NoiseChem 1
June 2004
Quality of Life and Management of Living Resources
Noise and Industrial Chemicals:
Interaction Effects on Hearing and
Balance
NoiseChem
Key Action 4: Environment and Health
2001-2004
FINAL REPORT
„ Prof. Deepak Prasher (UK): Co-ordinator
„ Dr Pierre Campo (France)
„ Prof. Laurence Fechter (USA)
„ Dr Ann-Christin Johnson (Sweden)
„ Dr Soren Peter Lund (Denmark)
„ Dr Thais Morata (USA)
„ Prof. Krystyna Pawlas (Poland)
„ Prof. Mariola Sliwinska-Kowalska (Poland)
„ Prof. Jukka Starck (Finland)
„ Prof. Wieslaw Sulkowski (Poland)
June 2004
NoiseChem 2
NoiseChem Background
The effects of combinations of environmental factors on workers’ health requires much
research attention as this reflects more closely the work conditions and little is known about
how individual toxic agents in mixed exposures interact to increase or modify the likelihood
of adverse health effects. Most work environments consist of a myriad of physical and
chemical agents that are potentially hazardous to health. The results of studies of isolated
work place hazards are often used to develop occupational safety criteria that may not be
adequate for protecting workers in environments where simultaneous or sequential exposures
to a variety of agents occur.
Around 30 million people in Europe work in noise environments hazardous to hearing, and a
further 10 million work with industrial chemicals considered to be oto-toxic such as solvents,
heavy metals, and asphyxiants. A considerable proportion of these people work with
industrial chemicals in a noisy environment thereby enhancing the risks. The risk to hearing
from noise exposure is well established but that due to exposure from industrial chemicals is
limited although laboratory studies in animals and some occupational exposure studies
suggest that simultaneous exposure to noise and chemicals produces a hearing loss, which is
significantly greater than the sum of either acting alone. In other words there are synergistic
effects observed with a combined exposure to noise and chemicals. This may mean that
individually the noise and chemicals may be within the exposure limits but in combination
may pose a greater risk. If this synergism is verified in Humans major changes will be
required in the limits that are set for occupational hazards in order to prevent occupational
hearing loss. In addition to the synergistic effects, the chemicals also affect the balance
system and the auditory central nervous system function in a way that noise appears not to do
so. It is important therefore to examine the interaction effects of chemicals and noise on both
hearing and balance systems of workers exposed to noise and chemicals together and each
independently.
The Table on the next page summarises the applications, known general health effects, audiovestibular findings and current exposure limits in some countries for the chemicals studied by
the NoiseChem group.
The need for research in this area is further heightened by the fact that there are no guidelines
or standards for combined exposures of workers to chemicals and noise on hearing and
balance functions. The aim of this group was to examine study designs, hearing assessment
alternatives and strategies for the analysis of combined effects and on the basis of the agreed
protocols conduct epidemiological studies across East/west Europe.
This study had two research groups; those working with animals to determine the mechanisms
of oto-toxic damage due to noise and chemical interactions through laboratory investigations
and those examining the effects on human audio-vestibular systems using systematic
standardised procedures through epidemiological investigations in factories in Sweden,
Finland, Poland and the UK.
NoiseChem 3
June 2004
Chemicals
studied
Toluene
Xylene
Applications
Biomonitoring
Health Effects
Glues,
Paints:
Many
industrial Processes
Hippuric Acid in Urine
Cognitive
neurological
dysfunction,
Hearing loss
and
dizziness,
Audio-vestibular
Findings
in
Animals / Humans
Altered
auditory
brainstem
responses, relative risk to hearing x5
for toluene and x11 for toluene and
noise in workers
Limit values ppm (TWA)
ACGIH 100;NIOSH100;OSHA 100
Netherlands100;UK 100;France
100
Belgium 100;Japan 100;Finland
100
Germany 100;Denmark 50;Sweden 50
Poland
ACGIH 20; NIOSH50; OSHA 100
UK 100; France50; Belgium 50
Japan 50; Netherlands 25; Denmark25
Sweden 25; Finland 20; Germany 20
Poland 25
Paints,
degreasers,
solvents for resins, gum
and rubber
Medical and Industrial
applications
Plastics, Latex paints and
coatings,
polyesters,
synthetic rubber. Wide use
in Packaging industry
Methylhippuric acid in
urine
Decreased
peripheral
nerve function, CNS
symptoms
Reduced auditory sensitivity
Mandelic acid and
PGA in urine or styrene
by gas chromotography
Changes in cerebral
activities,
dizziness,
hearing loss, hepatotoxic
Cortical
responses
affected;
vestibular
disturbances;
slight
hearing changes reported; % outside
upper limit of hearing were 12% for
noise and 33% for noise and styrene
Carbon
disulfide
Pesticide,
Fumigant,
Manufacture of viscoserayon, vulcanisation of
rubber, oils and waxes
2-thiothiazolidine-4carboxylic
acid
(TTCA) in urine
Delayed ABR, High frequency
hearing loss, synergistic action with
noise
Solvent
Mixtures
Common
in
many
industrial applications
Test
for
main
component in urine
Extensive neurological
deficits, sensory and
motor neuropathy ;
imbalance and hearing
loss
Neurotoxicity
Solvent
and Noise
Industrial plants
noisy machinery
solvent use
Test for solvent in
urine
and
noise
exposure
Styrene
with
and
Hearing loss and specific
solvent effects
Impaired
Interrupted
speech
discrimination, cortical response
abnormalities
Synergistic effect on hearing in
shipyard workers, paper mills,
printers, rayon manufacturers
ACGIH 50;NIOSH 100;OSHA
Netherlands100;UK100;France
Belgium 100;Japan100;Finland
Germany100;Denmark50;Sweden
Poland 25
200
100
100
50
June 2004
NoiseChem 4
Previous Work
The ototoxic effects of industrial solvents on workers’ hearing and balance have only been reported for a
relatively short time (Odkvist et al 1980, 1983, Bergstrom and Nystrom 1986, Moller et al 1990), in their long
history of industrial use. Numerous animal experiments and field studies have highlighted the ototoxic effects,
yet whilst risk to hearing of exposure to high levels of noise is well understood, the risk to hearing and balance
of exposure to industrial chemicals remains less well defined.
Solvents used in industry that are known to affect hearing and balance are toluene, styrene, carbon disulfide,
trichloroethylene, and xylene. The neurotoxic effects of such chemicals on exposed workers can be acute, in the
form of narcosis, central nervous system (CNS) depression, respiratory arrest, unconsciousness and even death
(NIOSH 1987). The evidence for chronic effects points to peripheral neuropathy and mild toxic encephalopathy
in solvent-exposed workers (NIOSH 1987).
Epidemiological studies have shown that impaired memory, emotional instability, dizziness, nausea, and
impaired reaction time are some of the side effects of solvent exposure.
In this review consideration will be given to the specific chemicals listed in the Table on page 3.
Toluene
Toluene is one of the most extensively researched of the industrial solvents. Toluene is an aromatic hydrocarbon
(C6H5CH3). It is a non-corrosive, volatile liquid with an aromatic odour. Worldwide production is estimated to
be 10 million tonnes and stems from two main sources, the catalytic conversion of petroleum and by-product of
the coke industry (WHO 2000).
Toluene is most frequently used for the manufacture of paints, other chemicals, thinners, adhesives, rubber, in
rotogravure printing and leather tanning. Toxicity from Toluene can occur from accidental or deliberate
inhalation of fumes, ingestion or absorption through the skin.
Toluene affects the central nervous system primarily. Acute exposure to high doses can provoke euphoria and
excitability, which is followed by a depressant response. Long-term exposure can lead to impaired memory and
concentration levels, emotional disturbance and impaired reaction times. Balance disturbances and damage to the
cerebellum have also been reported among toluene abusers (Fornazzari et al 1983).
Occupational Exposure Limits (OELs) for Toluene
Occupational Exposure Limits (OELs) for Toluene vary between 35 parts per million (ppm) in Denmark to
100ppm in the UK and the USA. The National Institute for Occupational Safety and Health (NIOSH)
recommended exposure limit (REL) is 100ppm as an 8hour time-weighted average (TWA) with a 10minute
ceiling of 200ppm. The former Occupational Safety and Health Administration (OSHA) standard for toluene
was 200ppm as an 8 hour TWA limit but in 1989 proposed the acceptable exposure level to be 100ppm with
150ppm as a short term exposure limit (STEL). This limit was expected to prevent adverse effects on the
peripheral and central nervous system and the reproductive system. However, evidence suggesting that adverse
effects occur at or below these exposure concentrations. As a consequence NIOSH is now reviewing these
recommendations (NIOSH 2003). The American Conference of Governmental Industrial Hygienists (ACGIH)
has a TWA of 50ppm. Toluene levels of 2000ppm are considered dangerous to life and health.
Hippuric acid, a metabolite of toluene is a biological marker for toluene and can be measured in urine.
4
June 2004
NoiseChem 5
Animal experiments – The effects of toluene on the auditory system
The ototoxic effects of toluene were first noted when weanling male Fischer rats exposed to 1400 parts per
million (ppm) or 1200ppm toluene, 14hrs/day, 7 days/week for 5 weeks (Pryor et al 1983) demonstrated a
slightly impaired hearing loss at 8kHz which was markedly impaired at 12kHz and above. No hearing loss was
noted at 4kHz. Although Pryor et al (1983) described these findings as a high-frequency hearing loss no testing
was done beyond 20kHz. In later studies (Campo et al 1996) this frequency area for the rat is described as the
mid-frequency range.
The above findings prompted further research regarding the hazards of exposure and abuse of toluene.
Electrophysiological testing of these rats 2.5months following exposure (Rebert et al 1983) revealed auditory
brainstem responses (ABR) consistent with a sensori-neural hearing loss. These were deemed the first results
indicating ototoxic effects of toluene on experimental animals.
Electrophysiological and behavioural audiometry (Pryor et al 1984a) on weanling or young adult male Fischer
rats exposed to 1200ppm toluene, 14hrs/day, 7 days/week for 5 weeks confirmed what Pryor et al (1984a)
describe as a high frequency hearing deficit, with a peak at 16kHz, which was more pronounced in the younger
rats. Missing and/or damaged hair cells were also found in the basal turn of the cochlea of the younger rats
during preliminary morphological examination (the only part of the cochlea investigated, in a non-quantified
study). Subsequent experiments showed that high exposure levels of 1000ppm for 14 hours/day for 2 weeks
were ototoxic but lower levels of 400 and 700ppm had no effect after a period of 16 weeks (Pryor et al 1984b).
The morphological effects on the cochlea of these levels of exposure, however, were not investigated. Pryor and
Howard (1986) pointed out that noise emanating from the inhalation system in previous experiments was not the
major factor in toluene-induced hearing loss. Using a non-inhalation route of exposure also resulted in hearing
losses in the frequency range 8-20kHz in male Fischer rats injected with 1.5 or 1.7 g/kg of toluene for a period of
a week. Possible interactions between noise and toluene exposure require investigating.
Sullivan et al (1988) attempted to relate histological data to electrophysiological data. In this experiment
Sprague-Dawley rats were gavaged with daily doses of 0.5ml and 1.0ml toluene per kg body weight for 21 days
consecutively. ABR thresholds were elevated for toluene-treated rats typically in the frequency regions 2-8kHz
described by Sullivan et al (1988) as midfrequency regions. Outer hair cell (OHC) loss occurred in the middle
and basal turn of the cochlea. The greatest loss was seen in the third row and became progressively less moving
to the second and first rows of OHCs.
The ototoxic effects of toluene noted by Sullivan et al (1988) are in contrast to those seen with other ototoxins
such as aminoglycosides (broad spectrum antibiotics) where the first row of OHCs are the primary target with
damage progressing to the second and third row of OHCs. The basal turn of the cochlea is preferentially affected
by aminoglycosides with damage spreading apically with increasing dosage in a linear fashion. The suggested
auditory effects of toluene exposure (Pryor et al 1983) has been correlated by further animal experiments which
have taken a more detailed approach.
Studies using male Sprague-Dawley rats exposed to toluene by inhalation, 1400ppm, 16hrs/day for 8days
(Johnson and Canlon 1994a, 1994b) support the interpretation that OHCs are progressively affected by toluene
exposure in the rat. A tendency towards lowered distortion product otoacoustic emission (DPOAE) amplitudes
(a sensitive means of assessing the integrity of the auditory periphery) and elevated ABRs were noted 3 days
after the start of exposure. Statistically significant lowered DPOAE amplitudes and elevated ABR thresholds
were noted after 5 days of exposure. 4 days post exposure DPOAEs were very much reduced with ABR
thresholds elevated by around 40dB across the 1.6-20kHz frequency range with maximal hearing loss between
6.3- 12.5kHz. The DPOAE results point to OHCs being mainly affected by toluene exposure, reflecting cochlear
damage, which increases with time of exposure in the rat.
Minimal damage to the third row of OHCs in the middle turn of the cochlea was seen after 5 days of exposure.
Ototoxic effects at similar levels of exposure by Pryor et al (1984b) were noted after 3 days exposure but the rats
used were younger and possibly more sensitive to the effects.
With more prolonged exposure and post exposure, damage had spread to all three rows of OHCs in the upper,
middle and part of the basal turn of the cochlea, with some inner hair cell (IHC) loss in the 4kHz area in (n = 2)
animals. The pattern of damage to OHCs is similar to that reported by Sullivan et al (1988), but the reports
disagree on toluene-induced lesions in the cochlea, (Sullivan low-mid frequency, Johnson and Canlon mid-high
5
NoiseChem 6
June 2004
frequency area of the organ of Corti). These differences may be attributed to a number of different causes such
as concentration and duration of exposure, different species of rat, and gavage in the case of Sullivan as opposed
to inhalation in the case of Johnson and Canlon. Both studies suggest that the ototoxic affects of toluene act in a
non-linear fashion.
The studies by Johnson and Canlon (1994a,b) suggest that damage progresses after the end of toluene exposure,
similar to the effects of both noise exposure and aminoglycoside exposure. The concentration of toluene
(1400ppm) used for these experiments are much higher than levels permitted in the working environment in
Europe or the USA, but were chosen because it was known to produce a considerable loss of auditory sensitivity
in the rat (Pryor et al 1983, 1984b) without causing any apparent affects on the general condition of the rat.
In the previous studies noted auditory testing in rats has been mainly within the 2-20kHz range. Whilst hearing
losses have been described as high frequency in the 8-20kHz range (Pryor et al 1983) and mid-frequency range
for 2-8kHz (Sullivan et al 1988) the fact that the rat has a hearing range of 0.25-80kHz has led to a questioning
of these descriptions with hearing losses previously named high frequency, now being referred to as midfrequency (Yano et al 1992, Johnson 1993, Crofton and Zhao 1993).
A study comparing the effects of a number of different solvents suggests evidence of a solvent-induced selective
mid-frequency (8-24kHz) hearing loss in rats exposed to concentrations in the 1000-4000ppm range (Crofton et
al 1994). This would suggest a distinctive mechanism of action which would take account of the non-linear
nature of cochlear damage caused by solvent exposure in rats. However, the results from the study by Crofton et
al (1994) were obtained using behavioural audiometry and no morphological examinations of the cochlea were
noted.
