Zamyslowska-Szmytke, Ewa MD, PhD; Politanski, Piotr MSc; Sliwinska-Kowalska, Mariola MD, PhD
Vertigo and dizziness have been previously introduced to the natural history of chronic toxic encephalopathy (CTE).1 Already in the 80s, animal studies suggested that vestibular and optooculomotor system tests may be used for assessing the influence of solvents on the central nervous system (CNS). Tham et al2 showed that organic solvents had a subclinical excitatory (benzene and toluene) or depressive (trichloromethane) effect on the vestibulooculomotor reflex (VOR), while some other, for example, hexane, did not influence the VOR at concentrations lower than those known to be toxic to CNS.
It has been suggested that depression of the VOR may be a very early sign of general depression of the reticular formation in the CNS.3 In 2004, Pascual4 reported a decreased number of neural cells in several brainstem regions, including vestibular cochlear nuclei in rats exposed to toluene. Excitation of the VOR may be the result of lower cerebellar control in exposed animals.2 Similar conclusions were drawn by Ödkvist et al5 on the basis of their electronystagmography (ENG) study in healthy volunteers exposed to toluene and styrene.5
The results of the studies on workers with long-term exposure to organic solvents are equivocal. In the majority of human studies, workers were examined because of suspected or confirmed CTE. Those patients were characterized by the lesions of numerous parts of CNS, including the structures of posterior fosse (brainstem and cerebellum). Visual Suppression of Postrotatory Nystagmus and Smooth Pursuit (SP) were the most relevant tests to reveal the solvent-induced lesions.6–9 In patients with CTE, the SP test displayed the significantly lower gain compared with the controls, the visual suppression tests displayed the higher gain. Square waves occurred much more frequently in exposed group than in the reference group. Hyperreactivity was also observed during the caloric testing.9 In the group of exposed subjects with CTE, magnetic resonance imaging scans revealed brain atrophy in the posterior fosse structures, the parietal lobes, and pericisternal white matter.7 In three-dimensional brain metabolic imaging of patients with CTE, Callender et al10 showed most frequent abnormalities in temporal lobes (67.7%), frontal lobes (61.3%), and basal ganglia (45.2%).
There are only few studies where non-CTE workers with long-term exposure to organic solvents were involved. In the Copenhagen Epidemiological Questionnaire study11 comprising 3387 men exposed to organic solvents for more than 5 years, a trend toward a higher prevalence of vertigo was found in a subgroup occupationally exposed to solvent. Decreased concentration and defective memory were the most frequent complaints in this subgroup. In the population of paint-and-lacquer–factory workers, Sulkowski et al12 observed decreased duration, amplitude, and slow-phase velocity (SPV) of rotatory-induced nystagmus in the ENG evaluation. The findings correlated with the total exposure to the solvent mixture. However, the age was not included in the analyses. On the contrary, Antti-Poika et al13 did not notice any dependency between ENG findings and exposure parameters, although the findings classified as abnormal were slightly more common in the exposed than in control group.
Static posturography is a useful tool to measure postural balance and functional aspects of nervous system. Static posturography comprises tests revealing somatosensory input and visual interference with functions of VOR and balance control. In dynamic posturography, some tools to assess the active balance control on moving platform and with visual conflict are added. Posturography findings in the volunteers exposed to m-xylene (200 ppm) revealed longer reaction time (RT) and greater postural sway than in nonexposed group.14 The similar abnormalities were found in CTE workers exposed for many years to xylene and toluene. In non-CTE workers, the results were inconclusive.13
Kuriwaka et al15 observed that long-term solvent exposure produced disturbances in dopaminergic neurons of the basal ganglia, resulting in the loss of postural reflexes, retropulsion, and gait abnormalities, which were sensitive to levodope.
Taking into account the relatively limited number of studies on chronically low-exposed workers, the aim of this study was to assess the balance system in workers occupationally exposed to low concentrations of organic solvent mixture at the workplace.
