Ann. occup. Hyg., Vol. 47, No. 6, pp. 493-502, 2003
© 2003 British Occupational Hygiene Society
Published by Oxford University Press
Occupational Toluene Exposure and Auditory Function: Results from a Follow-up Study
Institute for Occupational Physiology, University of Dortmund, Ardeystrasse 67, D-44139 Dortmund, Germany
Received 21 November 2002; in final form 20 March 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT PROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The ototoxicity of occupational exposure to toluene below 50 p.p.m. was investigated in a longitudinal study over 5 yr with four repeated examinations starting with 333 male workers from rotogravure printing plants. Past lifetime weighted average exposures (LWAE) to toluene and noise were determined from individual work histories; recent individual exposures were measured 10 times during the study (toluene, active sampling; noise, stationary measurements). The auditory thresholds were measured with pure tone audiometry. The mean LWAE exposures to toluene and noise were 45 ± 17 p.p.m. plus 82 ± 7 dB(A) for printers (high toluene intensity) and 10 ± 7 p.p.m. plus 82 ± 4 dB(A) for end-processors (low toluene intensity). The mean current exposures to toluene and noise during the study were 26 ± 20 p.p.m. plus 81 ± 4 dB(A) for printers and 3 ± 3 p.p.m. plus 82 ± 4 dB(A) for end-processors. Repeated measurement analyses (grouping factors: toluene intensity, exposure duration and noise intensity) and logistic regressions did not reveal significant effects of toluene intensity, of exposure duration and of interactions between toluene intensity and noise intensity. The stratification dependent on noise intensity itself [79 ± 3 versus 84 ± 1 dB(A)] was significantly associated with the auditory thresholds. Regarding the missing toluene effects, it was concluded that the threshold level for developing a hearing loss as a result of occupational toluene exposure might be above the actual limit of 50 p.p.m.
Keywords: biomarker; follow-up study; noise; rotogravure printing; toluene
|
PROBLEM |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
During recent decades the ototoxicity of solvents and their interaction with noise has become evident (Barregard and Axelsson, 1984). For example, occupational exposures to solvents like trichloroethylene (Szulc-Kuberska et al., 1976), styrene (Muijser et al., 1988) and mixtures of solvents (Jacobsen et al., 1993; Bergstrom and Nystrom, 1986) were reported to lead to auditory impairments. Patterns of interactions with noise have been reported and were considered in recent toxicity evaluations of solvents, e.g. of toluene (DEPA, 2001).
The ototoxicity of toluene was first described by Pryor et al. (1983a,b, 1984) in animals. The relationship between toluene concentration, exposure pattern and hearing loss was investigated in a series of experiments with rats. Two weeks of exposure to 1000 p.p.m. toluene for 14 h/day caused hearing loss. Lower concentrations (400 and 700 p.p.m.) were ineffective even after 16 weeks of exposure. Three day exposures to 1500 p.p.m. for 14 h/day or to 2000 p.p.m. for 8 h/day were ototoxic. Single exposures to 4000 p.p.m. for 4 h or to 2000 p.p.m. for 8 h were without effect. Intermittent exposure to 3000 p.p.m. for 30 min every hour for 8 h/day caused hearing loss within 2 weeks, but a similar exposure schedule for 4 h/day was ineffective even after 9 weeks. Therefore, toluene seemed to be ototoxic in rats but only at relatively high concentrations of exposure. In subsequent studies in rats and mice electrophysiological as well as behavioural and morphological methods were utilized to determine the ototoxic effects of toluene on hearing. Searching for a more sensitive animal species McWilliams et al. (2000) found transient auditory impairments in guinea pigs, starting at exposure levels of 250 p.p.m. for 8 h/day for 5 days. However, no permanent impairment of hearing was identified in guinea pigs (L. Fechter, personal communication).
The interactions of exposure to toluene and noise in rats were first analysed by Johnson et al. (1988, 1990) and Lataye and Campo (1997). Lataye and Campo found synergistic effects for simultaneous exposures to toluene and noise, whereas the experiments of Johnson et al. reported such effects for consecutive exposures to toluene and noise, but not vice versa. Later, Brandt-Lassen et al. (2000) reported that rats exposed to either 0 or 500 p.p.m. toluene (6 h/10 days) plus noise (2 h/10 days, 96 dB white noise) developed a small significant auditory threshold shift. Without additional noise, 500 and 1000 p.p.m. toluene did not produce such a threshold shift. At toluene concentrations of 1000, 1500 and 2000 p.p.m. clear and significant impairing interactions between noise and toluene were observed.