As a result of their findings Crofton et al (1994) suggest that the monitoring of high frequency hearing thresholds
in humans may not be the most effective and/or sensitive means of measuring auditory dysfunction resulting
from exposure to industrial solvents.
Differences in findings regarding frequency location of toluene-induced hearing loss stimulated Campo et al
(1997) to clarify the results of previous studies. Groups of adult male pigmented rats were exposed to
concentrations of toluene of 1000ppm, 1250ppm, 1500ppm, 1750ppm and 2000ppm respectively for 6hrs/day,
5days/week over a period of 4 weeks. Electrophysiological data (inferior colliculus potentials-ICPs) showed that
toluene concentrations of 1500-2000ppm produced significant auditory threshold shifts. (see Table I)
Table I: The effect of exposure levels of toluene on amplitude and frequency of auditory threshold shift. (Campo
et al 1997)
Toluene at
2000ppm
Toluene at
1750ppm
Toluene at
1500ppm
23dB
14dB
4dB
16kHz
16kHz
20kHz
8-12kHz
__
__
Peak amplitude of auditory threshold shift
Frequency of peak amplitude of auditory
threshold shift
Narrow plateau of auditory threshold shift
No significant threshold shift was noted at 32kHz or below 6kHz whatever the dose. These results are indicative
of no high frequency hearing loss in the rat, with the low frequency regions also being spared.
Histological data showed that toluene has a toxic effect on the cochlea. Generally, as noted by Sullivan et al
(1988), the most significant hair cell loss occurred in the third row of OHCs, with the second row less damaged
and the first row least damaged. IHCs were relatively well preserved. Interestingly toluene exposure resulted in
two peaks of damage, one at approximately 4kHz (the mid-apical turn of the rat cochlea) the second in the 1622kHz range (the middle turn). More damage was noted in the 4kHz region than at 16-20kHz region for all
toluene concentrations, but only a weak threshold shift was noted in this area. There was no obvious difference
in OHC loss between cochleograms for 1750ppm and 2000ppm.
6
June 2004
NoiseChem 7
The inconsistencies between the functional and structural results at 4kHz can have a number of explanations
including anatomical structure of the rat cochlea, sensitivity of the tests at different frequencies and the plasticity
of the response properties of the inferior colliculus (see Campo et al 1997, Lataye et al 1999). It is clear however
in this study that evoked potentials from the central auditory pathway do not mirror structural changes within the
cochlea, therefore audiometric thresholds following toluene-induced hearing loss need to be studied with caution
in terms of frequency localization.
Findings by Lataye et al (1999) measuring acute compound action potentials (CAP) on adult Long-Evans rats
exposed to 1750ppm toluene 6hrs/day, 5days/week for 4 weeks not only showed a significant auditory deficit of
20.8dB at 16kHz but also approximately 12dB at 3,4 and 5kHz, correlated by histological findings and with the
histological findings of a previous study by Campo et al (1997). Again, anatomical differences within the rat
cochlea and inferior colliculus may explain the different results using ABR (Campo et al 1997) and CAP (Lataye
et al 1999). The greatest number of afferent nerves (3000) is positioned halfway between the base and apex of
the rat cochlea, with only 800 at the apical turn. The number of neural tissues exhibiting evoked neural activity is
3-4 times higher in the inferior colliculus for high frequencies than for low frequencies, thus CAP may be a more
sensitive means of testing the low frequency regions.
The results of the study by Lataye et al (1999) suggest that toluene-induced hearing loss can occur over a
broader range of frequencies than previously reported.
In an attempt to determine the route of solvent intoxication, namely toluene and styrene, in the rat cochlea,
Campo et al (1999) found traces of toluene and styrene in the blood, brain, auditory nerve and cochlea of
exposed male Long-Evans rats following 10hrs of exposure over 2 days to 1750ppm toluene or styrene. No
traces of the solvents were found in the cerebral spinal fluids (CSF) or inner ear liquids (IEL). These results
along with the well documented pattern of damage to outer hair cells led Campo et al (1999) to propose that the
most likely preferential route for intoxication is via blood from the stria vascularis or spiral prominence diffusing
through the outer sulcus, thus reaching Hensen’s cells which are in close connection with Deiters cells located
under the OHCs.
The mechanism of action for toluene is less known but physical or chemical reactions in the hair cells are
suspected. In vitro studies (Lui and Fetcher 1997) indicate that toluene exposure leads to an increase in
intracellular calcium levels, thereby disrupting the motility of the hair cells, which in turn affects sensitivity to
sound.
The ototoxic effects of toluene and other solvents have been found to be species-dependant (Fechter 1993,
McWilliams et al 2000, Davis et al 2002, Cappaert et al 2002, Lataye et al 2003). Guinea pigs do not appear to
be affected by styrene or ethyl benzene inhalation, nor does there appear to be a synergistic affect between noise
and styrene on the guinea pigs auditory function. Chinchillas are markedly less sensitive to the effects of solvent
exposure than rats and mice. Guinea pigs have been noted to exhibit a temporary sensitivity to toluene
(McWilliams et al 2000) even at low levels of exposure 500ppm (threshold limit values in some countries is
100ppm), but these results were not confirmed in a later study (Lataye et al 2003). These variations in
susceptibility could be due to differences in levels in uptake of solvents between species, differences in solvent
metabolism between species or morphological differences in OHC structures.
Animal experiments – The effects of toluene on the vestibular system
The effect of toluene and other industrial solvents on the vestibular system in experimental animals is less well
documented than the effects on the auditory system, but studies point to neurotoxic effects rather than ototoxic
effects, that is the end organs of balance are not affected.
Studies have pointed to similarities in action of solvents producing different results with the conclusion that
solvents influence the vestibulo-occulomotor system, but by different mechanisms (Odkvist et al 1979,
Niklasson et al 1993). It was suggested by Odkvist et al (1979) that positional nystagmus elicited by solvents is
caused in vestibular subcortical pathways either by facilitation or by blocking inhibition i.e., acting on the
cerebellum. Tham et al (1984) also suggested that solvent exposure results in excitation or depression of the
vestibulo-ocular reflex (VOR) by interacting with the central pathways in the reticular formation and the
cerebellum, the effect being related to the blood levels of the solvents. Toluene was noted to have an excitatory
effect.
7
June 2004
NoiseChem 8
Short-term exposure to toluene (1500ppm 50minutes) resulted in exposed rats exhibiting longer nystagmus
reaction time to rotary acceleration with increased gain. The mean slow phase velocity was lower for exposed
rats at higher stimulation velocities. Optokinetic nystagmus after- nystagmus (OKNAN) and optokinetic
nystagmus after-after nystagmus (OKAAN) were significantly prolonged, likely due to loss of normal cerebellar
activity. (Larsby et al 1986). Decreased gain in optokinetic nystagmus is also consistent with cerebellar damage.
Results of a study by Nylen et al (1991) looking at long-term exposure to toluene (1000ppm, 21hrs/day, 6or 11
weeks) in rats tested one month after 6 or 11 weeks exposure indicated that long-term exposure to toluene causes
a long-lasting, possibly permanent lesion within the vestibulo-cerebellum, with no effects on the peripheral
vestibular or visual function.
Animal experiments – Interaction between noise and toluene
One of the more critical aspects of industrial exposure to toluene concerns interaction with noise. Barregard and
Axelsson (1984) suggested a possible ototraumatic interaction between noise and solvents when a small case
study (4 shipyard spray painters) exposed to both noise and solvents presented with more pronounced sensorineural hearing losses (SNHL) than would be expected from noise exposure alone. The study was expanded to
look at an additional 30 shipyard workers but no obvious cluster of inexplicable SNHL could be found.
In a follow-up of a 20 year longitudinal study of employees at a timber processing firm Bergstrom and Nystrom
(1986) found what they described as a remarkably large proportion of workers in the chemical division with a
hearing loss: 5-8% of workers with a history of exposure to noise levels of 95-100dBA compared with 23% of
workers in the chemical division exposed to noise levels of 80-90dBA.
The sequence of exposure to noise and toluene can have an effect on the degree of auditory dysfunction in the rat
(Johnson et al 1988, 1990). ABR responses in male Sprague-Dawley rats exposed to noise (100Leq, 10 hrs/day,
7or 5days/week, 4weeks) and toluene (100ppm, 16hrs/day, 7 or 5 days/week, 2weeks) were more elevated when
toluene exposure was followed by noise exposure than in the reverse sequence. (see fig.1 Johnson et al 1990).
Toluene followed by noise resulted in a synergistic reaction between the two agents i.e. threshold shifts exceeded
the summated loss caused by noise and toluene alone, particularly at 3.15 and 6.3kHz, even though the exposure
duration was shorter (5 days). The results from the reverse sequence showed an additive reaction between the
two agents. In both cases exposure to both agents resulted in a more severe loss of auditory sensitivity than
exposure to each agent alone.
An experiment on male Long Evans rats exposed to toluene (2000ppm, 6hrs/day, 5days/week, 4 weeks) and
noise (92dBSPL) by Lataye and Campo (1997) set out to look at the effects of simultaneous exposure to noise
and toluene. The results indicated permanent threshold shifts greater than the summated losses of noise and
toluene alone across the entire frequency range (2-32kHz) in comparison to 3.15 and 6.3kHz noted by Johnson
et al (1988).
8
NoiseChem 9
June 2004
It is difficult to compare the results of the experiments due to different testing parameters. However, both clearly
indicate a synergistic effect of noise and toluene on the auditory function in rats. The occupational impact of this
will be discussed later in the paper.
Histological data (Lataye and Campo 1997) indicates that the pattern of damage from simultaneous noise and
toluene exposure is similar to that of toluene alone but more pronounced i.e. OHC3>OHC2>OHC1. (see table
II, III ) This is a different pattern of that seen in NIHL where OHC1>IHC>OHC2>OHC3. This would suggest
that the chemical process induced by toluene exposure is compounded by noise exposure and the damage caused
by simultaneous exposure to both agents does not involve a new ototraumatic mechanism.
Styrene
Styrene is another industrial solvent that has been extensively researched (see Morata and Campo 2002 for a
review). Styrene is an aromatic hydrocarbon with a chemical structure of C8 H8 or C6H5CH=CH2. It is a
colourless to yellow, volatile, viscous liquid widely used in industry in the production of plastics, paints, resins,
synthetic rubber and insulation. It has a characteristic pungent odour.
Styrene can be absorbed in the body either through inhalation or absorption through the skin. Short term
exposure to styrene can cause irritation to the eyes, skin and respiratory tract and possible lowering of
consciousness. Long term or repeated exposure can lead to skin sensitisation, asthma and effects on the central
nervous system, including weakness, unsteadiness, slowing of reaction time and vestibulomotor dysfunction at
exposure limits of around 50ppm. Styrene is also known to have ototoxic effects (WHO 2000).
18-20 kHz region
4-5kHz region
Toluene 2000ppm
OHC3 OHC2 OHC1
73%
42%
25%
87%
59%
30%
Table II depicts percentages of outer hair cell loss for toluene-exposed animals (Lataye and Campo 1997)
9
June 2004
NoiseChem 10
Toluene 2000ppm + noise 92dbSPL
OHC3
OHC2
OHC1
18-24 kHz region 98%
86%
60%
4-5kHz region
89%
74%
41%
Table III depicts hair cell loss for simultaneous noise and toluene-exposed animals (Lataye and Campo 1997)
Occupational Exposure Limits for Styrene
Occupational Exposure Limits for Styrene vary between 20ppm in Sweden and 100ppm in the UK and USA.
NIOSH has a recommended exposure limit of 50ppm as an 8hour time-weighted average (TWA) with a short
term exposure limit (STEL) of 100ppm. OSHA has a permissible exposure limit (PEL) of 100ppm as an 8 hour
TWA limit with a ceiling of 200ppm. And 150ppm as a short term exposure limit (STEL). The American
Conference of Governmental Industrial Hygienists (ACGIH) recommends a TWA of 30ppm (NIOSH 2002).
Mandelic acid, a metabolite of styrene is a biological marker for styrene and can be measured in urine.
Animal experiments – The effects of styrene on the auditory system
The ototoxic effects of styrene were clearly demonstrated by Pryor et al (1987) when weanling male Fischer rats
were exposed to 800, 1000 and 1200ppm daily 14hrs/day, 6 weeks for mixed xylenes and 3 weeks for styrene.
ABR and behavioural audiometry results demonstrated a more potent effect of the two solvents compared to
toluene ranging from 2-20kHz (the highest frequency tested).
Results of ABR and histological data from young male rats exposed to styrene (800ppm, 14hrs/day, 5days/week
for 3 weeks) also indicated the ototoxic effects of styrene (Yano et al 1992). ABRs were minimally affected at
4kHz, more severely affected at 8 and 16kHz and to a lesser extent 30 kHz, indicating a mid–frequency hearing
loss in the rat. A loss of OHCs occurred in the basal and lower middle turn of the cochlea. The pattern of OHC
loss matched that found with toluene i.e. OHC3>OHC2>OHC1. Some IHC loss was noted in the regions of
extensive OHC loss.
At this stage of investigations into styrene ototoxicity similarities can be seen between the effects of toluene and
styrene exposure. However, with lower levels of concentrates over shorter duration periods styrene is seen to
have a more damaging effect over a wider range of frequencies in the rat cochlea.
The effects of styrene were noted at 8 and 16kHz by Crofton et al (1994) on rats exposed to styrene (1600ppm,
8hrs/day for 5days) using behavioural audiometry. Whilst this is consistent with previous findings at these two
frequencies, the lack of effect seen at higher and lower frequencies could be due to the different exposure
regimes and test parameters also behavioural audiometry may not be the most sensitive means of testing solvent
ototoxicity.
As with toluene Campo et al (1999) found traces of styrene in the blood, brain, auditory nerve and cochlea of
exposed male Long-Evans rats following 10hrs of exposure over 2 days to 1750ppm styrene whilst investigating
toluene and styrene intoxication route in the rat cochlea. No traces of the solvents were found in the cerebral
spinal fluids (CSF) or inner ear liquids (IEL). However, the uptake of styrene was significantly higher than that
of toluene, in both the blood and tissue regardless of the tissue examined. The preferential route of intoxication
for styrene proposed by Campo et al (1999) is thought to be the same as for toluene. Lataye et al (2001)
proposed a second route of intoxication for styrene. With higher concentrations of styrene damage was also
noted in the spiral ganglion cells (SPG) predominantly in the middle and mid-basal turn of the cochlea along
with damage to OHCs..
A comparison of toluene-induced and styrene-induced hearing losses (Loquet et al 1999) clearly demonstrates
significant auditory threshold shifts as a function of the concentration of both solvents. For toluene the peak
threshold shift occurred at 16-20kHz, for styrene it occurred at 12-16kHz, and at 4kHz and 32kHz. Unlike the
experiment by Yano et al (1992) no threshold shift was noted at 4kHz and 30kHz at styrene exposure levels of
850ppm. Both styrene and toluene exposure result in the same pattern of damage to OHCs, peaking at 4kHz and
20kHz. However styrene is more potent than toluene for example 2000ppm toluene resulted in 22.5dB
permanent threshold shift (PTS) at 16kHz, whilst a 20dB and 35dB PTS was obtained with styrene 850ppm (a
2.4 times lower dosage than toluene) and 1500ppm respectively. Styrene concentrates of 1000ppm and above
10
June 2004
NoiseChem 11
cause auditory threshold shifts in the rat at mid, mid-low, and high frequencies i.e. styrene is frequency
independent at high concentrations (Loquet et al 1999).