SUBJECTS AND METHODS
In total, 170 workers were examined. This group comprised 60 workers of a paint-and-lacquer factory (42 men and 18 women), aged 38 ± 7 years, and exposed to organic solvents, predominantly to xylene, toluene, and ethyl acetate. The reference group consisted of 110 nonexposed subjects (71 men and 39 women), aged 37 ± 9 years, and white collars and maintenance service, working in the same factory. Mean age did not differ significantly between the groups. In reference group, ENG test was performed in a subgroup of only 23 subjects. Study groups and subgroup are described in Table 1.
The questionnaire applied in the study included detailed inquires on medical history, occupational history, lifestyle, and ototoxic exposures not related to current or former occupation. Medical history evaluated vertigo and dizziness, subjective hearing impairment, past middle-ear diseases and surgery, chronic systemic diseases, cholesterol levels, arterial hypertension, head trauma, and current and past medications having an ototoxic potential. The detailed questionnaire was published earlier.16 The authors had an access to database containing hearing evaluation data of the workers. The careful medical examination and hearing assessment were used to exclude the nonoccupational balance disorders. The following inclusion criteria were applied to the study: no history of middle-ear diseases; normal tympanic membrane appearance on ENT examination; presence of tympanogram type A; presence of stapedial reflex; and no signs of retrocochlear hearing loss.
Subjects selected for the study were working in contact with solvents. They were mill operators, colorists, laboratory-of-quality supervision workers, dry-component mixers, and dispenser operators. The exposure assessments were based on data provided by local authorities responsible for hygienic and safety control in the enterprise. Moreover, an individual dosimetry method was used to evaluate current occupational exposures in 1999. The eluate was analyzed with gas chromatography (gas chromatograph HP-5890 series II, Hewlett Packard, Palo Alto, CA). Detailed description of solvent assessment procedure is available elsewhere.16 In the paint-and-lacquer company, the study participants were exposed to a solvent mixture with xylene isomers (14% of the mixture compounds), toluene (4%), ethyl acetate (36%), and white spirit as main compounds. The content of remaining chemicals constituted very low proportion of the entire mixture mass.
As a given employee was exposed to different solvent concentrations at various workplaces and during various employment periods, individual exposure over the whole working life was evaluated. Two cumulated exposure indices—cumulated concentration of given solvent and cumulated exposure index for all compounds of a mixture (a sum of the measured air concentration of a given chemical divided by its occupational exposure limit [OEL] value)—were calculated.16 Since the employees were exposed to different solvent concentrations at different workplaces and during various employment periods, the total working life exposure was calculated according to the following formula: [(a1 × b1) + …(an × bn)], where ai (i = 1…n) − solvent concentration (or exposure index for the mixture) at a given place of work, over the time period between two consecutive measurements of exposure, bi (i = 1…n) − period of employment (in years) with solvent concentration (or exposure index) of ai. To assess mean work life exposure, the cumulated concentration (or cumulated exposure index) was divided by total years of exposure of every employee. The mean (averaged) solvent exposure greater than the Polish OEL, or exposure index for a mixture exceeding one, indicated overexposure to solvents. In the solvent-exposed group, 22% of the subjects were overexposed to solvent mixture, while the mean values of particular compounds (toluene, 8.0 mg/m3 and xylene, 11.3 mg/m3) did not exceed the maximum permissible level for 8-hour day per 5 days per week in Poland (100 mg/m3). The details of exposure are presented in Table 1.
In the subgroup of 50 workers, urine samples were collected at the end of working shift from the last 4 hours. Methylhippuric acid concentration (g/dm3) (xylene metabolite) was assessed by gas chromatography.
A daily noise-exposure level for a nominal 8-hour working day (LEX8h) for each individual worker was obtained from factory records. The mean LEX8h for the study group was 82 dBA (SD = 3). Fifteen percent of subjects were exposed to noise levels within the 85 dBA admissible in Poland. The details of exposure are presented in Table 1.