In humans, ototoxic results of toluene were found by Abbate et al. (1993) for a selected group of 40 toluene-exposed printing workers. Mean exposures of 97 p.p.m. toluene for 
13 yr were associated with alterations in acoustically evoked potentials compared with matched controls. The changes were interpreted as subclinical signs of auditory nervous system changes.
Morata et al. (1993) investigated the hearing abilities of four groups of workers (n = 190) exposed to noise, to noise and toluene and to a mixture of solvents including toluene and of an unexposed control group by pure tone audiometry and emittance audiometry. Data on individual toluene exposures varied from 75 to 365 p.p.m. for an average of at least 6 yr and noise exposures ranged from 88 to 98 dB(A) for an average of at least 8 yr. Multiple logistic regressions for bilateral high frequency hearing loss >25 dB revealed that the relative risk was highest in the toluene plus noise group, followed by the solvent group and the noise group, with each risk being significant. In another study by Morata et al. (1997) a group of 124 printing workers exposed to various levels of noise (71–93 dB) and a mixture of toluene (up to 241 p.p.m.), ethyl acetate (up to 653 p.p.m.) and ethanol (up to 753 p.p.m.) for an average of 8 yr was investigated with pure tone audiometry and emittance audiometry. Both hippuric acid in urine, a traditional biomarker for toluene, as well as age were significantly associated with high frequency hearing loss >25 dB in logistic regression models. The concentration of toluene in ambient air was not found to be significantly associated with hearing loss and no significant associations with hearing loss were noted for the other solvents or noise.
The ototoxicity of toluene above a certain distinctive level seems to be the obvious conclusion from the literature. However, the critical threshold for such effects, as well as the interaction with occupational noise exposure conditions, remains unclear. The results in rats are based on high concentrations and short exposures and cannot be used for extrapolations to workplace conditions. The results in humans with occupational exposures support the assumption of combined effects of toluene and noise. However, the data do have certain limitations. The occupational exposures were partly described as toluene plus other solvents. The individual work histories concerning long-term exposure to toluene and noise were not adequately documented apart from the current exposures during the examination time. Contrary to the animal studies, the cross-sectional design of the human studies was restricted with respect to the development of adverse effects and potential causes (e.g. toluene/noise exposure). The present study tries to overcome these restrictions. Both long-term and recently existing exposure conditions are described, based on a representative extent of data. Furthermore, the potential development of toluene- and noise-induced hearing loss was documented by repeated measurements in a follow-up design.
|
STUDY DESIGN AND SUBJECTS |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The study was conducted as a long-term study over 5 yr (1996–2001) with four repeated examinations (examination periods 1–4 at intervals of
1 yr). The study was approved by the relevant authorities (regional government, co-determination partners in the plants) and was supervised by a commission representing the employees and employers as well as the employers’ liability insurance association. The study covered physical, physiological and neuropsychological methods to reveal potential health effects of occupational exposure to toluene in modern rotogravure printing plants. An overview on the study can be obtained from the research report (Seeber et al., 2002). Employees from 14 German rotogravure printing plants were recruited to participate in the study voluntarily. The results concerning auditory effects will be reported. The design stratified the participants in a first step as ‘high exposed’ (printing division) versus ‘low exposed’ (end-processing division). The second stratification depended on job tenure in rotogravure printing, ‘short’ versus ‘long’, and finally returned four subgroups, ‘Short High’, ‘Long High’, ‘Short Low’ and ‘Long Low’. For descriptive data, see Table 1. Analysing auditory functions, a third stratification factor, i.e. intensity of current noise exposure (high versus low), was introduced with a cut-off point at the median of the noise data, 82 db(A). As a consequence of the working conditions it had to be taken into account that exposure duration meant duration of toluene as well as noise exposure of the participants.
|
The sample size went down from 333 (100%) in examination 1 to 278 (83.5%) in examination 2, 241 (72.4%) in examination 3 and 216 (64.9%) in examination 4. The 333 participants at the onset of the study represented about 4.5% of the employees in the corresponding working areas of the plants taking part in the study. For a subsample of 192 participants a complete repeated data set was available to fit the follow-up design of the study. The four subgroups of this repeated measures sample did not show significant differences from the four subgroups at the onset of the study concerning age and duration of exposure and the split of the whole sample into the four subgroups. Therefore, they were considered to adequately represent all participants in their initial data. Table 1 gives descriptive data for the total sample and the repeated measures sample, which will be denoted as ‘sample’ in the following.