A combination of exposure to both inhaled styrene (750ppm, 5days/week, 4 weeks) and ethanol gavaged,(4g/kg)
resulted in greater permanent threshold shifts than induced by styrene alone, ethanol having resulted in no
change in thresholds. Histological data also revealed greater OHC losses in the combined group. (Loquet et al
2000)
A more detailed look at the mechanisms of effect of styrene on OHC concluded that Hensen’s cells and Deiters’
cells, are the real target of styrene which then go on to cause the death of OHCs (Campo et al 2001). One week
after exposure to styrene (1000ppm, 6hrs/day, 5days/week) light microscopy showed partial loss of cytoplasm in
Hensen’s cells. After 2 weeks of exposure cytoplasm in Deiters’ cell was abnormal, the expansion of the
supporting cells obliterated the spaces of Nuel, OHC3 was missing and the base of OHC2 was no longer well
delineated. Campo et al (2001) suggest that “the disturbances of supporting cells could generate an excessive ion
build-up beneath the OHCs causing dramatic metabolic changes of the sensory cells. In fact, the OHCs could be
poisoned in this way.”
The abovementioned study also indicated that 1 week of styrene exposure (1000ppm) caused as much PTS as 2,3
or 4 weeks exposure. In addition, the threshold shift did not peak after one week’s exposure, but increased up to
15dB by 6 weeks post exposure. Thus trauma continued after exposure ceased. This process was also noted by
Crofton and Zhao (1997) in a study of the ototoxic effects of trichloroethylene (TCE). These effects are clearly
far reaching from the occupational view point.
Animal experiments – The effects of styrene on the vestibular system
Rabbits exposed to styrene demonstrated positional nystagmus, indicating vestibular disturbances. The incidence
of the positional nystagmus correlated well with the blood level of the solvent. (Larsby et al 1978)
In animal experiments looking at the effects of solvent exposure on the vestibular system (Odkvist et al 1979)
rabbits were given an intravenous infusion of either styrene, xylene, trichloroethylene(TCE) or
methylchloroform. Positional nystagmus was elicited when styrene blood concentration reached 40ppm, xylene
and TCE blood concentrations reached 30ppm and methylchloroform blood concentration reached 75ppm. The
fast phase of nystagmus was left beating in the right lateral position and right beating in the left lateral position.
Cats given either styrene or TCE showed no signs of nystagmus in the lateral positions but exhibited vertical
downbeat nystagmus in the prone position with TCE. Styrene had both an inhibiting (with high doses) and
facilitating (with low doses) effect on optokinetic nystagmus (OKN). Odkvist et al (1979) suggested two
possible sites of action, the first directly on the cerebellum, the second on the oculomotor system or reticular
formation. An important finding of this paper was the degree of potentiation between simultaneous doses of
styrene and TCE (see fig 2).
Animal experiments – Interaction between noise and styrene
As previously stated guinea pigs do not appear to be affected by styrene. Fetcher et al (1993) administered
styrene by injection or inhalation (500ppm/7hours) with no obvious deleterious affects on hearing. The addition
of 95dB broadband noise resulted in a noise induced hearing loss (NIHL) but this was no enhancement of NIHL
by styrene.
In the rat the pattern of trauma for simultaneous exposure to styrene and noise is very similar to that of toluene
and noise. In a group of experiments (Lataye et al 2000) adult Long-Evans rats exposed to noise at 97dBSPL
6hrs/day, 5days/week for 4 weeks presented with a NIHL in the frequency range 8-20kHz with a peak loss near
12kHz. Styrene exposure (750ppm 6hrs/day, 5days/week for 4 weeks) resulted in a hearing loss in narrower
frequency range of 16-20kHz. Simultaneous exposure to noise and styrene led to a hearing loss in the frequency
range 8-20kHz with a peak loss at 12kHz. Simultaneous exposure to noise and styrene also resulted in
significantly greater permanent threshold shifts compared to those caused by either noise or styrene alone. A
significant synergy appears between the two agents at 6-12kHz, with affects being additive at the other
frequencies.
11
June 2004
NoiseChem 12
The order of trauma for simultaneous exposure group followed that for styrene. OHC3 was more severely
affected than OHC2, which in turn was more severely affected than OHC1 but more pronounced. This is a
different pattern of that seen in NIHL where OHC1>IHC>OHC2>OHC3. Splayed stereocilia were observed on
IHCs at the most damaged frequency area, 8-20kHz for noise exposure and simultaneous noise and styrene
exposure.( see tables IV,V,VI )
Makitie et al (2003) looked at the effects of styrene, styrene and noise exposure at 1,2,4 and 8kHz (human
hearing is usually tested between 125Hz-8kHz), a much narrower frequency range than previously tested, along
with lower concentrations of 100, 300 and 600ppm (100ppm being the OEL for the UK and USA). Styrene alone
at 100ppm and 300ppm concentrations did not have an adverse affect on the auditory sensitivity in the rat,
600ppm styrene resulted in a 1-3 dB PTS at 8kHz. Simultaneous exposure to styrene (100ppm and 300ppm) and
noise (100-105dB) resulted in flat PTS of 5-10dB across all frequencies tested. 600ppm styrene with 100-105dB
noise resulted in more significant PTS across all frequencies, the greatest ranging from 31.4dB at 2kHz – 40dB
at 8kHz. These findings agree with those of Lataye et al (2000) in that a synergistic interaction between noise
and styrene occurred.
Carbon Disulfide
Carbon Disulfide (CS2) in its pure form is a colourless liquid with a smell similar to chloroform. The commonest
form used in industry is impure and yellowish in colour with an unpleasant odour. It is made by combining
carbon and sulphur at very high temperatures.
12
NoiseChem 13
June 2004
Carbon disulfide evaporates at room temperature and is highly inflammable. It is listed as an extremely
hazardous substance.
Noise 97dBSPL
OHC3 OHC2 OHC1
5%
3%
12%
17kHz
region
Table IV depicts percentages of outer hair cell loss for noise-exposed animals (Lataye et al 2000)
Styrene 750ppm
OHC3 OHC2 OHC1
20 kHz region 86%
25%
15%
4-5kHz region 70%
36%
17%
Table V depicts percentages of outer hair cell loss for styrene-exposed animals (Lataye et al 2000)
Styrene 750ppm+
Noise 97dBSPL
OHC3 OHC2
kHz 94%
42%
20
region
4kHz region 86%
62%
Table VI depicts hair cell loss for simultaneous noise and styrene-exposed animals (Lataye et al 2000)
It has played an important role in industry since the 1800s. It was first recognized as an occupational hazard in
the 1840s when cold vulcanisation (a strengthening process for rubber) was introduced.
Several incidences of neurotoxicity were noted in a number of countries using the process and as a result carbon
disulfide was eliminated from the process. (Rybak 1992)
Carbon disulfide has many useful properties and was previously used in many extraction processes. Pre 1985 it
was used as a grain fumigant. At present its most important industrial use in the manufacturing process for rayon
and cellophane, and as a solvent.
Exposure to carbon disulfide can be through inhalation, ingestion, skin or eye contact or it can be absorbed
through the skin. Acute and chronic forms of poisoning can result from exposure, at very high levels it can be
life threatening because of the effects on the heart and nervous system. The effects of exposure are non-specific
and require a diagnosis based on exposure history, signs or symptoms and exclusion of other diseases.
Occupational Exposure Limits for Carbon Disulfide
NIOSH has a recommended exposure limit of 1ppm as an 8hour time-weighted average (TWA) with a short
term exposure limit (STEL) of 10ppm. OSHA has a permissible exposure limit (PEL) of 20ppm as an 8 hour
TWA 2-thiothiazoladine-4-carboxylic acid, is a biological marker for carbon disulfide and can be measured in
urine.
Animal experiments – The effects of carbon disulfide on the auditory system
Experiments on Fischer-344 rats exposed to carbon disulfide (172, 286 and 400 mg/kg/5days/wk, 11weeks)
administered intraperitoneally resulted in ABR latencies in wave V but not in wave I, indicating an effect on
conduction within the central auditory pathway. No histological data was reported. (Rebert et al 1986)
ABR results from Long Evans rats exposed to carbon disulfide (400 or 800ppm/7hrs/day, 7days/week for 11
weeks) were also consistent with a retrocochlear pattern of hearing loss. Interpeak latency was also prolonged.
An additional peripheral loss of a conductive nature was suspected, possibly due to effects on eustachian tube
function. (Rebert and Becker 1986)
13
June 2004
NoiseChem 14
No significant effect was noted in pure-tone thresholds in rats exposed to carbon disulfide (500ppm 6hrs/day,
5days/week for 12 weeks). At this exposure level severe neuromuscular compromise occurred highlighting the
fact that pure-tone audiometry(PTA) is not sensitive to the early effects of carbon disulfide exposure. (Clerici
and Fetcher 1991)
The effects of carbon disulfide on the vestibular system, or the effect of simultaneous exposure to noise and
carbon disulfide has not been well documented in animals.
Xylene
Xylene, CH6H4(CH3)2 is an aromatic hydrocarbon. It is a colourless liquid with a sweet smell. It is primarily a
man-made chemical produced from petroleum and to a lesser extent coal.
There are three forms of xylene. Mixed xylene is a mixture of all three and usually contains 6-15% ethylbenzene.
It is often found in thinners, paints and varnishes, and is used as a solvent in printing, rubber and leather
industries. It is also used as a material in the chemical, plastic and synthetic fibre industries.(ATSDR 1990)
Xylene can be absorbed in the body either through inhalation or absorption through the skin. Short term
exposure to xylene can cause irritation to the eyes, skin and respiratory tract, breathing difficulties and impaired
memory. Long term or repeated exposure can lead effects on the nervous system, dizziness and lack of muscle
coordination. Xylene is known to be ototoxic.
Occupational Exposure Limits for Xylene
Occupational Exposure Limits for xylene vary between 35ppm in Denmark and 100ppm in the UK, Poland,
Japan and USA.
Methylhippuric acid, a metabolite of xylene is a biological marker for xylene and can be measured in urine.
Animal experiments – The effects of xylene on the auditory system
As previously stated mixed xylene was found to have marked effect on the hearing sensitivity in exposed rats
(Prior et al 1987).
A slight loss of hearing sensitivity was noted 2 days after the end of exposure by Nylen and Hagman (1994) in
rats exposed to xylene(1000ppm 18hrs/day 7days/wk during 61 days). Simultaneous exposure with 1000ppm nhexane resulted in a persistent loss of auditory sensitivity of 7-17dB.
In experiments looking at the three different forms of xylene(meta-xylene, ortho-xylene, para-xylene) only
para-xylene at 900 and 1800ppm was noted to have ototoxic effect. Exposure levels for male Sprague-Dawley
rats were 450,900 and 1800ppm 6hr/day 6days/wk for 13 weeks). ABR results indicated increased auditory
thresholds (35-38dB) at 2,4,8 and 16kHz in the group exposed to 1800ppm xylene. These were still present 8
weeks post exposure. Histological investigations found moderate and severe loss of outer hair cells for 900 and
1800ppm xylene.
The effects of xylene on the vestibular system (see Odkvist et al 1979), or the effect of simultaneous exposure to
noise and xylene have not been well documented in animals.
Ethyl Benzene
Ethyl Benzene (C8H8) is an aromatic hydrocarbon. It is used as a solvent, in the production of styrene and also in
the rubber, plastics and petroleum industries. It is also present in mixed xylene.
Occupational Exposure Limits for Ethyl Benzene
The general occupational exposure limit for ethyl benzene is 100ppm.
Animal experiments – The effects of ethyl benzene on the auditory system
Experiments involving male rats exposed to ethyl benzene (800ppm, 8hrs/day, 5days) concluded that ethyl
benzene does induce a permanent hearing loss in rats (Cappeart et al 1999). Evidence from reflex modification
audiometry(RMA), CAP, and histological data all point to similar effects to toluene and styrene, particularly
regarding pattern of outer hair cell loss, OHC3>OHC2>OHC1, with inner hair cells remaining undamaged.
14
June 2004
NoiseChem 15
(Sullivan et al 1988) (Campo et al 1997) (Yano et al 1992). OHC loss occurred in the upper basal and lower
middle turns ( the mid-frequency regions). Prior to this very little was noted regarding the site of ototoxic action
of ethyl benzene.
RMA results show that the frequency range 4-24kHz was significantly affected by about 25dB and CAP
thresholds increased significantly by 10-30dB at all frequencies tested. These results point to the ototoxic effects
of ethyl benzene being frequency independent at this level of exposure. Experiments by Campo et al (1997)
showed increasing doses of toluene leading to an increasing range of frequencies affected, this was also noted
by Pryor et al (1987) and Loquet et al (1999),. This would suggest that higher doses of styrene and toluene are
required to induce comparable effects to that found in this study. These in turn could lead to the conclusion that
ethyl benzene is an extremely ototoxic solvent.
Further experiments by Cappaert et al (2000) exposing rats to ethyl benzene (0, 300, 400 and 550ppm 8hrs/day,
5days) indicated a dose dependent loss of auditory sensitivity, measured using DPOAEs and CAP, following
exposure. 0 and 300ppm had no effect on auditory sensitivity, 400ppm affected 12 and 16kHz by 15 and 16dB
respectively, and 550ppm affected 8, 12 and 16kHz by 24, 31 22dB respectively. Histological data corresponded
with a mid-frequency hearing loss. The combined results of the experiments by Cappaert et al (1999,2000) and
Pryor and Rebert(1993) confirm the fact that ethyl benzene is probably one of the most ototoxic organic solvents
and yet it has not been extensively examined.
The effects of ethyl benzene on the vestibular system have not been well documented in animals.
Animal experiments – Interaction between noise and ethyl benzene
Some interesting results were noted by Cappaert et al (2001) when studying the simultaneous effects of low
levels of ethyl benzene and noise exposure on rats.
Exposure to ethyl benzene (300 and 4000ppp 8hrs/day, 5 days) did not result in significant threshold shifts in
DPOAE or CAP. This was not the case for 400ppm in a previous experiment by Cappaert et al (2000), but, OHC
losses, specifically in the third row, were noted in the mid-frequency regions of the cochlea. The affected region
was noted to broaden with increased doses. Noise exposure at 95dBSPL and 105dBSPL resulted in only minor
loss of OHCs in the first row at 105dBSPL. Combined exposure did not result in any threshold shifts as
measured by DPOAEs or CAP, indicating the absence of synergistic effects at the low levels of exposure.
However, morphological data did suggest a synergistic effect between the two agents.
Trichloroethylene (TCE)
Trichloroethylene (TCE) is a colourless, non-corrosive solvent first introduced as a degreaser in Germany during
the First World War. It is currently used as a grease remover, dry cleaning agent and is found as an ingredient in
rug cleaning solutions and as a spot remover. It is also used in the production of paints, waxes and pesticides.