Hearing and Balance Assessment
Air conduction pure-tone thresholds for the frequencies 1 to 8 kHz were obtained bilaterally by using an Interacoustics Traveller AA 222 audiometer/middle-ear analyzer with TDH-39P headphones. Tympanometry was carried out bilaterally with the same equipment. Hearing examination was performed at least 16 hours after the last exposure to noise. Because the majority of the factories’ outpatient clinics were not equipped with sound-proof booths, all of the solvent-exposed group and the unexposed control group were examined in a quiet office room where the background noise did not exceed 30 dB sound pressure level. The audiograms were classified as normal when none of the single-hearing thresholds in both ears exceeded 25 dB hearing level (HL).
The test comprised two parts: Modified Clinical Test of Sensory Integration on Balance (mCTSIB) and Limit of Stability (LOS). In mCTSIB, the central integration of the vestibular, visual, and somatosensory inflow was evaluated. Four testing conditions were introduced: (1) Romberg trial on the firm surface with eyes open (EO, firm); (2) Romberg trial on the firm surface with eyes closed (EC, firm); (3) the trial on the foam with eyes open (EO, foam); and (4) the same with eyes closed (EC, foam). The fourth trial abnormalities are characteristic for vestibular involvement. Computer recorded the sway velocity in every condition (degrees per second). In LOS test, patients should actively move the center of gravity in a specified direction. Patient started moving immediately after sound signal, his or her center of gravity should reach the specified point as quickly and precisely as possible. The following parameters were calculated: RT (seconds)-–time between the sound and movement; directional control (DCL) (%)-–percentage of movement in specified direction; maximal movement velocity (degrees per second); endpoint (EPE) (%)-–the percentage of movements when the subject's center of gravity reached specified point and he or she could keep that posture for some time; maximal excursions (%)-–percentage of sway required by the software. Tests were performed by using Basic Balance Master (NeuroCom Int, Inc, Clackamas, OR).
The results were compared to the standards supplied by equipment producer and were automatically adjusted for age, height, and gender. All tests were performed by the same person.
In ENG testing, the protocol routinely designed for occupational medicine purposes was used; no fixation and rotation tests were performed. Testing protocol comprised the following.
1. Saccade test-–fast voluntary eye movement registration when the eyes moved between two randomly alternating lights 20° apart from central position (right–left 40°); undershoots and overshoots were assessed only qualitatively.
2. Smooth Pursuit test —when the ability of the subject to follow a moving object was tested. The dot moved sinusoidally with the frequencies of 0.1 and 0.5 Hz, respectively, in two tests, with amplitude of 20° to the right and left from central position. The gain was calculated as a ratio of eye velocity and target velocity. Normative values of gain were not lower than 0.6.17
3. Optokinetic (OKN) test for full field stimulus was performed; vertical strips were moving on a circular screen with velocity 20 degrees per second. The gain between SPV and target velocity was analyzed. Normal value of gain exceeded 0.5.17
4. The analysis of spontaneous and positional nystagmus assessed in supine, supine right and left, and rose positions. Nystagmus was classified as central or peripheral according to Nylen degree; moreover, the latency, duration, and slow phase direction were essential to distinguish between central and peripheral nystagmus. Second-degree nystagmus was assumed as the most characteristics for peripheral lesions.18
5. Bithermal caloric test using warm (44°C) and cold (30°C) water was performed in the darkness according to Hallpike protocol. The SPV and duration time with latency (from the beginning of irrigation to 3-second period without beats) were assessed. The caloric asymmetry (CP), directional preponderance (DP), and reactivity (expressed as mean time of caloric nystagmus in trials right ear 30°, left ear 30°, right ear 44°, and left ear 44°) were calculated. According to literature data,19 normal values of caloric parameters were as follows: CP SPV = 19%; DP SPV = 21%; CP time = 10 seconds; and DP time = 18 seconds.