The fractions of participants leaving the study were not significantly different between the subgroups (
2 = 0.49, P = 0.92). The main reasons for leaving the study were ‘problems with the date of examination’, ‘no more interest’ and ‘sick leave on the date’, which did not differ significantly between high and low exposed. However, ‘cessation of employment’ happened more frequently among low than high exposed workers (11.1 versus 6.6%). This difference was caused by economically induced structural changes in the plants during the 5 yr and was not related to health differences, especially hearing threshold, between the remaining and the leaving persons (df 1/315, F = 1.37, P = 0.24).
In addition to the main aim of performing a long-term repeated measurement study, for a subsample (n = 80) logistic regression models were performed to verify the results of Morata et al. (1997) on the relation between high frequency hearing loss >25 dB and current exposure to toluene, indicated by biomarkers.
Examination of auditory functions
The medical and psychological examinations took place in the first or in the second half of a morning or an afternoon shift, leaving at least 3 h of exposure-free time before ascertaining the auditory thresholds. During these 3 h, different examinations were carried out. Most of the medical examinations, including audiometry, were performed by one physician, except for examination 1, when a second physician performed most of the examinations. The individual hearing thresholds of every worker were ascertained from pure tone audiometry with a Siemens SD 26 Diagnostic audiometer meeting the audiometric requirements given in IEC Publication 645 (IEC, 1979). The test frequencies were 125, 250, 500, 750, 1000, 1500, 2000, 3000, 4000, 6000, 8000 and 12 000 Hz. The tests were performed with the standard noise-excluding Siemens headset and, where available in the different plants, in a sound insulated chamber. In order to include the typical frequencies for noise-induced hearing loss and potential toluene-induced deficiencies the six individual hearing thresholds for frequencies from 1.5 to 8 kHz were analysed statistically. To eliminate natural age-induced increases in the auditory thresholds the data was age adjusted following ISO 7029 (International Organization for Standardization, 1984) before the repeated measures analyses.
As a part of the physical examination sequence, otoscopy was performed to screen for exposure non-related hearing deficits (i.e. otitis or a perforated tympanic membrane) and, additionally, in examination period 4 tympanometry was used. From this screening and the results of their audiograms, 28 persons were categorized as suffering from hearing defects that had not been induced by occupational exposure. The subcategories were ‘acute hearing loss’ (n = 2), ‘explosion trauma’ (n = 12), ‘conductive hearing loss’ (n = 14) and ‘unilateral hearing loss’ (n = 26), with multiple categorization possible. However, ascertaining data via tone audiometry was possible in these subjects. The numbers did not differ significantly between the groups with high versus low toluene intensity, short versus long exposure duration or high versus low noise intensity. Thus, these subjects were not excluded from the analysis.
During examination period 2, for 80 sample participants data on high frequency hearing loss >25 dB was analysed in logistic regression models. High frequency hearing loss was diagnosed if the audiogram of a participant showed a hearing loss of at least ‘25 dB at any of the tested frequencies, if the audiogram revealed a notch in one of the frequencies between 1 and 6 kHz or the thresholds were poorest in this frequency range’ (Morata et al., 1997, p. 293). The prevalence of bilateral hearing loss was 36% in the present sample.