High level exposure to TCE has resulted in decreased visual perception skills and motor skills. (NIOSH 1996)
Occupational Exposure Limits for Trichloroethylene
Occupational Exposure Limits for TCE vary between 25ppm and 100ppm. NIOSH has a recommended
2ppm/60minute ceiling when used as an anaesthetic and a 25ppm TWA for all other exposures. OSHA have a
PEL of 100ppm TWA, ACGIH have a 50ppm TWA. (NIOSH 1996)
Trichloroacetic acid and trichlorpethynol are biological markers for TCE and can be measured in urine.
Animal experiments – The effects of trichloroethylene on the auditory system
Initial experiments on male Long Evans and Fischer rats exposed to TCE (1600 or 3200ppm for 12 weeks)
demonstrated its ototoxic effects. (Rebert et al 1991). Increased ABR thresholds were noted in what is now
termed as the mid-frequencies in rats.
Jaspers et al (1993) noted that rats exposed to TCE (0,1500 or 3000ppm, 18hrs/day, 5days/wk for 3 weeks)
exhibited increased hearing thresholds at 20kHz for the 3000ppm group. There was no significant increase at 5
or 35kHz indicating a short-term, high-level effect of TCE exposure in the mid-frequency region.
15
June 2004
NoiseChem 16
Crofton and Zhao (1993, 1997) and Crofton et al (1994) also demonstrated, using reflex modification
audiometry(RMA), a mid-frequency hearing loss in rats exposed to TCE at 16kHz, 16kHz and 8 and 16kHz
respectively.
Whilst the abovementioned studies on the ototoxic effects of TCE confirm a clear pattern of the frequency range
affected they are unable to determine the site of trauma.
RMA thresholds of rats exposed to TCE (0 or 4000ppm, 6hrs/day, 5days) in a study by Fetcher et al (1997)
confirmed previous findings of a mid-frequency hearing loss (Crofton and Zhao 1993,1997) (Crofton et al 1994).
Auditory thresholds were elevated by 25dB at 8 and 16kHz in rats exposed to TCE. CAP thresholds showed an
impairment of 20dB restricted to 8 and 16kHz. The cochlear microphonic (CM which appears to arise from the
OHCs in the basal most turn of the cochlea) was not affected. CAP data and CM data suggest that TCE affects
the IHCs and/or the spiral ganglion cells.
Histological investigations from the experiment did reveal a marked decrease in the number of spiral ganglion
cells whilst OHCs were rarely seen to be missing. In a number of sections of the cochlea the loss of spiral
ganglion cells was obvious even though IHCs and OHCs in the same section were intact. Whilst within the
experiment it was not possible to quantify hair cell loss the subjective impression was that spiral ganglion cell
decline was consistent in all sections whilst hair cell loss was infrequent. TCE exposure decreased the numbers
of spiral ganglion cells by 47% in the lower middle turn and 43% in the upper middle turn of the cochlea.
Clearly whilst the region of damage (i.e. a mid-frequency cochlear site of action) caused by TCE is similar to
that caused by toluene, styrene, xylene and ethyl benzene, the mechanism of damage is quite different i.e.
toluene, styrene, xylene and ethyl benzene cause damage to outer hair cell, TCE to spiral ganglion cells. The
reasons for the two different patterns of morphological damage are not clearly understood.
Animal experiments – The effects of trichloroethylene on the vestibular system
Experiments in this field are quite limited and have been discussed in a previous section.
Animal experiments – Interaction between noise and trichloroethylene
There is relatively little information regarding this field, but an experiment by Muijser et al (2000) points to
combined exposure to TCE (3000ppm 18hrs/day, 5days/wk, 3 weeks) and noise (95dBSPL) having a synergistic
effect on the auditory sensitivity in rats at 4kHz and possibly 24kHz(results at 24kHz not confirmed
statistically). There was no histological data available for this experiment.
Main findings from the animal experiments
There is a considerable amount of evidence from animal studies, using a variety of test parameters (see tables
VII, VIII), pointing to the effects of industrial solvent on hearing and balance, the majority of which have been
carried out on rodent species.
Below are noted a number of the important findings.
• Industrial solvents can be ototoxic, causing a permanent hearing loss in rats.
• The effects of solvents are species dependent
• Rodents are more sensitive to the effects of solvents than guinea pigs and chinchillas and, as with
the case of styrene, metabolism in the human is more in-line with rodents than guinea pigs. (Morata
et al 2002).
• Studies show that the effects of solvents exposure, with the exception of carbon disulfide, are related
to cochlear damage.
• Exposure to solvents causes both functional and structural damage to the cochlea
• OHCs are more venerable to solvent exposure with the degree of damage to OHCs being
OHC3>OHC2>OHC1 , the exceptions being carbon disulfide for which no histological data was
found for this study, and TCE
• Spiral ganglion cells are most venerable to TCE exposure.
• The effects of carbon disulfide exposure suggest a retrocochlear pattern of hearing loss.
16
June 2004
•
•
•
•
•
•
•
NoiseChem 17
A mid-frequency hearing loss is most widely reported with evidence of a wider spread of
frequencies affected and greater threshold shifts with higher concentrations of solvents i.e. dose
dependent.
The ototoxic effects of solvents continue after exposure ceases.
Styrene and xylene are more potent than toluene, but ethyl benzene is even more potent.
There is evidence that combined exposure to solvents and noise has a synergistic effect. This can
happen when levels of exposure are below permitted levels.
Evidence points to the pattern of trauma from simultaneous exposure to noise and solvents mirroring
that of the solvent, but the scale of the trauma is more enhanced.
The damage caused by simultaneous noise and solvent exposure does not involve a new
ototraumatic mechanism.
There is a critical level when synergy occurs. It is important to find out that level for occupational
exposure.
17
NoiseChem 18
June 2004
Table VII Studies on animals exposed to toluene
Author
Toluene
Duration
exposure levels exposure
of Frequencies affected
Audiometric test
Location of OHC loss
(Pryor et al 1200
1983)
1400ppm
(Rebert et al 1200
1983)
1400ppm
Pryor et
1984a
or 14hrs/d, 7d/wk 8-16kHz
ABR
____
5wks
or 14hrs/d, 7d/wk 4 8-20kHz(highest
ABR
or wks
frequency tested) peak
____
at 16kHz
al 1200ppm
14hrs/d, 7d/wk 8-20kHz(highest
ABR/behavioural Basal turn weanlings
5wks
frequency tested) peak
at 16kHz
al 400 or 700ppm 16 weeks
no effect
ABR/behavioural
Pryor et
1984b
Pryor
and
Howard 1986
Sullivan et al
(1988)
injected
with 1 week
1.5 or 1.7 g/kg
gavaged daily 21 days
0.5ml and 1.0ml
/kg
(Johnson and 1400ppm,
16hrs/day
for
Canlon
8days
1994a,b)
Crofton et al 2500ppm
8hrs/day 5 days
1994
6hrs/day,
Campo et al 1500ppm,
5days/week
4
1750ppm
1997
weeks.
2000ppm
8-20kHz
behavioural
2-8kHz
ABR
1.6-20kHz
maximal ABR/DPOAE
between 6.3- 12.5kHz.
8-24kHz
behavioural
8-20kHz
ABR
2-32kHz (greater than ABR
+ 6hrs/day,
Lataye
and 2000ppm,
4 summated loss)
Campo (1997) noise 92dBSPL 5days/week,
weeks
Lataye et al 1750ppm
6hrs/day,
3-4kHz
CAP
1999
5days/week for 4 12-16kHz
weeks
Authors
description
of
hearing loss
High frequency
High frequency
middle and basal turn
Mid frequency
upper, middle, + part of mid-high
the basal turn some IHC frequency
loss at 4kHz
Mid frequency
mid-apical turn
middle turn
OHC
loss
pronounced
and Mid frequency
more
mid and mid-apical turn
mid and mid-low
frequency
18
NoiseChem 19
June 2004
Table VIII Studies on animals exposed to styrene
Author
Styrene
Duration
of
exposure
exposure
levels
Pryor et al 800, 1000 and daily 14hrs/day,
(1987)
1200ppm
3 weeks
Yano et al 800ppm,
14hrs/day,
(1992)
5days/week for 3
weeks
Crofton et 1600ppm,
8hrs/day
for
al (1994)
5days
6hrs/day,
Loquet et 850ppm
5days/week
4
al (1999)
weeks.
““
1000ppm
Frequencies affected
Audiometric
test
Location of OHC Authors description
loss
hearing loss
of
2-20kHz (highest frequency ABR
and
High frequency
tested)
behavioural
4-30kHz
ABR
basal and lower Mid frequency
middle turn
8 and 16kHz
behavioural
Mid frequency
12-20kHz
ABR
Mid frequency
12-16kHz +4kHz +32kHz
mid-low to high
““
““
““
6hrs/day,
5days/week
weeks
1500ppm
Lataye et 750ppm
al (2000)
750ppm
97dBSPL
Campo et 1000ppm
al (2001)
+
Lataye et 750ppm,
al (2001)
1000ppm,
1500ppm
Makitie et 100ppm
al (2003)
300ppm
600ppM
+ noise 100105 dB
16-20kHz
4
““
6hrs/day,
5days/week
weeks.
6hrs/day,
5days/week
weeks.
12hrs/day,
5days/week
weeks.
8-20KhZ(greater
summated loss)
16-20kHz
than ABR
ABR
1
12-16kHz
ABR
4
8khz ( highest frequency ABR
4 tested, only affected at
600ppm concentration)
flat loss 1-8khz (100ppm,
200ppm)
ABR
significant PTS 600ppm
“ “
frequency-independent
““
ABR
Middle
upper turn
and Mid frequency
SPGs affected in
middle and midbasal turn
Loss of OHC3 4.5-7.5mm
from round window (7mm
reflects site for 8kHz
Middle
and Loss of all OHCs in site
above
upper basal turn
19
June 2004
NoiseChem 20
• There is a synergistic effect of some solvents with other solvents, dependent upon their
molecular make-up.
• The preferential intoxication route would appear to be via blood from the stria vascularis
or spiral prominence diffusing through the outer sulcus.
• The effects of solvents on the vestibular system are not as well documented as the effects
on hearing.
• The effects on the vestibular system are neurotoxic, having an effect on the cerebellum.
• Solvents influence the vestibulo-occulomotor system, different solvents by different
mechanisms.
• There appears to be a synergistic effect of TCE and styrene on the vestibular system.
• Test parameters vary often making it difficult to compare results
• Toluene and styrene have been more extensively studied for ototoxic effects than other
solvents. In comparison very little is documented on the effects of TCE and xylene.
The above findings have far reaching implications for occupational exposure to industrial
solvents. Of greatest interest and concern is the evidence of a synergistic effect from
combined exposures to noise and solvents, a factor that has not really been taken into
consideration in industry previously and one that is not clearly recognised and understood.
Occupational studies-the effects of toluene on hearing and balance
Groups of workers in a printing factory exposed to high levels of toluene (100-365ppm) and
noise (88-98 dBA), noise alone (88-97dBA), and a group exposed to a mixture of solvents
were tested for hearing function (Morata 1998) compared to a control group. The majority of
average hearing losses in the groups exposed to solvents were up to 45dBHL. In the group
exposed to both toluene and noise the prevalence for sensorineural (permanent) hearing loss
was 53%. This was statistically higher than for the other groups. The prevalence in the
unexposed groups was 8%, in the noise exposed group 26% and the mixed solvent exposed
group 18%. Of the workers exposed to solvents only, 18% had a high frequency hearing loss
compared to 8% in the unexposed group. The adjusted relative risk for hearing loss in the
groups were 4 times greater for workers exposed to noise, 11 times greater for noise and
toluene, and 5 times greater for group exposed to mixed solvents. Acoustic reflex threshold
(ART) measurements pointed to lesions within the central auditory pathway. The indications
are that exposure to solvents can cause a hearing loss and there is a reaction between
simultaneous exposure to noise and solvents.
Exposure to toluene below 50ppm was studied in 333 rotogravure-printing workers over a
period of 5 yrs. involving 4 repeated examinations (Schaper et al 2003). The mean lifetime
weighted average exposure for toluene and noise were 45+/-17ppm + 82+/-7dBA respectively
for printers and 10+/-7ppm + 82+/-4dBA respectively for end processors. Exposure at the
time of testing for printers was 26+/-20ppm + 81+/-4dBA and 3+/-3ppm + 82+/-4dBA for
end processors.
The results of this study did not indicate any significant differences in the effects on the
auditory system as a result of the toluene exposure between the two groups. A control group
was not included in this study, but as noise exposure in both groups was almost identical any
differences in auditory thresholds between the two groups could be attributed to toluene.
However, the effect of noise was significant. It would appear from the data that the exposure
levels of toluene were not sufficiently high to induce a hearing loss and that the possible
threshold level for developing a hearing loss as a result of exposure to toluene could be above
the limit of 50ppm.
In 1981 Biscaldi et al carried out some neurophysiological studies on a small group of
workers accidentally exposed to toluene. Immediately after the accident abnormal brain
activities and a reduction in vestibular reflexes were present. Six months after the exposure
the vestibular findings were markedly improved but abnormal brain activities were still
present.
20
June 2004
NoiseChem 21
Small groups of volunteers exposed to a number of different solvents for a period of one hour
were tested to measure vestibular and occulomotor disturbances. (Odkvist et al 1983). The air
concentration values of the solvents varied between 1.3 and 3 times the legal threshold value.
Light exercise was performed to increase the intake. The test battery used included
electronystagmography (ENG) during sinusoidal oscillation with lights off, and fixation,
smooth pursuit, OKN, saccades and visual suppression. The group exposed to toluene
underwent additional rotatory tests with fixation and visual suppression.
The group exposed to styrene showed increased saccade speed and impaired visual
suppression (failure to suppress is usually a sign of cerebellar involvement), and high gain in
smooth pursuit during sinusoidal swing with fixation. Other tests were normal. The findings
for the group exposed to toluene were not as clear-cut as those for styrene. Styrene is more
potent than toluene (Pryor et al 1987, Yano et al 1992) and the dose for this experiment was
2-3 times the legal threshold value for styrene and 1.3-1.9 the legal threshold value for
toluene. The difference in potency and doses makes the two sets of results difficult to
compare. The group of volunteers exposed to TCE showed no significant disturbances in the
test battery. This could be explained by the differences in the molecular make up of aliphatic
solvents (TCE) and the aromatic solvents (toluene, styrene,)
A group of 11 workers with psycho-organic syndrome (POS) as a result of solvent exposure
were also included in the above study. The modern terminology for POS is chronic toxic
encephalopathy(CTE). The symptoms of CTE develop slowly. In the initial stages the
symptoms include vertigo, dysequilibrium, and nausea. Longstanding symptoms include
fatigue, poor concentration, intellectual reduction and dysequilibrium. (Odkvist et al 1992). 6
out the 11 workers had balance abnormalities when walking a line with eyes closed, 7 had
pathological spontaneous or positional nystagmus of low amplitude, 6 had pathological visual
suppression and 4 had above normal saccade speed. Out of 9 workers with less exposure to
industrial solvents 2 had abnormal findings in visual suppression and saccade speeds.
Tests on 15 workers exposed to toluene at threshold limit value (TLV) also showed impaired
visual suppression and increased saccade speed. (Hyden et al 1983)
A large-scale survey was undertaken in 4 cities in China to look at dose-dependant increase of
subjective symptoms as a result of exposure to toluene. 1316 workers exposed to toluene and
769 control subjects took part in the survey. Urine and blood samples were taken to monitor
the uptake of toluene.
