ME, Fehling C. Attewell RG. Skerfving S. Focal epilepsy and exposure to organic solvents: a case-referent study. J Occup Med. 1988:30:805-808. 29. Flodin U, Soderfeldt B, Norrlind-Brage H, et al. Multiple sclerosis. solvents and pets. A case-referent studv. Arch .Veurol. I988:45:620-623. 30. Hawkes CH. Cavanagh JB, Fox kl. Motoneuron disease: a disorder secondary to solvent exposure? Lancer 1989:1: 73-76. 31. Schwartz BS. Ford DP. Bolla KI. Agnew J. Rothman N, Bleecker ML. Solvent- associated decrements in olfactory function in paint manufacturing workers. Am J Ind Med. 1990:18:697-706. 32. Sandmark B, Broms I. Lofgren L, Ohl- JOM Volume 36, Number 10, October 1994 1091 son CG. Olfactory function in painters workers. Am J Ind 'Wed. 19!%18: and adverse pregnancy outcomes. Br J exposed to organic solvents. Scand J 697-706. Ind Med. 1988;45:193-197. Work Environ Health. 1989:1T60-63. 33. Mergler D, Blain L. Assessing color vi- 46. Mergler D. Blain L. Assessing color vi- 6 I . Brender JD. Suarez L. Paternal occupa- sion loss among solvent-exposed work- tion and anencephaly. Am J Epidemiol. sion loss among solvent-exposed work- ers. Am JInd Med. 1987:12:195-203. 199013i:5 17-521. ers. Am JIndMed. 1987:12:195-203. 47. Baker EL. Neurological disorders. In: 62. Kelsey KT, Wiencke JK, Little FF. 34. Ledin T, Odkvist LM, Moller C. Postu- Rom WN, ed. Environmental and Oc- Baker EL, Jr, Little JB. Sister chromatid rography findings in workers exposed to cupational Medicine, 2nd ed. Boston: exchange in painters recently exposed industrial solvents. k t a Otolaryngol Little, Brown: 1992. to solvents. Environ Res. 1989;50 (Stockh). 1989;107:357-36 1. 48. Demers RY, Markell BL. Wabeke R. 248-255. 35. Ledin T, Jansson E. Moller C, Odkvist Peripheral vibratory sense deficits in sol- 63. SalamancoGomez F, Moreta G, Palma LM. Chronic toxic encephalopathy in- vent-exposed painters. J Occup 'Wed. V, et al. Cytogenetic study in children vestigated using dynamic posturography. I99 I ;33:105 1-1054. chronically habituated to inhalation of Am J Otolaryngol. 1991:12:96-100. 49. Bove FJ, Letz R, Baker EL. Sensory volatile substances. Am J Med Genet. 36. Moller C, Odkvist LM, Thell J, Larsby thresholds among construction trade 1987;27:391-397. B, Hyden D. Otoneurological findings in painters: a cross-sectional study using 64. Olsson H, Brandt L. Risk of non-Hodg- psycho-organic syndrome caused by in- new methods for measuring temperature km's lymphoma among men occupa- dustrial solvent exposure. Acta Otolar- and vibration sensitivity. J Occup Med. tionally exposed to organic solvents. yngol (Stockh). 1989;107:5-12. 1989;31:320-325. Scand J Work Environ Health. 1988;14: 37. Dager SR, Holland JP, Cowley DS, Dun- 50. Morrow LA. Ryan CM, Goldstein G. 246-25 1 . ner DL. Panic disorder precipitated by Hodgson MJ. A distinct pattern of per- 65. Blair A, Stewart PA, Tolbert PE, et al. exposure to organic solvents in the work sonality disturbance following exposure Cancer and other causes of death among place. Am J Psychiatry. 1987;144: to mixtures of organic solvents. J Occup a cohort of dry cleaners. Br J Ind Med. 1056-1058. .Wed. 1989;31:743-746. 199047:162-168. 38. Orbaek P, Rosen I, Svensson K. Electro- 51. Ryan CM, Morrow LA. Hodgson M. 66. Brandt L, Kolstad H, Rasmussen K. neurographic findings in patients with Cacosmia and neurobehavioral dysfunc- Health risks of dry cleaning. UgeskrLae- solvent induced central nervous system tion associated with occupational expo- ger. l987;149:3 19-323. dysfunction. Br J Ind Med. 1988;45: sure to mixtures of organic solvents. .4m 67. Hernberg S. Kauppinen T, Riala R, Kor- 409-4 14. J Psychiatry. 