Toluene and noise exposures
Current toluene exposure in ambient air
The current individual toluene exposure for each participant in the study was measured twice per year. Air from the breathing zone was collected for a whole working day using active samplers consistent with TRGS (technical regulations governing hazardous substances) 402 (TRGS, 1993). Means and standard deviations of the measurements per exposure period are given in Table 2. Across the whole study, the average exposure level for toluene in the breathing zone was 25.7 ± 20.1 p.p.m. in the printing area (n = 106 persons) and 3.2 ± 3.1 p.p.m. in the end-processing area (n = 86 persons), revealing an exposure relation between the groups (ratio of means) of 8:1.
|
Lifetime weighted toluene exposure in ambient air
The calculations of the individual lifetime weighted average exposures (LWAEs; Bleecker et al., 1991) to toluene at the onset of the study were based on job exposure matrices with data from five representative German plants. The data indicated a decrease in the mean exposure for the printing area from 135 p.p.m. before 1975 to 40 p.p.m. in 1995 and for the end-processing area from 20 to 5 p.p.m. At examination 1 the work history of each participant was determined with a standardized interview. From the available information the individual cumulative exposure (CE) at the beginning of the study was calculated as the total of the products of particular exposure times and exposure levels during the working life. The LWAE was derived from the individual CE as the individual average exposure during the completed working years by dividing the CE by the summarized personal exposure time. For the 192 sample participants, the LWAE data of the groups are also given in Table 2. The relation between the groups ‘high’ versus ‘low’ (ratio of means) amounted to about 5:1 with respect to lifetime exposure.
Current noise exposure
The actual individual noise exposure for each participant in the study was measured together with his toluene exposure twice per year. The individual Leq [dB(A)] was assessed from the data from permanent stationary sound level metering at typical work stations and the individual work scheme following DIN 45645 (Deutsches Institut für Normung, 1977). The mean noise exposures did not differ significantly between the toluene exposure groups during the study. For the whole study the mean current noise exposure for the sample was 81.1 ± 3.5 dB(A) in the printing area (n = 106) and 81.6 ± 4.2 dB(A) in the end-processing area (n = 86). Means and standard deviations of the measurements per exposure period are given in Table 2. For the two groups dependent on the intensity of actual noise the mean levels of actual noise exposure were 79 ± 3 versus 84 ±1 dB(A) during the whole study, the difference being significant (P < 0.001).
Lifetime weighted noise exposure
The individual lifetime noise exposures at the onset of the study were calculated by a procedure similar to that used to calculate the lifetime exposures to toluene. Previous data from 11 printing plants was used to assess job exposure matrices for the same time periods as for toluene in order to use the same work histories to calculate the individual lifetime exposures to noise. The data indicated a decrease in the mean exposure for the printing area from 92.5 dB(A) before 1975 to 83 dB(A) in 1995 and for the end-processing area from 88 to 82.5 dB(A). The relation between the noise levels of the study groups ‘high’ versus ‘low’ (ratio of means) for their lifetime exposure was nearly 1, the differences not being significant. The LWAE values are also given in Table 2.
Biomarker data on current toluene exposure
During examination period 2, for 80 sample participants concentrations of hippuric acid and o-cresol, two biomarkers for current total toluene exposure (ACGIH, 2002; DFG, 2002), were ascertained from after-shift urine samples. The analyses of the urine samples were performed according to the standards published by the German Research Foundation (DFG, 1999). The mean ± standard deviation of the biomarker subsample were:
demographics: 39 ± 9 (range 24–56) yr for age, 13 ± 9 (range 3–38) yr for tenure, 15 ± 15 (range 1–69) p.p.m. for toluene in ambient air, 82 ± 3 (range 70–87) dB(A) for current noise;
biomarkers: 1.8 ± 1.6 (range 0.1–8.9) g/l urine for hippuric acid (BEI 1.6 g/g creatinine) and 1.0 ± 1.2 (range 0–6.0) mg/l urine for o-cresol (BEI 0.5 mg/l, German BAT 3.0 mg/l).
Statistical procedures
All statistical computations were performed with SAS and SPSS routines. The six individual hearing thresholds for the frequencies from 1.5 to 8 kHz were log transformed and analysed by repeated measures analysis of variance. The between factors in this analysis were toluene intensity (high versus low), duration of exposure (long versus short) and noise intensity (high versus low). The within factors were examination period (4), ear (2) and frequency (6).