Tests were performed by using Electronystagmograph AM produced by Farum Medical Equipment Manufacture. All tests were performed by the same technician. The ENG waveforms were recorded on a system comprising paper chart recorder with a paper speed of 15 mm/s. Calibration was adjusted such that the subject's 20° eye movement produced 10-mm waveform on a chart. Caloric-induced–slow-phase nystagmus velocity was calculated by one examiner as average SPV of three to five beats within the period of maximum activity. The CP was calculated by using the formula of Jongkees.
Covariance analysis was used to compare the mean values of variables adjusted for age and gender (tests with multiple comparison). For frequency analysis, chi-square test or Fisher test was used. The linear multiple regression test (analysis of variance [ANOVA]) was incorporated to establish the linear relationships between exposure parameters (mean cumulated concentration of the solvents, exposure index, and metabolite excretion) and balance test results within entire exposed group; age and gender were confounding factors in ENG analyses, and height, age, and gender were confounding factors in posturography analyses.
The results were assumed to be statistically significant at P < 0.05. Statistica 6 pl (StatSoft Polska, Krakow, Poland) and Systat 11 (Systat Software, Inc, Chicago, IL) software were used for testing.
Audiograms for each individual were classified as normal or abnormal. An abnormal audiogram was defined as one showing at least one audiometric threshold for a single frequency in either ear greater than 25 dB HL. According to this definition, 42% of solvent-exposed and 27% of nonexposed subjects had an abnormal audiogram. The differences between groups were statistically significant. Solvent-exposed subjects exhibited significantly poorer audiometric hearing thresholds than nonexposed subjects at 3000, 4000, and 6000 Hz in the right ear (Fig. 1).
Vertigo or dizziness was reported by 5 (8%) of solvent-exposed and 14 (12%) of nonexposed subjects; the frequency of these symptoms was not significantly different between groups.
The first part of the test (mCTSiB) did not reveal differences in the frequency of abnormal results between solvent-exposed and control groups (Table 2). There were also no statistically significant differences of mean values of sway velocities between exposed and nonexposed workers (Fig. 2). In the LOS tests, abnormal RT was more common (30%) in solvent-exposed group than in nonexposed group (9%) (Table 2). Mean value of RT was also longer in solvent-exposed subjects (mean, 0.8; SE, 0.02 seconds vs mean, 0.7; SE, 0.03 seconds; P < 0.05) (Fig. 2a). Mean values of the following parameters were significantly poorer in solvent-exposed group than in nonexposed: movement velocity (mean, 3.9; SE, 0.13 degrees per second vs mean, 4.8; SE, 0.20 degrees per second), EPE (mean, 79.7; SE, 1.66% vs mean, 83.9; SE, 0.89%) and directional control (mean, 85.3; SE, 0.46% vs mean, 84.0; SE, 0.35%, respectively) (Figs. 3a and 3b).
Frequencies of abnormal results in exposed and nonexposed groups are presented in Table 3. Square waves were observed in exposed group only (23% of subjects). Abnormal gains of OKN were more common in exposed group (53% vs 26%) (P < 0.05); moreover, the mean values of SPV in OKN test were slower in this group than in nonexposed controls; for the right-side movement, the difference reached statistical significance (mean, 7.1; SE, 0.02 degrees per second vs mean, 8.3; SE, 0.04 degrees per second in exposed and nonexposed groups, respectively) (Fig. 4). In SP test, abnormal gain was also more frequent in exposed group (43% vs 22%) but the difference was not significant (P < 0.1). Second-degree nystagmus according to Nylen's classification was observed in 22% of solvent-exposed workers and 26% of nonexposed individuals. In caloric test, such parameters as CP-time, CP–SPV, and DPs did not differ in solvent-exposed group versus nonexposed group (Table 3). Mean values of nystagmus time were slightly lower in three trials (R30°, L30°, R44°) and significantly lower in the fourth (L44°). Reactivity (mean value of caloric response of all trials) was significantly lower in solvent-exposed group than in nonexposed group (mean, 136; SE, 2.9 seconds vs mean, 148; SE, 3.3 seconds, respectively) (Fig. 5). Mean values of reactivity calculated for SPV were similar in two groups of examined workers (17.1 degrees per second and 17.0 degrees per second).