1988;145:1442-1445. kala ML, Asikainen U. Increased risk for 39. Hazemann P, Jeftic M. Lille F. Somato- 52. McCrank E, Rabheru K. Four cases of primary liver cancer among women ex- sensory evoked potentials in alcoholics progressive supranuclear palsy in pa- posed to solvents. Scand J WorkEnviron and patients occupationally exposed to tients exposed to organic solvents. Can J Healrh. 1988;14:356-365. solvents and lead. Electromyogr Clin Psychiatry. 1989;34:934-936. 68. Brandt L, Kristoffersson U. Olsson H. Neurophysiol. 1987;27:183-187. 53. Monstad P. Nissen T. Sulg IA, et al. Mitelman F. Relation between occupa- 40. Massioui JFE. Lille F, Lesevre N, Haze- Sleep apnea and organic solvent expo- tional exposure to organic solvents and mann P, Gamier R. Dally S. Sensory sure. J ,Veurol. 1987;234:152-154. chromosome aberrations in non-Hodg- and cognitive even related potentials in 54. Lindbohm ML. Taskinen H. Sallmen M. km's lymphoma. Eur J Haernatol. 1989: workers chronically exposed to solvents. Hemminki K. Spontaneous abortions 42:298-302. J Toxicol Clin Toxicol. 199028: among women exposed to organic 69. Hashimoto DM. Kelsey KT. Seitz T. et 203-2 19. solvents. Am J Ind .Wed. 1990;17: al. The presence of urinary cellular sedi- 41. Moen BE, Kyvik KR. Engelsen BA. 449-463. ment and albuminuria in newspaper Riise T. Cerebrospinal fluid proteins and 55. Lipscomb JA. Fenster L. Wrensch M. pressworkers exposed to solvents. J Oc- free amino acids in patients with solvent Shusteman D. Swan S. Pregnancy out- cup .Wed. 1991:33:516-526. -n :h 1: J. re :>. .ge induced chronic toxic encephalopathy and health controls. Br J Ind .Wed. 1990: 47:277-280. 42. Barregard L. Wikkelso C. Rosengren LE, et al. Cerebrospinal fluid proteins in men with chronic encephalopathy after exposure to organic solvents. Scand J Uork Environ Health. 1990;16:423-427. 43. Orbaek P. Lindgren M, Olivecrona H. Haeger-Aronsen B. Computed tomography and psychometric test performances comes in women potentially exposed to occupational solvents and women working in the electronics industry. J Occup .Wed. 1991;33:597-604. 56. McDonald JC. Lavoie J. Cote R. McDonald AD. Chemical exposures at work in early pregnancy and congenital defect: a case-referent study. Br J Ind .%fed. 1987:44:527-533. 57. Tikkanen J. Heinonen OP. Cardiovascular malformations and organic solvent 70. H o u P. Pilliod J. Soderstrom D. et al. Relation between renal function tests and a retrospective organic solvent enposure score. Br J Ind ]Wed. 1989:46: 815-819. 7 I . Hamngton JM. Whitby H. Gray CN. et al. Renal disease and occupational exposure to organic solvents: a case referent approach. Br J Ind Med. 1989:46: 643-650. nd in patients with solvent induced chronic exposure during pregnancy in Finland. 72. Daniell WE. Couser WG. Rosenstock L. Ill. toxic encephalopathy and healthy con- Am J Ind.Cfed. 1988:14:1-8. Occupational solvent exposure and glo- trols. Br J Ind .Wed. l987:44: 175-179. 58. Eskenazi B. Gaylord L. Bracken MB. merulonephritis. A case report and re- Io- 44. Hagstadius S. Orbaek P. Risberg J. Lind- Brown D. In utero exposure to organic view of the literature. J.4M.4. 1 9 8 8 3 9 : It-) gren M. Regional cerebral blood flow at solvents and human neurodevelop- 2280-1283. ':I: the time of diagnosis of chronic toxic ment. Dev .2fpd Child .Velcro/. 1988:30: 73. Navarte J, Saba SR. Ramirez G . Occu- encephalopathy induced by organic-sol- 192-50 1. pational exposure to organic solvents vent exposure and after the cessation of 59. Eskenazi 8. Bracken MB. Holford TR. causing chronic tubulointerstitial ne- nt- exposure. Scand J Ubrk Environ Health. Grady J. Exposure to organic solvents phritis. Arch Intern ,Wed. 1989i149: nc- 1989;15:130-135. and hypertensive disorders of pregnancy. 154-158. -Im 45. Schwartz BS, Ford DP. Bolla KI. et al. ,4m JIndMed. 1988;14:177-188. 