In the second approach comprising biomarker data and information on high frequency hearing loss stepwise logistic regression models were used to estimate odds ratios analogous to Morata et al. (1997). The models were provided with the independent variables hippuric acid/o-cresol, exposure, age, tenure, actual toluene exposure in ambient air, actual noise exposure and information on severe ear infections. The variables were entered as continuous ones except for the information on ear infections (yes/no).
|
RESULTS |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Toluene intensity, exposure duration and all interactions of the stratification factors did not show significant effects on the auditory thresholds (Table 3, between subject effects). The main hypothesis of an effect of toluene exposure level on hearing was not supported. However, a significant effect of noise intensity (F = 4.5, P = 0.04) emerged. The mean values and standard errors for the hearing thresholds related to the factors toluene intensity, exposure duration and noise intensity are given in Fig. 1. There, the side result of significant differences between the participants’ left and right ears (F = 21.3, P = 0.00) is also demonstrated.
|
|
The main interest of the study, the analysis of the longitudinal design with its four repeated measures, revealed additional results. When applied to the full length (examination periods 1–4) of the study, significant effects were observed for the factor Time (examination period number) itself and for five of seven interactions of the between factors with the factor Time (Table 3, Within factor Time and interaction factor Time). This means that a general change in the auditory thresholds took place during the 5 yr of the study, indicated by the significant factor Time, and the patterns of change were significantly dissimilar for the different groups depending on toluene intensity, noise intensity, toluene intensity x exposure duration, noise intensity x exposure duration and toluene intensity x exposure duration x noise intensity.
Figure 2 provides the hearing thresholds dependent on toluene/noise intensity for every examination period illustrating the origin of the significant interactions of toluene intensity x Time and noise intensity x Time. The figure shows that the data for examination period 1 deviated from the rest of the data, especially concerning the factor toluene intensity. Generally, for every examination period the threshold differences of the groups dependent on toluene intensity were about half the size of the differences dependent on noise intensity (1.5 versus 3–4 dB with significant between factor noise intensity) and the increase in the differences between groups was greater for the differentiation concerning noise intensity.
|
The combination of toluene intensity and noise intensity was not associated with significant changes in the auditory thresholds in the time course of the auditory data. This result, together with the missing between subjects effect of this factor combination, did not fit the hypothesis of a higher impact of exposure in the high toluene plus high noise exposed group. The origin of the significant three-factor interaction results is pointed out in Fig. 3. Similarity of the data structure from examination 2 to 4 was observed and again examination period 1 deviated. In the left figure, the groups with long exposure duration showed the highest auditory thresholds. Short exposure was associated with lower threshold values. In the right figure, the groups including high noise, independent of their exposure duration, showed the highest auditory impairments. The group with low noise and short exposure duration showed the best results. This clearly indicated the evidence of higher noise exposure on the auditory thresholds whereas such clear evidence for high toluene exposure was not observed.
|
Excluding examination 1 because of its dissimilarities to the rest of the study in a re-analysis, each two- and three-factor interaction became insignificant. The only crucial factors remaining significant were noise intensity (F = 5.4, P = 0.02) and the factor time (F = 5.0, P = 0.01). So, noise intensity was the only exposure factor to show a significant impact on the hearing thresholds in analysing either three or four repeated measures.
The stepwise regressions analysing the data of a subsample (n = 80) revealed that only age significantly elevated the risk of bilateral high frequency hearing loss. The intensity of toluene exposure in ambient air, the results of biological monitoring and noise level did not contribute significantly to the explanation of high frequency hearing loss. Thus, the results of Morata et al. (1997), who found both age and hippuric acid to be significant, were not replicated at the given exposure level. The complete results are displayed in Table 4 for each model (left, hippuric acid; right, o-cresol).
|
Summarizing the results of the analyses, level or duration of toluene exposure did not show significant effects on the auditory thresholds of the study participants. The only significant exposure variable was intensity of current noise exposure for the age-adjusted data in the repeated measures analysis and the variable age for the logistic regressions.
|
DISCUSSION |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The discussion will try to integrate the present study into the existing results in the literature and to point out potential shortcomings of the study.