The Relationships Between Organic Solvent Exposure and Balance Tests Results
The results of linear regression analysis within solvent-exposed group are presented in Table 4. Cumulated toluene concentration was negatively correlated with caloric nystagmus time in trials L44°, L30°, and R30° (not significant) and total response expressed as reactivity. Cumulated toluene concentration was also correlated with DP calculated from nystagmus time and SPV value. Moreover, cumulated exposure index was correlated with DP SPV. There were no relations between xylene exposure parameters and ENG tests results. In static posturography, mean exposure index was correlated with RT in LOS test and methylhippuric excretion with sway velocity of the first trial (EO, firm) of mCTSiB.
Occupational environmental monitoring indicated that employees of paint-and-lacquer factory were exposed to rather moderate concentrations of solvents, with toluene, xylene, ethylbenzene, and white spirit as the major compounds. Mean concentrations of toluene and xylene were 8 mg/m3 and 11 mg/m3, respectively. In Poland, OELs for both solvents are 100 mg/m3. Mean exposure index was 0.6 while overexposure starts over 1. Mean noise exposure (82 dB) did not exceed the normative value of 85 dBA, although 15% of workers were overexposed to noise. Despite exposure mostly below normative values, hearing impairment was more frequent (42%) in solvent-exposed than in nonexposed group (27%). Hearing impairment more frequent due to solvent exposure was detected in our earlier study,16 although in our current study, it is quite likely that hearing impairment might be due to the combined exposure to noise and solvents. However, considering that the main goal of our study was the influence of organic solvents on balance tests, interactions between solvents and hearing were given less attention.
The main aim of our study was the assessment vestibulooculomotor, celebellooculomotor, and vestibulospinal reflexes and the posture control in workers occupationally exposed to organic solvent mixture.
Long-term exposure to organic solvents may evoke dizziness, which is common in CTE.1,11 Vertigo, as characteristic for vestibular lesions, may be rather the result of acute exposure. In our study, the insignificant number of vertigo and balance disturbances in employees of lacquer factory may be explained by the long time of exposure and slow onset, which allows sufficient activation of neural plasticity of the CNS to permit for vestibular compensation. This compensation phenomenon may also explain our results of posturography, in which no increased sway was observed in any test of mCTSiB in solvent-exposed group as compared with the nonexposed group. In our study, posturographic abnormal results were observed in LOS test, where psychomotor skills and attention may affect the results of basic (vestibulospinal) reflex assessment. In solvent-exposed group, longer RTs and inability to sway the body in certain direction and to reach the EPE were more frequent than in nonexposed group. Longer RT was also observed in CTE group in the study by Niklasson et al,9 in which this parameter correlated with some behavioral tests results as well as solvent-exposure duration. In our study, the linear relation between RT and mean exposure index was at the borderline of significance (P = 0.05). The psychomotor test abnormalities are very common in subjects with organic brain lesions; however, in our study, similar results were observed in exposed workers with normal result of neurological examination. Moreover, we found a positive correlation between methylhippuric acid (xylene metabolite) urine concentration and sway velocity in the first mCTSiB trial (EO, firm), which may suggest the relation between the postural control ability and exposure to xylene.