Results showed an increase in subjective symptoms in the exposed group, which was closely
associated with the intensity of exposure levels to toluene; the threshold concentration
appeared to be 100ppm . The subjective symptoms included dizziness, floating feeling and
loss of hearing capacity. No diagnostic tests were undertaken and noise levels were not taken
in to consideration. (Ukai et al 1993).
21
June 2004
NoiseChem 22
Occupational studies-the effects of styrene on hearing and balance
The effects of simultaneous exposure to styrene and noise were examined by Sass-Kortsak et
al (1995) in a study of 299 workers in 14 different fibre-reinforced manufacturing plants.
Although noise and styrene levels differed between the plants noise levels generally fell
within the 85-90dBA range and styrene was generally below 50ppm.
The results of the study indicated minimal effects of styrene on pure-tone thresholds for
simultaneous exposure to noise and styrene when noise and styrene lifetime exposures were
taken into consideration. Age and noise however were positively associated with hearing loss.
Exposure to styrene only approached significance for 4 and 6kHz in the left ear.
More recent studies have looked at the effects of styrene and styrene and noise on hearing
(Morioka et al 1999, 2000). The first study looked at 115 workers in 7 factories. In 6 of the
factories workers were exposed to styrene, toluene and methanol in the production of plastic
buttons, in the other factory workers were exposed to styrene and acetone, producing fibrereinforced plastic baths. The mean duration of exposure was 9.4+/- 8.9yrs. Styrene
concentrations varied from 0.1 to 91.6ppm, outside of the permitted TWA of 50ppm.
Exposure to noise was below 85 dBA.
Conventional audiometry indicated no change as a result of styrene exposure. However, tests
to determine the upper limit of hearing of the exposed workers1 revealed evidence that
exposure to organic solvents for 5 years or more may cause a reduction in the upper limits of
hearing even for workers exposed to low levels of styrene. The results also pointed to the
effects of styrene on the upper limits of hearing being dose-dependent.
These results would suggest that styrene induces a hearing loss which progresses from higher
to lower frequencies in humans and that tests for the upper limit of hearing may be a useful
diagnostic tool for detecting early ototoxic effects of styrene exposure. This does not correlate
with findings of a mid-frequency hearing loss in animals.
As the effects of noise were not taken into consideration in the first study the second study
(2000) comprised of 3 groups according to exposure conditions. The first group were exposed
to low levels of styrene(2.9-28.9ppm), methanol (4.7-34.3ppm) and methanol (9.1-69.7ppm)
and noise (69-76dBA). The second group was exposed to noise (82-86dBA), and the third
was an unexposed control group.
The prevalence of reduction in the upper limit of hearing was higher in the combined group
than expected even when exposures were within the OELs. In conventional audiometry no
significant findings were noted.
In (2001a) a study by Sliwinska-Kowalska et al also looked at combined exposure to styrene
and noise in the plastics industry. The study involved 3 groups of workers matched by age,
gender and personal traits. One group was exposed to styrene and noise, one group was
exposed to noise at twice the level of the combined group, and one group was exposed to
neither styrene nor noise.
Workers in the combined group had a significantly increased average hearing loss at 1-8kHz
compared to the other groups. The risk of hearing loss in the combined group was 7 times
higher than the non-exposed group and 4 times higher than the noise exposed group,
suggesting that styrene has an additional adverse affect to noise on hearing.
An extensive audilogical test battery was carried out on 313 workers exposed to styrene
(below recommended values) and noise, styrene, noise, and unexposed controls. Pure tone
audiometry showed significantly worse thresholds at 2, 3, 4 and 6kHz for workers exposed to
styrene, or styrene and noise than the other two groups. The three exposed groups showed
significantly poorer results on the speech in noise test, but no significant differences between
the groups were seen in the interrupted speech test or cortical response audiometry. DPOAE
results though not fully analysed do not appear to show any differences between the groups.
The results also showed that the amount of mandelic acid in urine was more likely to increase
the odds ratio for hearing loss than age or noise exposure. (Johnson et al 2002).
1
a test in which the subject listens to a tonal stimuli which can range from 0.5-50kHz, the
upper limit of hearing being the highest frequency the subject can perceive as a tone
22
June 2004
NoiseChem 23
Human volunteers exposed to styrene showed no indications of positional or spontaneous
nystagmus at blood/styrene concentration at 8.5ppm or less. (Odkvist et at 1979).
Further experiments, correlating with the above results, were carried out on 10 healthy
volunteers exposed to styrene for one hour (Odkvist et al 1982). Pulmonary uptake and blood
levels were carefully monitored and the blood/styrene concentration was equivalent to that
reached after several hours in a working environment with a styrene concentration level
within permitted limits. 8 out of the 10 subjects exhibited disturbed visual suppression. The
speed of saccades was significantly enhanced in 8 out of 10 volunteers. There were no
significant differences in rotatory and optokinetic nystagmus as a result of the exposure. No
spontaneous nystagmus was observed.
Postural control was investigated in 18 workers exposed to styrene for 6-15 years in the
plastic boat building industry where styrene evaporation considerable and ventilation was
poor, but levels were below permitted levels. 9 painters, with a diagnosis of CTE as a result of
solvent exposure were also included. These workers were mainly painters with an exposure
history from 8-30 years, exposure levels were unknown. The reference group consisted of 52
workers in the construction industry with no exposure to industrial solvents. (Ledin et al
1989)
The two exposed groups had significantly larger sway areas with eyes open and closed than
the reference group. The two exposed groups also showed significant lower ability to
suppress vestibular nystagmus than the reference group. The correlation between static
posturography and otoneurological tests were positive for the group exposed to styrene. These
results point to cerebellar involvement.
In another small study looking at the effects of exposure to styrene on balance and hearing,
Moller et al (1990) found evidence of central components in both audiometric and vestibular
test results amongst 18 workers with long-term exposure to low levels of styrene(25ppm).
Although there was no indication of hearing loss being caused by anything other than noise 7
of the 18 workers displayed disturbances in the central auditory pathway. The styreneexposed workers also displayed significantly larger sway areas than the reference group and
poorer ability to suppress vestibular nystagmus.
Occupational studies-the effects of carbon disulfide on hearing and balance
Carbon disulfide is thought to have an effect on conduction within the central auditory
pathway in rats (Rebert et al 1996). There have also been a number of studies looking at the
effects of carbon disulfide on hearing and balance in humans.
Morata et al (1989) set out to explore the effects of simultaneous exposure to noise and
carbon disulfide on two groups of workers in a rayon factory in Brazil. At the time of the
study although permissible standards of exposure existed in Brazil, threshold limits of most
hazardous agents were not respected due to pressures for industrial development.
Group A consisted of volunteers who reported no exposure to noise or carbon disulfide prior
to working in the factory, group B were randomly chosen following health examinations
which included hearing tests for all workers in the factory, these workers had no exposure to
noise or carbon disulfide prior to their current employment . All subjects underwent PTA at
least 12 hours post last exposure to avoid temporary threshold shifts as a result of industrial
noise. Group two underwent balance tests. All subjects were interviewed regarding health and
history of noise and carbon disulfide exposure and had PTA.
The PTA results showed a large proportion of workers with a hearing loss.(71.7% in group A,
66.7% in group B) Hearing losses were higher than expected for all age groups suggesting
that hearing was not simply attributable to ageing process. Further studies revealed that 12.7%
of workers exposed to carbon disulfide had hearing losses affecting both higher and lower
frequencies compared with 3.5% of workers exposed to noise alone. The indications are that
hearing loss occurs earlier and is more serious among workers exposed to carbon disulfide
and noise, than exposure to noise alone. Hearing losses increased with exposure time to
carbon disulfide and noise and could be observed after 3 years simultaneous exposure.
23
June 2004
NoiseChem 24
Balance screening was limited, but the largest group that failed had a hearing loss. Results
also showed a marked increase in failures on balance test following 3-5 years simultaneous
exposure to carbon disulfide and noise(3.1% 0-2 years exposure, 30.3% 3-5 years exposure).
These results suggest a possible relationship between hearing loss, balance disturbance and
exposure to carbon disulfide.
A group of 80 workers in a viscose fibre spinning mill with clinically observed carbon
disulfide poisoning and 40 people exposed to carbon disulfide but without subjective or
objective symptoms of carbon disulfide poisoning were studied to assess changes in the
auditory system resulting from carbon disulfide exposure. Both groups were of similar age
with similar durations of employment. Exposure to carbon disulfide varied in time between
10 and 35mg/m3 and continuous noise ranging from 88-92dBA. A control group of 40
workers had no contact with carbon disulfide but were exposed to similar levels of noise
ranging from 86-93dBA. (Kowalska et al 2000).
Bilateral retrocochlear hearing loss associate with central vestibular disturbances were noted
in 97.5% of the subjects with clinically observed carbon disulfide poisoning. In the exposed
group with no symptoms 45% had a retrocochlear hearing loss and 32% had a cochlear
hearing loss. 22% of those under study had normal hearing. In the control group cochlear
hearing loss typical of acoustic trauma without concomitant vestibular disorders was evident.
The results show a dominating carbon disulfide toxic effect on the hearing system in subjects
with clinically observed carbon disulfide poisoning and exposed to noise. Hearing impairment
and vestibular disorders that occur are central rather than peripheral. The variabilities in
subjects exposed to carbon disulfide and noise is probably due to individual susceptibility to
the harmful effects of both agents.
Variabilities in results on individuals expose to solvents have also been noted by Toppila et al
(2002) and Morioka et al (1999, 2000)
The effects of noise and carbon disulfide exposure were also noted by Chang et al (2003).
Approximately 80% of workers exposed to carbon disulfide(1.6-20.1ppm) and noise (8091dBHL) had a hearing loss >25dBHL compared to 32.4% of workers exposed to noise-only
(83-90dBHL) and 23% of workers exposed to lower levels of noise(75-82dBHL). Workers
exposed to carbon disulfide and noise had a greater hearing impairment at 0.5, 1, and 2kHz
than the noise exposed groups, the noise-only group had a stronger effect at 4kHz which is
characteristic of a NIHL.
A dose-response effect was also noted in this study suggesting that chronic exposure to
carbon disulfide >10ppm should be avoided in order to reduce the toxic effects on hearing in
workers.
In a group of 37 patients with chronic carbon disulfide intoxication complaining of vertigo
72% were found to have a postural stability disorder. These results correlated with ENG
results which confirmed damage to the central part of the vestibular system, due to carbon
disulfide intoxication (Sulkowski et al 1992)
Occupational studies-the effects of xylene on hearing and balance
Groups of workers in the paint and lacquer industry were studied to assess the risk and
incidence of hearing loss as a result of exposure to organic solvents whose main components
were xylene and ethyl acetate. Levels of exposure depended on individual work posts. PTA
results were compared with workers exposed to noise above the threshold limit value and a
group exposed to neither noise nor solvents. Hearing loss was found in 30% of workers
exposed to solvents, 20% in noise-exposed subjects and 6% in the non-exposed. The highest
PTA thresholds were found in the solvent exposed group, lower for the noise exposed group
and lowest for the non-exposed group. The relative risk of hearing impairment was also
greater in the solvent-exposed group. (Sliwinska-Kowalska et al 200) These findings were
confirmed in a larger study of 731 workers in the dockyard, paint and lacquer industries.
(Sliwinska-Kowalska et al 2002) The study comprised 4 groups exposed to organic solvents
alone (the main component being styrene), organic solvents + noise>80dBA, noise only
>80dBA, and one group exposed to neither noise nor organic solvents. The highest incidence
of hearing loss was found in workers exposed to organic solvents (48%), and organic
24
June 2004
NoiseChem 25
solvents + noise (42%) compared to the noise-exposed group26% and the control group
(17%). The range of frequencies affected in the solvent and solvent + noise group ranged
from 3-8kHz. The frequency affected by noise alone was 4kHz.
Xylene has been found to have an effect on postural sway in exposed workers. A TWA
xylene concentration of 100ppm with 200ppm peaks resulted in no observable effects on
balance but a TWA of 200ppm with 400ppm peaks did. The effects were more noticeable
with eyes closed. (Savolainen and Linnavuo 1979)
Xylene at concentrations of 636 and 1,218 mg/m3 for 4 hours had very little effect on the
vestibular system. When given with a single dose of 0.4 and 0.8 g/kg ethyl alcohol the effects
of the lower dose of xylene on body sway were additive and the higher dose of xylene
antagonized the effects of alcohol on positional nystagmus and body sway. Alcohol also
significantly increased the uptake of xylene.(Savolainen et al 1980) A dose related effect of
xylene on balance was also noted. (Savolainen et al 1985) Changes in body sway with eyes
open/closed correlated closely with xylene/blood concentrations.
Occupational studies-the effects of TCE on balance
150 workers in a jet engine section of an air force base were studied primarily for pulmonary
effects of metal dusts. The workers were also exposed to chlorinated and fluorinated solvents
so the effects of solvent exposure were also studied. (Kilburn 1999)
The level of exposure to solvents apparently was not measured but it was known that large
amounts of TCE and other chlorinated solvents had been assigned to the airbase. Most
workers from within the study came from areas where solvents were used to clean metals. A
significant difference was noted in sway speed (balance) with eyes open/closed in the exposed
group compared to the non-exposed group.
Occupational studies-the effects of mixed solvents on hearing and balance
Whilst a number of studies have looked at the effects of specific solvents, solvents + noise on
hearing and balance many workers are exposed to a mixture of solvents often with noise.
Otoneurological findings in workers with CTE due to exposure to organic solvents may have
impaired equilibrium and central vestibular disorders as seen in impaired visual suppression,
increased saccade speed, abnormal gain in smooth pursuit and pathological nystagmus. In
some cases balance performance deteriorated 5 years after exposure ceased. (Ledin et al
1991) Tests of central auditory function such as discrimination of interrupted speech and
evoked cortical responses to frequency glides (CRA-delta-f) indicate retrocochlear disorders.
PTA thresholds (tests of peripheral hearing function), and speech discrimination tests on the
other hand often show minimal effects. (Moller et al 1989, Ledin et al 1989, Ledin et al 1991,
Odkvist et al 1992, Pollastrini et al 1994, Niklasson et al 1997,1998) These studies do not
support the assumption that long-term exposure to industrial solvents causes cochlear
damage.
Long-term exposure to aliphatic and aromatic solvents and jet fuel was studied in 3 groups of
workers by (Odkvist et al (1987). Group A consisted of 16 subjects diagnosed with CTE,
group B 7 subjects with suspected CTE, both groups with long-term exposure to aliphatic and
aromatic solvents. Group C consisted of 8 workers with long-term exposure to jet fuel.
Discrimination of interrupted speech and CRA-delta-f produced significant abnormal results
in all groups, particularly group A. PTA and speech discrimination results correlated with
each other, there was no indication of hearing loss which could not be attributed to age or
exposure to noise. ABR and stapedius reflex testing did not show significant abnormalities.
Abnormal vestibular findings included positional nystagmus, reduced vestibulo-ocular reflex
suppression (VORS) and reduced gain in the smooth pursuit test. Vestibular pathology was
highest in group A and lowest in group C.