Classification of the exposure level of the study
The experiments in rats and guinea pigs and the studies in humans revealed different degrees of impairment due to toluene exposure alone and combined with noise dependent on the levels of exposure. In rats, ototoxic effects were shown for high toluene concentrations. Permanent damage of outer hair cells in the cochlea seemed to be dose related above a certain threshold limit. Searching for a critical daily dose for auditory impairment a level of 9800 p.p.m. h per day was described (Pryor et al., 1984). A similar critical level emerged from the study of Brandt-Lassen et al. (2000). In rats, hearing loss was revealed at a dose of 9000 and 12 000 p.p.m. h toluene alone for 10 days. Simultaneous exposures of 12 000 p.p.m. h and 92 db(A) for 20 days (Lataye and Campo, 1997) and sequential exposures to noise after toluene at 16 000 p.p.m. h for 10 days and 1000 dB(A)h for 28 days (Johnson et al., 1988, 1990) had a synergistic effect in rats. On the other hand, consecutively combined exposure to at least 6000 p.p.m. h and 96 dB white noise for 2 h over 10 days revealed similar results (Brandt-Lassen et al., 2000). In guinea pigs, McWilliams et al. (2000) found transient auditory impairments at 2000 p.p.m. h over 5 days.
The first study of Morata et al. (1993) indicated significantly elevated risks of developing hearing impairment after combined occupational exposures to toluene and noise. For the toluene plus noise group, the mean dose per work day was 88–98 dB(A) plus 1500 p.p.m. h resulting from a calculated mean of 184 p.p.m. per work shift (Morata et al., 1993, table 1, p. 246). For the present study the mean dose per work day was 81 dB(A) plus 208 p.p.m. h for the high exposed group resulting from a mean of 26 p.p.m. per work shift. This is about a seventh of the toluene exposure and less than a seventh of the physical noise energy for the high exposed compared with the Morata sample, with an increase of 3 dB meaning a doubling of physical energy. Therefore, a threshold for auditory impairment based on the combination of toluene and noise exposure is probably located at a level between the exposures of the two studies.
In the second study of Morata et al. (1997) a significantly increased risk for hearing loss >25 dB was observed at the measured concentrations of hippuric acid. This biomarker-based result was extrapolated to estimate an odds ratio >2 for a toluene exposure of 50 p.p.m. in ambient air. To compare the results of Morata et al. and the present study the scales for hippuric acid had to be harmonized. The mean value of the Morata et al. study was 2.3 g/g creatinine (BEI 1.6 g/g creatinine). The corresponding value of the present study was 1.0 g/g creatinine. Accepting hippuric acid as a marker of toluene exposure and given that the prevalence of bilateral hearing loss amounted to 49% in the Morata et al. sample in comparison to 36% in the present one, the different odds ratios (Morata et al., 1.76, CI, 1.00–2.98; this study, 1.28, CI, 0.75–2.18) are reasonable.
Finally, the present study is classified by the information on the participant’s LWAE regarding toluene and noise. For the group with high and long exposure the mean LWAE levels were 59 p.p.m. plus 86 dB(A) for 22 yr and the LWAE levels for about 68% of the participants in this group were up to values of 72 p.p.m. toluene and 88 dB(A) noise exposure for 22 yr. Thus, the information on current exposures of 25 p.p.m. plus 81 dB(A) for the group with high and long exposure was supplemented with valuable information, especially taking into account that the study did not reveal any toluene-induced hearing impairments.
Critical remarks on the study design
The study did not replicate the findings of the initially cited studies concerning ototoxic effects of toluene exposure below the level of 50 p.p.m. toluene, even in conjunction with noise. Therefore, potential shortcomings have to be discussed.
The significant within factor results in the ANOVA analysis of the repeated dataset of examination periods 1–4 are obviously related to dissimilarities between the data of examination period 1 and the remaining periods. Therefore, these results were not considered to be exposure related and were left out of the final conclusions.
A possible disadvantage of the study was not including non-exposed participants, either to noise or to toluene. However, the current mean exposure to noise was almost equal for the two toluene exposure groups. Therefore, a potential difference in the auditory thresholds between the high and low toluene-exposed groups would have been ascribed to toluene alone. Additionally, the actual mean toluene exposure for the low exposed group was around the olfactory threshold of 3 p.p.m. (Amoore and Hautala, 1983), where no effects on physical health are to be expected. Thus, from the viewpoint of matching for noise, the type of sample seemed to be a reasonable approach.