74. S e w R. Stolley PD. Salgado N. Pereira Solvent-associated decrements in olfac- 60. Daniell WE. Vaughan TL. Paternal em- AB, Ramos OL. Exposure to hydrocar- )hl- tory function in paint manufacturing ployment in solvent-related occupations bons and rapidly progressive glomerulo- 1092 Review of Exposure to Solvents Baker nephritis. Braz J Med Bioi Res. 1990; 231225233, 75. Redlich CA, Beckett WS, Sparer J, et al. Liver disease associated with occupational exposure to the solvent dimethylformamide. Ann Inrern Med. 1988;108: 680-686. 76. Harrison R. Letz G . Pasternak G, Blanc P. Fulminant hepatic failure after occupational exposure to 2-nitropropane. Ann Inrern Med. 1987;107:466-468. 77. Hodgson MJ, Hey1 AE, Van-Thiel DH. Liver disease associated with exposure to 1, I , 1-trichloroethane. Arch Intern Med. 1989:149: 1793-1 798. 78. Cordes DH. Brown WD, Quinn KM. Chemically induced hepatitis after inhaling organic solvents. War J Med. 1988: 148:458-460. 79. Chen JD. Wang JD, Jang JP, Chen W. Exposure to mixtures ofsolvents among paint workers and biochemical alterations of liver function. Br J Ind Med. 1991:43:696-701. 80. Dossing M. Changes in hepatic microsomal enzyme function in workers ex- posed to mixtures of chemicals. Clin Pharmacol Ther. 1982;32:340-346. 81. Franco G. Fonte R, Tempini G, Candura F. Serum bile acid concentrations as a liver function test in workers occupation- ally exposed to organic solvents. In1 Arch Occup Environ Health. 1986;58: 157- 164. 82. Edling C, Tageson C. Raised serum bile acid concentrations after occupational exposure to styrene: a possible sign of hepatotoxicity. Br J Ind Med. 1984:8: 137- 140. 83. Harving H. Dah1 R, Molhave L. Lung function and brochical reactivity in asthmatics during exposure to volatile organic compounds. Am Rev Respir Dis. 199I :143:75 1-752. 84. Lerman S, Kipen H. Reactive airways dysfunction syndrome. .tm Fam Physician. 1988;38:135-138. 85. Wilcosky TC. Simonson NR. Solvent exposure and cardiovasculardisease. Am J Ind .Wed. 199I ; 19:569-586. 86. Claydon SM. Myocardial degeneration in chronic solvent abuse. Med Sci Law. 1989:28:2 17-2 18. 87. Yakes B, Kelsey KT. Seitz T. Hashimoto D. Feldman HA, Christiani DC.Occupational skin disease in newspaper pressroom workers. J Occup Med. 1991;33: 7 I 1-7 17. 88. Beving H, Tornling G. Olsson P. In- creased erythrocyte volume in car repair painters and car mechanics. Br J Ind .Wed. 1991;48:499-501. 89. Morton HG. Occurrence and treatment of solvent abuse in children and adolescents. Phannacol Ther. 1987;33: 449469. 90. Fidler AT, Baker EL, Letz RE. Neuro- behavioral effects of occupational exposure to organicsolventsamong construction painters. Br J Ind Med. 1987:44: 292-308. 9 1. Hazemann P. Jeftic M. Lille F. Somatosensory evoked potentials in alcoholics and patients occupationally exposed to solvents and lead. Electromyogr Clin ,Veurophysiol. 198727:183-187. 92. Parkinson DK, Bromat EJ. Cohen S. Dunn LO, Dew MA, Ryan C, Schwartz JE. Health effects of long-term solvent exposure among women in blue-collar occupations. Am J Ind .Wed 1990:17: 66 1-675. 93. Cherry N. Hutchins H. Pace T. Waldron HA. Neurobehavioural effects of repeated occupational exposure to toluene and paint solvents. Br J Ind .Med 1985;42:29 1-300. 94. Maizlish NA. Langolf GD. Whitehead LW. Fine U.Albers JW. Goldberg J. Smith P. Behavioural evaluation of workers exposed to mixtures of organic solvents. B r J Ind .Wed 1985:42:579-590. Interesting Quotations . . .the case was only one incident in one physician's experience. Although his memory of that experience remains vivid. it is best viewed as an anecdote from which it would be a mistake to generalize. The plural of "anecdote" is not "data." From Enemies of Patients, by R. Macklin. New York: Oxford Press; 1993, p 8.