The majority of studies concerning the influence of solvent exposure on balance control are performed in healthy volunteers or workers suspected to CTE. Smith et al20 described one of the few studies in group of workers exposed for a long time to low concentrations of solvents, namely jet fuel (naphtha and low proportions of benzene, toluene, and xylene). They found a statistically significant association between solvents, as cumulated concentration of benzene, toluene, and xylene, and increased postural sway. The strongest association was noticed between JP-8 benzene and sway length in fourth posturography trial (EC, foam). He also identified several cofactors in the regression modeling—the most important was gender, age, and caffeine intake. However, in his study, age was incorporated to cumulated exposure parameters, so the interpretation of the relations between tests results and exposure should be done with caution. There was no reference group in the study by Smith et al.20 Also in the study by Iwata et al 21 among the solvent workers, transversal and sagittal sways with EO and tremor intensity were significantly related to toluene exposure. On the contrary, Niklasson et al9 in their CTE workers group did not find any relations between body sway and exposure duration or intensity.
Our ENG findings in the group of workers exposed to organic solvent mixture referred mostly to the central part of balance system. In solvent-exposed group, tests revealed the following results: more common square waves; slightly lower gain in smooth pursuit test; and significantly lower gain in OKN. Symmetrical low-degree gain reduction in SP may be the sign of cerebellar abnormalities, but these test results are strongly dependent on age, gender, drug intake, and problems with attention, especially in older subjects.7,9 Our study concerned workers whose age did not exceed 65 years, with no neurological deficits; age and gender were included into statistical analyses. Thus, the results in exposed group indicate that the solvents were responsible for CNS abnormalities. Abnormal OKN results were observed in 53% of solvent-exposed workers, and abnormal SP results were noted in 43% of this group. Similar data were published by Sulkowski et al12; they noticed abnormal results of OKN and SP in 25% and 13% of workers exposed to organic solvent mixture, respectively. The OKN test is less sensitive but more specific for central lesions assessment. Symmetric OKN gain impairment in subjects who revealed slight or no abnormalities of SP may suggest the early stage of degenerative lesions in the cerebellum/brainstem area.
Sulkowski et al12 showed that central abnormalities are more frequent in subjects working longer. Niklasson et al9 found the linear correlation between SP gain and duration of exposure, although in both studies, age might be a significant factor, while it was not included in the statistical analyses. In our study, we demonstrated nothing more than strong correlation between urinary excretion of methylhippuric acid and SP gain.
In caloric test, we found more frequent and significantly shorter mean response (reactivity) to caloric stimuli in solvent-exposed group as compared with nonexposed subjects. Moreover, this hyporeactivity was negatively correlated to toluene cumulative concentration. The vestibular hyporeactivity may be of peripheral or central origin. Tham et al2,3 suggested that toluene caused excitation of the VOR reflex by interacting with the central pathways in reticular formation and the cerebellum, which might result in nystagmus reduction. Unfortunately, fixation index-–the vulnerable tool to distinguish between peripheral and central abnormalities-–was not included in our protocol. The likelihood of vestibular involvement was discussed in previous studies12,22 and should also be considered in our study.
In summary, the Numerous Balance test abnormalities were found in group of subclinical workers occupationally exposed to organic solvents mixture. Our results revealed that central part of vestibular system and body movement coordination were the most involved. The shortening of caloric response may have the central origin, although it would not be reasonable to exclude the vestibular hyporeactivity.
The authors thank Professor Wieslaw Szymczak, PhD, for the contribution to statistical analyses.
This study was supported by Nofer Institute Research Project.
1. Wenngren B, Ödkvist LM. Vestibulo-oculomotor disturbances caused by occupational hazards. Acta Otolaryngol. 1988;(suppl 455):7–10.
2. Tham R, Bunnfors I, Eriksson B, Larsby B, Lindgren S, Odkvist LM. Vestibulo-ocular disturbances in rats exposed to organic solvents. Acta Pharmacol Toxicol (Copenh). 1984;54:58–63.
3. Tham R, Larsby B, Eriksson B, Niklasson M. The effect of toluene on the vestibulo- and opto-oculomotor system in rats, pretreated with GABAergic drugs. Neurotoxicol Teratol. 1990;12:307–311.