The problem of a ‘healthy worker effect’ may exist, because workers who left the plants before the onset of the study were not investigated. However, during the study period of 5 yr the participant’s reasons for leaving their firm were equally distributed between the subgroups except for ‘cessation of employment’, as previously explained. Furthermore, those participants (n = 141) who left the study sample during the 5 yr were checked concerning their auditory thresholds against the remaining participants (n = 192). This analysis did not reveal a significant difference (df 1/315, F = 1.37, P = 0.24).
The only significant exposure-related difference between the two groups with high and low current noise intensity that required additional information, concerned the use of personal noise protection. Up to the onset of the study, the percentages for ‘never wearing ear protection’ went down from 76% before 1975 to 30% in 1996 and for ‘always wearing ear protection’ went up from 4 to 28%. At the end of the study these percentages were 23% (never) and 22% (always). Corresponding information for the later Morata et al. study concerning recent habits regarding noise protection were ‘only 11% of the workers exposed to noise above 85 dB(A) reporting using hearing protectors’ (Morata et al., 1997, p. 293). Under these conditions and a mean tenure of 7 yr (range 1–25 yr) Morata et al. did not detect a noise effect despite the information that ‘60% of the study population was exposed to noise doses considered to be high enough to cause hearing loss’ (Morata et al., 1997, p. 295). Obviously, the longer tenure in our study (13 yr) was the reason for the group differences concerning noise.
|
CONCLUSION |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The initially cited statements on hearing loss due to occupational toluene exposure can now be more closely specified regarding the NOAEL (no adverse effect level). The combination of a current mean exposure level of
25 p.p.m., an average exposure duration of 22 yr and a mean current noise level of 81 db(A) was not associated with hearing loss. Additionally, taking into account the information on LWAE as described in this study [59 p.p.m. plus 86 dB(A)], the threshold level for developing a hearing loss due to toluene exposure might be above the actual limit level of 50 p.p.m. Acknowledgements—A team from the Berufsgenossenschaft für Druck und Papierverarbeitung (Printing and Paper conversion) and from the Central Institute for Research and Testing of the Berufsgenossenschaften provided the raw data for the exposure analyses. The authors appreciate especially the excellent support by the cooperation partners A. Glöckle, Dr E. Cuno and Dr J. Seibel. Additionally, the authors thankfully appreciate the cooperation of the study participants in following all requirements of the study. We thank Professor L. Fechter, University of Oklahoma, for his comments on and corrections to the manuscript. The study was supported financially by different institutions of the German Berufsgenossenschaften (Employer’s Liability Insurance Association).
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: +49-231-1084-415; fax: +49-231-1084-308; e-mail: schaeper{at}ifado.de
|
REFERENCES |
|---|
|
TOPABSTRACTPROBLEMSTUDY DESIGN AND SUBJECTSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
ACHGIH. (2002) TLVs and BEIs: threshold limit values for chemical substances and physical agents and biological exposure indices. Cincinnati, OH: American Conference of Governmental Industrial Hygienists.
Abbate C, Giorgianni C, Munao F, Brecciaroli R. (1993) Neurotoxicity induced by exposure to toluene. An electrophysiologic study. Int Arch Occup Environ Health; 64: 389–92.[CrossRef][Web of Science][Medline]
Amoore JE, Hautala E. (1983) Odor as an aid to chemical safety: odour threshold compared with threshold limit values and volatilities for 214 industrial chemicals in water and air solution. J Appl Toxicol; 3: 272–89.[Medline]
Barregard L, Axelsson A. (1984) Is there an ototraumatic interaction between noise and solvents? Scand Audiol; 13: 151–5.[Web of Science][Medline]
Bergstrom B, Nystrom B. (1986) Development of hearing loss during long-term exposure to occupational noise. A 20-year follow-up study. Scand Audiol; 15: 227–34.[Web of Science][Medline]
Bleecker ML, Bolla KI, Agnew J, Schwartz, BS, Ford DP. (1991) Dose-related sublinical neurobehavioral effects of chronic exposure to low levels of organic solvents. Am J Ind Med; 19: 715–28.[Web of Science][Medline]
Brandt-Lassen R, Lund SP, Jepsen GB. (2000) Rats exposed to toluene and noise may develop loss of auditory sensitivity due to synergistic interaction. Noise Health; 3(9): 33–44.[Medline]
DEPA. (2001) Risk assessment toluene, final report—July 2001. Copenhagen: Danish Environmental Protection Agency.