The Copenhagen male study (Jacobsen et al 1993), based on self assessment, reported that
exposure to solvents for 5 years or more resulted in a higher risk of hearing impairment than
for non-exposed in men not exposed to noise. Exposure to noise for 5 years or resulted in
twice the risk of hearing impairment than exposure to solvents. In men exposed to both noise
and solvents the effect of noise dominated and solvents were not assessed as having an
25
June 2004
NoiseChem 26
additive effect. Although the relative effect of solvent exposure was assessed as moderate in
this study, the attributable risk of a damaging effect from long-term exposure to solvents was
considerable.
Thirty-three workers with a varied level of exposure to solvents, both in type of solvent and
length of exposure underwent a comprehensive battery of audiological tests. Apart from a
mild to moderate SNHL found in some subjects (some typical for noise exposure) peripheral
tests were normal in most subjects. Central test, filtered speech and mismatch negativity
showed varying degrees of abnormality which was felt to reflect the differing levels of
disease resulting from solvent exposure.
Further studies by Morata et al (1997) specifically focussed on occupational hearing loss
among 438 refinery workers. The workers were exposed to solvents, which included benzene,
toluene, xylene and ethyl benzene, and noise. In the majority of cases exposure levels were
within permitted exposure levels.
The results show that for the group of workers with simultaneous exposure to noise and
solvents there was a increased prevalence of a high frequency hearing loss (42-50%).
Acoustic reflex decay testing pointed to an element of retrocochlear involvement.
The data suggested that workers exposed to solvents within permitted TWA levels were 2.4
times more likely to develop a hearing loss than non-exposed workers. Workers exposed to
solvents and level of noise considered high enough to cause a hearing loss were 3 times more
likely to develop a hearing loss than non-exposed workers and 1.8 times more likely to
develop a hearing loss than workers exposed to lower levels of noise and solvents.
Similar findings were also noted by Sliwinska-Kowalska et al (2001b) where the relative risk
factor of hearing loss in solvent-exposed workers was significantly increased in exposures to
solvents and noise in a wide range of frequencies (2-8kHz). In groups exposed to solvents,
and solvents and noise hearing thresholds at 1-8kHz were significantly poorer than in the nonexposed group. The poorest thresholds in the solvent and noise group were 2-4kHz.
In a field study (Sulkowski et al 2002) of workers exposed to a mixture of solvents hearing
and vestibular function was assessed in 61 workers and results compared with a control
group. Vestibular dysfunction was present in 47.5% of the exposed group and 5% of the
control group. A high frequency SNHL hearing loss was found in 42% of the exposed group
and 5% of the controls, identified using PTA and DPOAEs. The findings corresponded
closely with the rate of total exposure to the solvents.
Increased postural sway responses were also noted by Smith et al (1997) in US Air Force
personnel exposed to jet fuel. It was also noted in workers in a shoe, sandal and leather
factory exposed to n-hexane, xylene and toluene (Yokoyama et al 1997). In this study where
the mean level for n-hexane was 40ppm, exposed workers demonstrated significantly larger
sway than a control group of non-exposed workers. The pattern of changes in sway in the
exposed group suggested that the vestibulocerebellar and spinocerebellar systems were
affected by n-hexane. The exposed workers did not exhibit any other specific signs or
symptoms of solvent poisoning or neurological disorders. The results of this study would
suggest that measurements of postural sway may be sensitive to, and useful for testing, the
effects of exposure to solvents on balance.
26
June 2004
NoiseChem 27
Main findings of the occupational studies
The occupational effects on hearing and balance as a result of exposure to industrial solvents
with and without noise have not been as well studied as the effects on animals. Some of the
main findings are listed below
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Human studies suggest that solvents are ototoxic.
Exposure to industrial solvents can cause permanent hearing loss, both peripheral
and central, though results vary.
Hearing loss can occur at exposure levels within the permitted OEL.
There is evidence of a dose-response relationship between hearing thresholds and
levels of exposure.
A synergistic effect of combined noise and solvent exposure, even at levels below
permitted OELs has been noted though results are variable.
It is very difficult to separate out the individual effects in combined exposures.
Minimal effects are often noted on PTA thresholds.
PTA results can indicate a high-frequency loss, or a hearing loss which affects a
wider range of frequencies.
The indications are that hearing loss occurs earlier and is more serious among
workers exposed to carbon disulfide and noise
Effects of solvents continue after exposure ceases
Exposure to solvents can result in CNS disturbance.
Exposure to solvents can result in balance abnormalities.
Exposure to solvents can result in vestibular disturbance. Long-term exposure can
cause cerebellar lesions as indicated by decreased VORS, pathological
spontaneous and positional nystagmus, pathological increased saccade speed,
abnormal gain in smooth pursuit.
Vestibulo-oculomotor testing is valuable in testing CNS lesions in solvent
exposed workers
Study parameters vary often making it difficult to compare results
27
June 2004
NoiseChem 28
General consideration of Previous work
The main findings from the animal experiments are very important for human studies,
however extrapolating data from animal experiments to human cases is very complex.
In animal experiments although different test parameters have been used which make it
difficult to compare one set of results with another, the variables are fewer than in human
studies. Many additional factors need to be taken into account for humans which include,
increasing age, levels of noise exposure throughout a working life, personal traits, health
history, exposure history including and duration of exposure to solvents, individual
susceptibility to solvents and solvent uptake, all which could have an adverse effect on
hearing. There is a danger of confusing the relationship between exposure and hearing loss if
they are not carefully considered in studies.
The majority of results from animal experiments and human studies indicate that exposure to
solvents results in a permanent hearing loss. Negative results are few (Pryor et al 1984b,
Schaper et al 2003). The hearing loss in rats it is a mid-frequency cochlear loss with the
exception of the effects of TCE exposure. In human studies a high- frequency loss is noted
(Morata 1988, Morata et al1997) or effects on a wider range of frequencies are noted
(Johnson et al 2002, Sliwinska-Kowalska et al 2001b). Evidence of both cochlear and central
elements are noted in human studies (Laukli et al 1995, Morata et al 1997, 1998, Johnson et al
2002, Sliwinska-Kowalska et al 2000,2002).
The effect of industrial solvents on balance has not been extensively studied. However, there
do seem to be similarities in the site of action for both animals and humans. In both groups
solvents appear to act on the cerebellum, causing pathological nystagmus and effects on the
vestibulo-ocular reflex (Larsby et al 1978, Odkvist et al 1979, Niklasson et al 1993, Ledin et
al 1989, Odkvist et al 1992).
Animal experiments have concentrated on individual solvents with or without noise exposure.
Whilst some occupational studies have also looked at the effects of specific solvents with or
without noise, the majority of workers are exposed to a mixture or mixtures of solvents and
noise that may interact with each other in an additive or synergistic manner. The permutations
are numerous and very difficult to monitor and measure the effects of. Further more detailed
studies are needed in this area.
The combination of agents of greatest concern is simultaneous exposure to solvents and noise.
There are a number of examples of a synergistic reaction between these two agents in
experiments on rats (Lataye and Campo 1997, Lataye et al 2000, Johnson 1990.) There is also
some evidence in human studies of a synergistic reaction between the two agents (Morata et
al 1998, Sliwinska-Kowalska et al 2001). This could have far reaching implications in the
workplace. Exposure levels for each agent may currently be within permitted OELs, but can
still result in damage to hearing and balance when combined. If the presence of such synergy
were confirmed in humans OELs would need to be revised for both solvents and noise to
prevent future occupational hearing loss occurring.
In both animal and human studies there is evidence of a dose-response relationship and a
critical level at which trauma occurs. These have yet to be confirmed, but are critical in
securing safe levels of exposure for solvents alone and in combination with noise. If hearing
losses are developing at levels at or below the current PELs this would suggest that they are
already too high.
Animal experiments point to different susceptibility to solvent exposure between species.
Human studies point to different susceptibility between individuals. A better understanding of
these differences would help to determine which individuals would be more prone to the
adverse effects of solvent exposure.
In many countries the only agent considered ototraumatic is high intensity noise. With so
many workers exposed to solvents (10million in Europe alone) it would suggest that a high
number of unmet needs regarding hearing conservation exist. Few workers exposed to
solvents would be required to have a hearing test because the level of noise would not be
deemed hazardous so the extent of the problem may go unnoticed. Whilst it is now
recommended that workers exposed to solvents have their hearing tested there are no
regulations requiring this.
28
June 2004
NoiseChem 29
There are periods during a working life when workers are exposed to very high
concentrations for short periods of time. Exposure concentrations in the animal experiments
far exceed the TWA for occupational exposure in the majority of experiments. It may be more
relevant at this stage for experiments to use exposure levels closer to permitted OELs, taking
into consideration the safety factor considered relevant to human exposure. Duration of
exposure also needs careful consideration in further experiments, whilst very short in animal
experiments workers are usually exposed for a period of many years.
Balance problems resulting from solvent exposure have been neglected to a large extent, yet
evidence suggests it can be a real problem for exposed workers, and could potentially be very
dangerous. However, research in this field is quite limited making the need for further studies
of paramount importance.
An agreed regular standardized battery of tests sympathetic to both peripheral and central
elements of hearing and balance would be an effective means of monitoring the effects of
solvent exposure and aiding research purposes. These tests would need to be quick and easy
to administer and be acceptable to the work force.
Materials and Methods
NoiseChem group felt it necessary that standardised procedures be developed for effective
field evaluation of noise and industrial chemicals exposure and their effects and interactions
on hearing and balance function.
A detailed questionnaire was developed which would provide a thorough exposure, personal,
and medical history which could then be used to extract exposure to both noise and chemicals
over a lifetime of work for all individuals. The medical and personal histories allowed
researchers to exclude certain workers on the basis of their previous medical history which
may have an impact on the association which may be developed between exposure and results
of tests.
From previous work it was felt that pure tone audiometric assessments routinely conducted
for the evaluation of noise exposure effects may be an inadequate measure of the solvent
exposure effects particularly as central auditory pathway disturbance may occur.
A questionnaire was devised (appendix1). This was also implemented on the computer. The
software incorporated the questionnaire information and the test results. A number of
individual screens were provided which included administration, personal information, work
history, armed forces information, exposure information including leisure, medical history,
clinical tests including pure tone audiometry, tympanometry, acoustic reflex thresholds,
otoacoustic emissions, auditory brainstem responses, cortical responses, cognitive potentials,
electro-nystgomgraphy, posturography, urine analysis. Additional tests conducted by some
labs were speech-in-noise, frequency ramp cortical responses, and masking level difference.
A number of audio-vestibular tests were considered as core in the battery of tests across the
laboratories, others were only implemented by certain labs. The core consisted of otoscopy,
tympanometry, pure tone audiometry, acoustic reflex thresholds, otoacoustic emissions,
auditory brainstem responses, and posturography.
The otoacoustic emissions which provide direct information about the outer hair cell activity
were to be evaluated as an additional tool for the screening of auditory damage and as an
early indicator of noise damage. In addition the brainstem response would provide
information regarding any effect of solvent exposure on the central auditory pathway.
29
NoiseChem 30
June 2004
The table below provides details of the number, mean age, noise exposure levels and styrene
measures for the subjects examined by the six laboratories (Lab 1-6).
Lab Exposure group
1
control
noise (n)
styrene
styr. + n
Total
2
control
noise
styrene
styr. + n
solv. mix
solv. mix + n
Total
3
control
noise
styrene
solv. mix + n
Total
4
noise
styrene
Total
5
solvent mix
solv. mix + n
Total
6
control
noise
solv. mix
solv. mix + n
Total
N
Age
59
78
65
88
290
26.5 (11.2)
32 (9.1)
31.3 (10.4)
31 (6.3)
97
17
41
62
20
51
288
26.5 (11.1)
43.3 (11.1)
31.5 (9.9)
32 (8.8)
27.2 (2.7)
30.6 (6.4)
110
154
114
137
515
37.5 (9.1)
39.5 (9.3)
35.2 (8.9)
41.1 (8.3)
180
126
306
63
199
262
39
153
13
174
379
47.6 (14.8)
53.3 (7.8)
49.6 (14.2)
47.4 (7.5)
30
June 2004
NoiseChem 31
Results
In this section the results from each laboratory are presented individually and then a summary
of the results is presented together with an overall analysis of the data.
Firstly data from animal experiments is presented and then the epidemiological data from 6
centres. Each section is presented independently and subsequently the impact of joint data is
considered.
31
NoiseChem 32
June 2004
NoiseChem
Animal Studies Lab Reports
32
Animal Lab 1: Pierre Campo
Institut National de Recherche et de Sécurité,
France
June 2004
NoiseChem 34
General introduction
In the NoiseChem project, the participants wanted to examine the interaction effects of chemicals and
noise on hearing system of workers exposed to noise and chemicals together and each independently.
The need for research in this area was heightened by the fact that there are no guidelines or standards
for combined exposure of workers to chemicals and noise on hearing.
In this project, the participant 7 was involved in animal research.
Based on that, several options were chosen in the experiments reported in the present report.
Choice of the animal model
Because rat has a solvent metabolism close to that of human, rat can be used to study both solvent–
induced hearing loss and noise–induced hearing loss. Moreover, rat can be also considered as a wellsuited model in aging research, Long-Evans rat was chosen as an animal model.
Choice of the toxic solvent
The highest occupational exposures to styrene occur during the production of glass-reinforced
polyester products, especially large items such as boats. For this reason, styrene was chosen as a
representative aromatic solvent in all the experiments reported in the present report. Because
inhalation is the main intoxication route for humans, the animals will be always exposed inside
inhalation chambers. Each inhalation chambers (200liters) designed to sustain dynamic and adjustable
airflow could contain up to eight animals. Styrene was also the targeted solvent in the epidemiological
studies.
Choice of the noise
Because noise can be accurately controlled in laboratory experiments, the rats were exposed to an
octave band noise centered at 8 kHz. The duration of the noise exposure will be identical to that of
solvent The spectrum of the noise exposure was chosen to cause hearing losses :
(1)
where auditory sensitivity is the highest and
(2)
in the frequency range impaired by aromatic solvents.
The workplan of the participant 7 can be summarized by several questions or avenues of
research such as :
-Is DPOAE a relevant early indicator for solvent-induced hearing loss ?
-Is there a difference of susceptibility to noise- and solvent-induced hearing loss between
young and aged subjects ?
-Could the plasticity of the CNS explain the non functional effects of small cochlear trauma?
-Are the standard values [85dB(A), 50ppm for styrene or 100ppm for toluene] recommended
by the Labour Ministry pertinent in case of complex environment?
Which are the mechanisms involved in the solvent ototoxicity?
All these questions summarized quite accurately the objectives targeted by the participant number 7 of
the NoiseChem project.
34
June 2004
NoiseChem 35
Chapter 1: Is DPOAE a relevant early indicator for solvent-induced hearing loss?
The current study was carried out to test whether or not cubic distortion otoacoustic emissions (2f1-f2)
were more efficient than auditory (inferior colliculus) evoked potentials (AEP) for assessing solventinduced hearing losses in Long-Evans rats. For the purposes of comparison between the two
techniques, changes in cubic distortion product otoacoustic emissions (∆DPOAE), auditory evoked
potential permanent threshold shifts (PTS) and outer hair cell losses were measured in a population of
solvent-treated rats.