4. Pascual J. Toluene alters p75NTR expression in the rat brainstem. Ind Health. 2004;42:75–78.
5. Ödkvist LM, Larsby B., Tham R, Hyden D. Vestibulo-oculomotor disturbances caused by industrial solvents. Otolaryngol Head Neck Surg. 1983;91:537–539.
6. Ödkvist LM, Arlinger SD, Edling C, Larsby B, Bergholz LM. Audiological and vestibulo-oculomotor findings in workers exposed to solvents and jet fuel. Scand Audiol. 1987;16:75–81.
7. Ödkvist LM, Möller C, Thuomas K. Otoneurologic disturbances caused by solvent pollution. Otolaryngol Head Neck Surg. 1992;106:687–692.
8. Möller C, Ödkvist LM, Thell J, Larsby B, Hyden D, Bergholz LM. Otoneurological findings in psycho-organic syndrome caused by industrial solvent exposure. Acta Otolaryngol. 1989;107:5–12.
9. Niklasson M, Möller C, Ödkvist LM, et al.. Are deficits in the equilibrium system relevant to the clinical investigation of solvent-induced neurotoxicity? Scand J Work Environ Health 1997;23:206–213.
10. Callender TJ, Morrow L, Subramanian K, Duhon D, Ristovv M. Three-dimensional brain metabolic imaging in patients with toxic encephalopathy. Environ Res. 1993;60:295–319.
11. Hein HO, Suadicani P, Gyntelberg F. Cerebral symptoms in 3,387 men and occupational exposure to organic solvents. An epidemiological study. Ugeskr Laeger. 1991;153:493–496.
12. Sulkowski WJ, Kowalska S, Matyja W, et al.. Effects of occupational exposure to a mixture of solvents on the inner ear: a field study. Int J Occup Med Environ Health. 2002;15:247–256.
13. Antti-Poika M, Ojala M, Matikainen E, Vaheri E, Juntunen J. Occupational exposure to solvents and cerebellar, brainstem and vestibular functions. Int Arch Occup Environ Health. 1989;61:397–401.
14. Laine A, Savolainen K, Riihimaki V, Matikainen E, Salmi T, Juntunen J. Acute effects of m-xylene inhalation on body sway, reaction times and sleep in man. Int Arch Occup Environ Health. 1993;65:179–188.
15. Kuriwaka R, Mitsui T, Fujiwara S, Nishida Y, Matsumoto T. Loss of postural reflexes in long-term occupational solvent exposure. Eur Neurol. 2002;47:85–87.
16. Sliwinska-Kowalska M, Zamyslowska-Szmytke E, Szymczak W, et al.. Hearing loss among workers exposed to moderate concentrations of solvents. Scand J Work Environ Health. 2001;27:335–342.
17. Van der Stappen A, Wuyts F, Van de Heyning P. Computerized electronystagmography: normative data revisited. Acta Otolaryngol. 2000;120:724–730.
18. Janczewski G, Latkowski B. Otoneurologia. Warszawa: Bel Corp. Scientific Publications; 1998.
19. Wuyts F, Boniver R. Normative data in ENG and VNG. B-ENT. 2008;4:45–47.
20. Smith LB, Bhattacharya A, Lemasters G, et al.. Effect of chronic low-level exposure to jet fuel on postural balance of US Air Force personnel. J Occup Environ Med. 1997;39:623–632.
21. Iwata T, Mori H, Dakeishi M, Murata K. Effects of mixed organic solvents on neuromotor functions among workers in Buddhist altar manufacturing factories. J Occup Health. 2005;47:145–148.
22. Calabrese G, Martini A, Sessa G, et al.. Otoneurological study in workers exposed to styrene in the fiberglass industry. Int Arch Occup Environ Health. 1996;68:219–223.