Deutsches Institut für Normung. (1977) Einheitliche Ermittlung des Beurteilungspegels für Geräuschimmissionen, DIN 45645. Berlin: Deutsches Institut für Normung.
DFG. (1999) Analyses of hazardous substances in biological materials. Deutsche Forschgsgemeinschaft. Weinheim: Wiley-VCH.
DFG. (2002) MAK- und BAT-Werte-Liste 2002. Deutsche Forschgsgemeinschaft. Weinheim: Wiley-VCH.
IEC. (1979) IEC 645, Audiometers. Geneva: Central Office of the International Electrotechnical Commission.
International Organization for Standardization. (1984) Acoustics—threshold of hearing by air conduction as a function of age and sex for otologically normal persons, ISO 7029. Geneva: International Organization for Standardization.
Jacobsen P, Hein HO, Suadicani P, Parving A, Gyntelberg F. (1993) Mixed solvent exposure and hearing impairment: an epidemiological study of 3284 men. The Copenhagen male study. Occup Med; 43: 180–4.
Johnson AC, Juntunen L, Nylen P, Borg E, Hoglund G. (1988) Effect of interaction between noise and toluene on auditory function in the rat. Acta Otolaryngol; 105: 56–63.[Medline]
Johnson AC, Nylen P, Borg E, Hoglund G. (1990) Sequence of exposure to noise and toluene can determine loss of auditory sensitivity in the rat. Acta Otolaryngol; 109: 34–40.[Medline]
Lataye R, Campo P. (1997) Combined effects of a simultaneous exposure to noise and toluene on hearing function. Neurotoxicol Teratol; 19: 373–82.[CrossRef][Web of Science][Medline]
McWilliams ML, Chen GD, Fechter LD. (2000) Low level toluene disrupts auditory function in guinea pigs. Toxicol Appl Pharmacol; 167: 18–29.[CrossRef][Web of Science][Medline]
Morata TC, Dunn DE, Kretschmer LW, Lemasters GK, Keith RW. (1993) Effects of occupational exposure to organic solvents and noise on hearing. Scand J Work Environ Health; 19: 245–54.[Web of Science][Medline]
Morata TC, Fiorini AC, Fischer FM, Colacioppo S, Wallingford KM, Krieg EF, Dunn DE, Gozzoli L, Padrao MA, Cesar CL. (1997) Toluene-induced hearing loss among rotogravure printing workers. Scand J Work Environ Health; 23: 289–98.[Web of Science][Medline]
Muijser H, Hoogendijk EM, Hooisma J. (1988) The effects of occupational exposure to styrene on high-frequency hearing thresholds. Toxicology; 49: 331–40.[CrossRef][Web of Science][Medline]
Pryor GT, Dickinson J, Howd RA, Rebert CS. (1983a) Neurobehavioral effects of subchronic exposure of weanling rats to toluene or hexane. Neurobehav Toxicol Teratol; 5: 47–52.[Web of Science][Medline]
Pryor GT, Dickinson J, Howd RA, Rebert CS. (1983b) Transient cognitive deficits and high-frequency hearing loss in weanling rats exposed to toluene. Neurobehav Toxicol Teratol; 5: 53–7.[Web of Science][Medline]
Pryor GT, Rebert CS, Dickinson J, Feeney EM. (1984) Factors affecting toluene induced ototoxicity in rats. Neurobehav Toxicol Teratol; 6: 223–38.[Web of Science][Medline]
Seeber A, Blaszkevicz M, Demes P, Kiesswetter E, Schäfer M, Sietmann C, Thriel Chv, Zupanic M (2002) Abschlussbericht zum Forschungsprojekt ‘Toluol in Tiefdruckereien’. Sankt Augustin: HVBG.
Szulc-Kuberska J, Tronczynska J, Latkowski B. (1976) Otoneurological investigations of chronic trichloroethylene poisoning. Min Otorinolaryngol; 26: 108–12.
TRGS. (1993) TRGS 402. Bundesministerium für Arbeit und Sozialordnung, BArbBL, 9/1993. Technische Regeln für Gefahrstoffe.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Sliwinska-Kowalska Organic solvent exposure and hearing loss Occup. Environ. Med., April 1, 2008; 65(4): 222 - 223. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||