Hearing testing
The animals were equipped with electrodes placed with the aid of a stereotaxic table. Of course, the
animals was first anaesthetized with a mixture of ketamine (50mg/kg) and xylazine (3 mg/kg). The
authors would want to mention that they adhere to the Guide for Care and Use of Laboratory Animal,
as promulgated by the French Conseil d'Etat through the Décret N° 87-848 published in the French
Journal Officiel on 20 October 1987. A tungsten microelectrode was surgically implanted to record
auditory potentials picked up from the inferior colliculus, whereas a ground electrode was implanted
in the nasal region. After the surgery, the animals was allowed to recover for 4 weeks before any
testing.
One month after the surgery, audiometric testing was performed in an audiometric room on awake rats
placed in a restraining device (Figure 1). Auditory evoked potentials (AEP) were recorded on awake
rats. The acoustic stimuli (two cycles for the rise/fall ramp, four cycles for the plateau) were filtered
clicks from 2 to 32 kHz presented at a rate of 20Hz and the analysis window lasted 30ms. The
electrical signal from the implanted electrode was amplified (X 2000) and filtered between 30 and
3000 Hz. Averaged AEPs were obtained from 260 presentations. As shown in Figure 1, an amplitude
trough-to-peak of 15 V of the response was considered as threshold value in our experimental
conditions. Hearing testing with AEP is detailed in Campo et al. (1997).
Figure 1 : Auditory evoked potentials (AEP) recorded in rats.
An amplitude trough-to-peak of 15µV of the response was
considered as threshold value in our experimental conditions
For each animal, an audiogram was performed prior the styrene exposure (T1), the day following the
end of the exposure (T2) and six-week post-exposure (T3). The compound and the permanent
threshold shifts were defined respectively as: CTS = T2-T1; PTS = T3-T1
35
PTS = T3-T1
CTS = T2-T1
EAPIC
DPOAE1
styrene
650, 700, 750 ppm
(T1)
2
4
EAPIC
DPOAE3
EAPIC
DPOAE2
6
(T3)
(T2)
10
8
Animals
surgery
12
14 weeks
Histology
Figure 2: Experimental protocol
Deep anesthesia was required to measure DPOAEs in the rat [Figure 3A]. The anesthetized rats were
placed on a temperature-controlled blanket to maintain normal body temperature (38°C). Primary
tones, f1 and f2, were presented at equal levels (L1=L2) and the levels were decreased in 10 dB steps
from 60 dB to 10 dB in order to obtain DPOAEs input/output functions (DPOAEs I/O).
A
B
Figure 3: Anesthetized animal was required to measure DPOAEs
The highest level was deliberately limited at 60dB SPL to preserve the auditory function. On the other
hand, the ratio of f1 to f2 was 1.2. f1 and f2 were produced by dual-channel frequency synthesizer and
attenuated under computer control by matched Tucker Davis Technologies, programmable attenuators
(PA4), and transduced by two miniature speakers (Knowles Electronics EA 1842). The frequencies of
the primaries were selected to yield what is commonly referred to as cubic difference tones at the 2f1f2 frequency [Figure 3B. The cochlear response was recorded with a microphone (Sennheiser KE4211.2) fitted into the probe. The three transducers were enclosed in the probe whose tip was gently
pressed against the opening of the ear canal.
The emitted response were averaged (N=100) to determine the 2f1-f2 amplitude. All DPOAEs I/O
curves were plotted as a function of the geometric mean frequency of the primaries from 2 to 16 kHz.
For each animal, DPgram [Figure 4] was performed prior the styrene exposure (DPOAE1), the day
following the end of the exposure (DPOAE2) and 6 weeks post-exposure (DPOAE3). ∆DPOAE was
defined as DPOAE1-DPOAE3.
NoiseChem 37
June 2004
Each rat was exposed to either 650, 700 or 750ppm of styrene for 4 weeks, 5 days per week, 6 hours
per day
Hearing testing with DPOAEs is detailed in Pouyatos et al. (2002).
Exploitation and dissemination of results
Figure 4A
DPgram
PDgram
DPOAE amplitude (dB)
20
Threshold shift (dB)
Figure 4B
L1 = L2 = 40 dB SPL
650ppm
15
10
5
750ppm
0
700ppm
-5
-10
Frequency (kHz)
-15
-20
5
40
35
30
25
20
15
10
5
0
-5
-10
6
8
10
12
16
0 ppm - PTS
750ppm
650 ppm - PTS
700 ppm - PTS
750 ppm - PTS
700ppm
650ppm
Frequency (kHz)
2
3
4
5
6
8
10 12 16 20 24 32
Comparison between DPgram (Figure 4A) and audiogram (Figure 4B) shows that DPOAEs are as
much sensitive to styrene as the audiometry performed with auditory evoked potentials, but not more.
As reported in Table 1, high coefficients of correlation [0.84≤r≤0.91] between ∆DPOAE and PTS
were obtained across the styrene-induced effects for the frequencies ranging from 5 to 12 kHz.
5 kHz
6 kHz
8 kHz
10 kHz
12 kHz
16 kHz
37
NoiseChem 38
June 2004
y30=0.52x-2.25
0.79x-0.47
0.83x-0.33
0.89x-5.87
0.56x+2.12
0.53x-0.50
r30=0.48
0.84
0.89
0.81
0.75
0.60
y40=0.98x-1.28
0.78x+0.05
0.71x+0.10
0.78x-3.21
0.65x-0.61
0.65x+0.98
r40=0.84
0.86
0.91
0.89
0.88
0.66
0.47x+1.63
0.90x-2.02
0.86x-3.81
0.80x-1.91
0.84x-4.58
r50=0.66
0.86
0.89
0.87
0.83
0.87
y60=0.29x+0.32
0.54x-1.56
0.55x-1.39
0.61x-1.82
0.70x-2.92
0.45x+0.77
r60=0.49
0.79
0.82
0.78
0.75
0.79
y50=0.40x+1.96
Table 1: Equations and correlation coefficients of the linear regression lines obtained from the scatter
plots [PTS vs. ∆DPOAE] with equal level primaries emitted at 30dB, 40dB, 50dB, 60dB.
rn: correlation coefficient.
Discussion
This first experiment demonstrates that although DPOAEs are not more sensitive than AEP in the rat,
they could be used to monitor the styrene ototoxicity. Moreover, the findings suggest that normal
DPOAEs may not guarantee normal cochlear status and, therefore, results of DPOAE measurements
should be interpreted cautiously. Because the ratio PTS/∆DPOAE could help determine the nature of
the contributor of hearing losses and DPOAEs have non-invasive and objective characteristics, their
use as diagnostic audiometric tool has implications for workplace safety.
If DPOAEs are not more sensitive than AEPs, they might vary earlier than AEPs. Such an
assumption deserve to be tested in a near future to close definitively the investigations on the right
indicators to monitor the ototoxicity of styrene.
Policy related benefits
This experiment shows that DPOAEs, at low-stimulus levels, can be good predictors of the auditory
thresholds. DPOAEs could therefore be used to monitor the ototoxicity of styrene.
It is likely that the use of DPOAEs could be generalized to the whole family of the aromatic solvents.
The use of both audiometric techniques and the determination of the ratio PTS/∆DPOAEs could be
very helpful to determine the origin of cochlear trauma: mechanical or chemical process.
Literature cited
Campo P., Lataye R., Cossec B., Placidi V. (1997) Toluene-induced hearing loss: a mid-frequency
location of the cochlear lesions. Neurotox. & Teratology 19(2) 129-140.
Pouyatos B., Campo P., Lataye R. (2002) Use of DPOAEs for assessing hearing loss caused by
styrene in the rat. Hear. Res. 165 : 165-164.
38
Chapter 2 : Is there a difference of susceptibility to noise- and solvent-induced hearing loss
between young and aged subjects ?
The number of aged people has increased appreciably the last decade in developed countries and the
tendency is expected to continue into the next century. This demographic change has profound effects
on social and health care systems for elderly people.
One of the most prevalent functional implications of ageing is deterioration of hearing. The hearing
loss which is related to ageing is called age-related hearing loss or presbycusis (Schuknecht, 1993).
Because aged workers are often considered as a population with an increased risk of worked-related
injuries, the ageing effects on the peripheral and central auditory systems have been studied with
animals. To the best of our knowledge, the interaction of age with solvent on hearing had never been
studied. Epidemiological studies have shown that chronic exposure to styrene can cause auditory
deficits in workers. Similarly, numerous animal experiments have shown that styrene can severely
disrupt the auditory function (Morata and Campo, 2001).
Because of the lack of data in the literature regarding age-styrene interactions, the main goal of
the present investigation was to compare the noise effects with those of styrene on hearing in
young adult and aged populations of Long-Evans rats.
Subjects
Three months old and 25 months old rats were respectively exposed either to noise or styrene.
Hearing testing
Auditory function was tested by recording the auditory-evoked potentials from the inferior colliculus
[Figure 1] as described in the first part of the present report.
Protocol
The animals were exposed for 6 hours per day, 5 days per week for 4 weeks either to a broad –band
noise centered at 8 kHz with an intensity of 92 or 97dB SPL [Figure 5a,b], or to 700 ppm styrene
[Figure 5c].
As previously [Figure 2] an audiogram was obtained prior to the exposure (T1), at the end of the
exposure (T2), and six weeks post-exposure (T3).
Compound and permanent threshold shifts were respectively defined as :
CTS = T2-T1 and PTS = T3-T1.
Exploitation and dissemination of results
Noise-induced hearing loss
Figure 5a shows significant CTS and PTS within the 92 dB SPL noise-exposed group compared to the
age-matched controls. The threshold shifts were located in a frequency range from 8 to 16 kHz. The
maximum hearing loss occurred between 10-12 kHz which corresponds to approximately one-half
octave (11.3 kHz) above 8 kHz: the central frequency of the exposure.
At 92 dB SPL intensity, there were no differences between the young and aged groups in terms of
CTS or PTS.
Conversely, at 97 dB SPL, both the CTS and PTS were significant from 8 kHz up to 32 kHz regardless
of the age of the animals (Figure 5b).
June 2004
NoiseChem 40
Figure 5: Compound (CTS) and permanent threshold shift (PTS) versus frequencies, ranging from 2 to
32 kHz, obtained from noise-exposed rats (n=13 exposed and control young rats, n=14 exposed and
control aged rats). The noise was an octave band noise centered at 8kHz at (a) 92 dB or (b) 97 dB
SPL, 6h/d, 5d/w, 4w.
40
Styrene-induced hearing loss :
(c)
Figure 5c: CTS and PTS versus frequency obtained from styrene-exposed rat. Styrene exposure :
700ppm, 6h/d, 5d/w, 4w.
Figure 5c shows CTS and PTS as a result of styrene exposure for young and aged rats. Only the young
styrene-exposed rats show significant PTS. A 15 dB styrene-induced hearing loss was located in the
region of 16-20 kHz. Surprisingly, no difference was obtained between controls and aged styrenetreated rats. No significant recovery could be measured from the CTS induced by styrene at the end of
the six-week post-exposure period.
Discussion
Noise effects
As expected from the literature, there was a strong causal relationship between noise exposure and
hearing loss in this study. Indeed, the electrophysiological data showed that 92 and 97 dB SPL octave
band noises centered at 8 kHz clearly caused a permanent hearing deficit (Figure 5a-b) whose
maximum amplitude was positioned half an octave (10-12 kHz) above the exposure frequency.
If aged rats were equally vulnerable to 92-dB noise, 97-dB noise was more damaging in aged than in
young rats, specifically in the high frequency region.
Styrene effects
To the best of our knowledge, no data had been published on styrene effects as a function of age of the
exposed subjects. Based on the electrophysiological results reported in the present study, aged rats
were insensitive to 700 ppm styrene compared to young animals contrary to all expectations (Figure
5c).
As previously reported, styrene-induced hearing loss was located in the mid-frequency range, namely
in the vicinity of the 16-20 kHz area. For young animals, a 15-dB PTS was observed in conjunction
with large hair cell losses.
The aged animals showed no significant threshold shift due to the styrene exposure although hair cell
losses (22.3 %) occurred in the third row of OHCs. Such a loss of OHCs was modest with regard to
that stated in the young rats, and apparently insufficient to modify the functional results. This
statement associated with others in previous experiments in which the third row of OHCs was
damaged without causing any threshold shift, suggests that the third row of OHC does not seem to
influence auditory sensitivity in the rat.
Noise versus styrene effects: a comparison
NoiseChem 42
June 2004
Based on the histological data, noise caused less hair cell loss than styrene, although the magnitudes of
PTSs were greater following the noise exposure than those following the styrene exposure. These
results can easily be explained by the respective mechanisms responsible for the noise- or styreneinduced hearing losses.
Figure 6: Scanning electron micrograph of the OC from a rat exposed to OBN:8kHz at 97 dB. Upper
panels correspond to control stereocilia, whereas bottom panel correspond to splayed and broken
stereocilia observed following the noise exposure.
Control inner hair
cell
Inner hair
cell
Outer
hair cell
Noise-induced hearing loss can be due to a stereocilia pathology, but styrene-induced hearing loss
would be caused by a poisoning of the organ of Corti which would disorganize the membranous
structures of the outer hair cells (Campo et al., 2001).
The effects of styrene are therefore different from that of noise and are mainly caused by a chemical
process. When the cells are present they look in a standard shape.
Figure 7 : Styrene-induced hearing loss could be summarized by a poisoning effect
Scanning electron microscopy of
a control organ of Corti
Scanning electron microscopy of
a styrene-exposed organ of Corti
42
June 2004
NoiseChem 43
The difference of vulnerability between young and aged subjects
The concept of reduced metabolic activity is often mentioned to explain the differences in sensitivity
between aged and young subjects. If the decrease in metabolic activity would be the reason of the
difference of vulnerability between young and aged subjects, the aged ear should be more sensitive to
trauma from environmental insults, including noise and drug stress. In fact, if the concept of reduced
metabolic activity can explain the noise effects, it could not explain the styrene effects since the aged
ears were less sensitive to styrene than the young ear. The difference in sensitivity to styrene between
groups could be due to either a weight difference between animals, or to a maturity difference. Indeed,
the aged rats weighted more than the young rats (500 g vs. 300 g) at the beginning of exposure. The
maturity of the cochlea could also play a major role in the sensitivity between the age groups. The
existence of a critical period for styrene-induced hearing and hair cell loss will also be examined in
future studies.
Policy related benefits
Age is a parameter which should be taken into consideration for the risk assessment concerning people
exposed to noise. Indeed, the aged ear should be more sensitive to trauma from environmental insults,
Literature cited
Campo P., Pouyatos B., Lataye R. (2003) Is the aged rat ear more susceptible to noise or styrene
damage than the young ear ? Noise & Health 5 :1-19
Morata T, Campo P. (2001) Auditory function after single or combined exposure to styrene : a review.
Noise-induced hearing loss/ Basic mechanisms, prevention and control. Eds nRn publications,
London, 293-304.
Schuknecht H., Gacek M. (1993) Cochlear pathology in presbycusis. Ann. Otol. Rhinol. Laryngol.
102: 1-16.
The results of the study carried out with young and aged rats lead to an additional question :
43
Chapter 3: Could the plasticity of the central nervous system (CNS) explain the non functional
effects of small cochlear trauma?
The main goal of the present investigation was to compare the noise effects with those of styrene on
GABAergic neurotransmission in the inferior colliculus [IC] of a population of young-adult Long–
Evans rats. The purpose which motivated this experiment was to know whether the CNS was capable
of modulating the inhibitor circuits to enhance the inputs coming from the cochlea. More accurately,
the main goal of the study was to determine whether styrene-induced OHC loss could also be
counterbalanced by a modification of GABAergic mechanisms in the IC.
In the present study, inhibition changes were assessed by estimating the 67-kDa isoform of the
glutamate decarboxylase (GAD67), the synthesizing enzyme of GABA. GAD67 levels are associated
with neuronal metabolism through the GABA shunt, and may be sensitive to long-term changes in
GABA levels due to different physiological conditions. Since the GABA is a general inhibitor
throughout the CNS, the GAD concentration is a good indicator to evaluate the inhibition performance
in young and aged subjects
Subjects
A total of 29 male rats was used in this investigation. The rats were 7 weeks old and their weight
ranged between 180 and 200 g when they arrived in the animal facility.
Hearing testing
Auditory function was tested by recording the near field auditory evoked potentials from the IC.
Exposure conditions
Subjects were exposed (6h/d, 5d/w for 4w) to either a 97-dB SPL octave band noise centered at 8 kHz,
or to 700 ppm styrene.
Changes in neuronal processing in the inferior colliculus
The dosage by indirect competitive enzyme-linked immunosorbent assay (ELISA) of the glutamate
decarboxylase (GAD) in the IC from young and aged rats has been developed to evaluate the plasticity
of the CNS.
The dosage is detailed in Pouyatos et al. (2004)
Exploitation and dissemination of results
in fe rio r co llicu lu s G AD67 co n ce n tratio n s
1 6 ,0
1 4 ,0
1 2 ,0
6 ,0
4 ,0
2 ,0
n=8
8 ,0
n=5
*
1 0 ,0
n=6
[GAD67] µg / g
Figure
8:
GAD67
concentrations in the
inferior colliculus of
controls,
97dB
SPL
noise-exposed and 700
ppm
styrene-exposed
rats. The bars represent
the standard deviation.
0 ,0
Controls
9 7 dB SPL
70 0 ppm
June 2004
NoiseChem 45
Discussion
The main findings of the present study support the hypothesis that the IC undergoes significant
changes in GABA neurotransmitter function following acoustic trauma. A significant decrease of
GAD67 levels (-37 %) can be observed 6 weeks following 97dB-noise-exposure. These results are
consistent with those obtained by Abbott and coworkers (1999) in rats exposed for 9 h to a 100-dB
continuous tone. Indeed, the authors measured by Western blotting a 39 % decrease of GAD67 levels
30 days post-acoustic exposure. Thus, the ELISA technique would be as sensitive as the Western
blotting. However, the main disadvantage of the two former techniques is that they provide no idea of
the localization of the antigen in the nucleus of interest, by contrast to immunostaining on tissue slices
for instance. The finding of a reduced concentration of GAD is consistent with a growing number of
studies which suggest that inhibitory neurotransmission may be reduced in the auditory brainstem
when peripheral excitatory drive has been compromised. Bledsoe et al. (1995) found a decreased
number of GABAergic neurons in the IC 21 days after cochlear ablation. Interestingly, Milbrandt et al.
(2000), showed a 41 % decrease of GAD65 levels in IC after a 10h-exposure to 106 dB SPL (12 kHz),
with a complete recovery at 30 days post-exposure. This last result supports the hypothesis that
GAD65 is present in the neuron as an inactive reservoir and is likely to be responsive to short-term
changes in requirements for GABA. On the other hand, GAD67 may preferentially contribute to the
GABA shunt, and would be subject to long-term physiological regulation (Feldblum et al., 1995). The
decrease of the GAD67 concentration one month post-exposure may be indicative of the establishment
of a new equilibrium following a deafferentation. Along with these neurochemical studies, many
functional results support the hypothesis of weaker inhibition in the IC following a noise-exposure.
For instance, Szczepaniak and Møller (1996) observed a decrease of the GABA neurotransmission
after a noise exposure, and Salvi et al. (1990) obtained enhanced evoked responses in the rat and the
chinchilla following an acoustic trauma. Even though the effects of noise on the inhibitory
neurotransmission have been extensively studied, to the best of our knowledge no studies have
focused on the effects of solvents on the central auditory system. Our results suggest that, contrary to
noise, styrene does not induce a modification of GAD67 concentration in the IC. This was unexpected
since styrene caused massive hair cell loss particularly in the 3rd row of OHC. It would have been
reasonable to think that such hair cell damage would have caused an inhibition to counterbalance the
subsequent loss of input. But obviously, it is not the case. So, how can we explain that noise-induced
cochlear damages can be compensated by the release of central inhibition, but not styrene-induced
injury ?
This discrepancy may be explained by the differences between the damaging effects of noise and
styrene on both the CNS and the organ of Corti. We cannot exclude the possibility that styrene could
directly affect IC function. However, we know also that styrene specifically damages OHCs, while
loud noises mainly cause a mechanical impairment of both OHC and IHC stereocilia. If the effect of
noise can be considered as a deafferentation, the effect of styrene may only result in minor loss of
afferent inputs since OHCs are only connected to 5 % of the afferent fibers. Based on a previous study
carried out in identical experimental conditions (Lataye et al., 2001), the dose used in the present study
was too low to cause a significant lost of spiral ganglion cells. That suggests that a central
compensation of peripheral damage is only achievable when most of the afferent fibers are impaired,
which is the case after acoustic trauma.
Policy related benefits
Central compensation for cochlear damage can preferably occur when afferent fibers are altered.
Therefore, central compensation for cochlear damage depends on the nature of the ototoxic agent. It is
reasonable to assume that noise and only noise causes a modification of inhibitory neurotransmission
within the inferior colliculus (third auditory nucleus) because of impairment of afferent supply to the
auditory brainstem.
Such findings prove that even though the audition function seems to be intact, the cochlea can be
injured. Because the central nervous system compensates for reasonable noise-induced hearing loss,
the trauma may not be identified. The risk for the injured people would be to suffer from earlier
presbycusis.
Literature cited
45
June 2004
NoiseChem 46
Abbott S., Hugues L.F., Bauer C.A., Salvi R., Caspary D. (1999) Detection of glutamate
decarboxylase isoforms in rat inferior colliculus following acoustic exposure. Neurosciences, 93(4):
1375-1381
Bledsoe SC., Nagase S., Miller JM., Altschuler RA. (1995) Deafness-induced plasticity in the mature
central auditory system. Neuroreport. 7(1), 225-229.
Milbrandt JC., Holde TM., Wilson MC., Salvi RJ., Caspary DM. (2000) GAD levels and muscimol
binding in rat inferior colliculus following acoustic trauma. Hear Res. Sep;147(1-2):251-60.
Feldblum S., Dumoulin A., Anoal M., Sandillon F., Privat A. (1995) Comparative distribution of
GAD65 and GAD67 mRNAs and proteins in the rat spinal cord supports a differential regulation of
these two glutamate decarboxylases in vivo. J. Neurosci. Res. 42(6), 742-757.
Szczepaniak WS., Moller AR. (1996) Evidence of neuronal plasticity within the inferior colliculus
after noise exposure: a study of evoked potentials in the rat. Electroencephalogr Clin Neurophysiol.
100(2):158-64.
Salvi RJ., Wang J. (1997) Evidence for rapid functional reorganization in inferior colliculus and
cochlear nucleus. In Acoustical Signal processing in the central auditory system. ed. Syka, J., Plenum,
NY pp 477-488.
Lataye R., Campo P., Barthelemy C., Loquet G., Bonnet P. (2001). Cochlear pathology induced by
styrene. Neurotoxicol Teratol. 23(1), 71-79.
Pouyatos B., Morel G., Lambert AM., Maguin K., Campo P. (2004) Consequences of noise- or
styrene-induced cochlear damages on glutamate decarboxylase levels in the rat inferior colliculus.
Hear. Res. 189 : 83-91.
46
NoiseChem 47
June 2004
Chapter 4 : Use of experimental data obtained with animals for establishing new limit exposure
values.
In order to test the noise exposure standards established for protecting people against noise-induced
hearing loss, we wanted to know whether or not a Lex,d of 85dB was pertinent when the noise was
combined with a styrene exposure. Although two animal investigations had recently studied the
combined effects of noise and styrene (Lataye et al., 2000; Mäkitie et al., 2003), the noise intensities in
both studies were high (an octave band noise centered at 8 kHz emitted at 97 dB SPL in the first study,
or industrial noise emitted at 100-105 dB SPL for the second study). Since these noises had a Lex,d >
85 dB, they were not pertinent according to people in charge of the legislation, people who can
propose a reduction of the threshold limit values based on animals experiments. With the purpose of
being more convincing and closer to the industrial conditions, the intensity of the chosen noise was
moderate : 86.2 dB for 6h/d, 5d/w for 4w. With such a schedule, the Lex,8h was equal to 85dB.
The styrene exposure had to be also relatively low , inferior to 500 ppm. As for the noise, the duration
was 6 hours per day, 5 days per week for 4 weeks.
Although it is technically simpler to use non-performing animals for carrying out a doses-effects
study, such conditions would have underestimated the solvent uptake of these animals. Indeed,
performing animals have a higher pulmonary ventilation and cardiac output compared to nonperforming animals. In fact, performing animals have a solvent uptake which would be closer to that
of workers exposed in their work environment.
For all these reasons, the main originality of the present experiment was to make the animals work
during the exposure period. To fulfill this condition, a home-made running wheel was carried out :
Figure 9
.
With this silent running wheel placed inside the inhalation chamber, we were capable of making rats
work during either noise (86.2dB), styrene (300-600ppm) or noise and styrene exposures.
The noise intensity (86.2 dB SPL) was chosen for testing the applicability of the Lex,8h =85dB for
estimating the hazard of noise regardless of the presence of other ototoxic agent.
As far as the styrene exposure was concerned, we knew that the threshold limit value of styrene
authorized in France for work environments is 50 ppm averaged over a 8 hours work day.
People in charge of the establishment of damage-risk criteria are used to take a safety factor for setting
permissible human exposures based upon laboratory animal data (c.f. Slikker et al., 1996). It is fairly
common to adopt a safety factor of 10.The size of the safety factor is justified by species differences,
extrapolation from sub-chronic to chronic exposures, and increased sensitivity of particular groups
within the population such as the elderly or very young.
47
NoiseChem 48
June 2004
Based on this principle, we tested concentrations ranging from 300 to 600 ppm of styrene. One single
concentration was above the adjusted threshold limit value [TLV] (60 ppm X 10), one concentration
was right at the adjusted TLV (50 ppm X 10) and two others were below the TLV (30 and 40ppm
X10).
Subjects
A total of 42 male rats was used in this investigation. 20 rats ware used for the doses-effects study and
22 for the combined exposure. The rats were 7 weeks old and their weight ranged between 180 and
200 g when they arrived in the animal facility.
Hearing testing
Auditory function was tested by recording the near field auditory evoked potentials from the inferior
colliculus [Figure 1 for details].
Exposure conditions
The styrene group of rats was exposed 6 hours per day, 5days per week for 4weeks to varying
concentrations of styrene ranging from 300 to 600 ppm.
The noise group was exposed to an octave band noise centered at 8 kHz emitted at 86.2 dB SPL during
the same schedule [Figure10].
OBN: 8kHz; 86.2 dB SPL
Figure 10: Relative amplitude of the spectrum of the noise (OBN: 8kHz; 86.2 dB SPL) and the ambient
noise
The combined group was exposed to both noise and styrene.
Statistical analysis
The auditory effects as a function of the styrene concentrations were testing by running a 2-way
ANOVA for which frequency was a within-subject and treatment a between-subject factor. The
interaction between noise and styrene was tested by running a linear model. The fixed effects studied
in the present investigation were those caused by either the styrene or the noise exposure or their
interaction.
Exploitation and dissemination of results
48
NoiseChem 49
June 2004
Electrophysiology: doses-effects study
(Figure 11): Styrene-induced hearing loss [SIHL] following exposure to varying concentrations.
Exposure period was 6h/d, 5d/w, 4w
A
Performing animals
B Sedentary animals (Loquet et al., 1999)
Figure 11 shows the PTS values obtained post-exposure. The threshold shifts increased significantly
[F(4,44)=16.94, p<0,001] as a function of the solvent concentrations. The comparison of the values
reported in the two graphs showed that the ototoxic potency of styrene depends on activity of the
animals. This is a determinant factor to take into consideration specially for the establishment of new
limit exposure values.
Base on these two graphs, we wanted to establish a predicting model yielding to the same amount of
styrene-induced hearing loss and depending on physical activity. To this end, we looked only at the
main injured frequencies: 10, 12, 16 and 24 kHz.
The idea was to evaluate how much we could decrease the solvent concentration to have the same
amount of hearing loss when animals are in the running wheel or sedentary (Loquet et al., 1999).
A
B
Figure 12: Predicting model that yields to the same SIHL depending on physical activity
By example, an average of 470ppm calculated from the 4 main injured frequencies was required to
obtain a 10 dB - SIHL (Figure 12A) when the rats were active, whereas 750ppm was needed to obtain
the same amount of PTS (Figure 12B) when the rats were sedentary. In other terms, performing rats
exposed to 470-ppm styrene have the same amount of SIHL that sedentary rats exposed to 750ppmstyrene.
Based on the abovementioned principle, we completed Table 2 of predicted values.
49
June 2004
NoiseChem 50
Doses-effects study : histology
Figure 13A: Performing rats were exposed to
styrene 6 h/d, 5 d/w for 4weeks.
Figure 13B: Sedentary animals were exposed
to styrene for 6 h/d, 5 d/w for 4w.
Figure 13 : Average cochleogram (n=4) Abscissa – upper trace : length (mm) of the entire spiral
course of the organ of Corti from the bottom to the hook.-Abscissa - lower trace : Frequency-map.
Ordinate: hair cell loss in percent. IHC: inner hair cells; OHC1: first row of outer hair cells, OHC2:
second row; OHC3: Third row.
50
As previously for the electrophysiological data, we wanted to establish a predicting model. This time,
the predicting model was established from the outer hair cell losses observed both in performing and
sedentary rats. Interestingly, the histological data obtained in Table 3 led to an average (223 ppm)
close to that obtained with electrophysiological data (213ppm).
Electrophysiology: combined exposure to noise and styrene :
Figure 14: Permanent threshold shift (PTS) versus frequency obtained after either 400ppm styrene
exposure, or (OBN: 8kHz at 86.2 dB SPL) or simultaneous noise and styrene. Exposure duration was
6h/day, 5 days/week, 4 weeks. The bars represent the 95% SD.
If there is a significant effect of the treatment [Ftreatment (3,33)=7.08, p = 0.0001], there is no significant
difference between the noise and the noise+styrene groups (post hoc test).
Histology : combined exposure to noise and styrene :
A
B
Figure 15: (A) Average cochleogram obtained from rats exposed to [OBN: 8kHz; 86.2 dB SPL]; (B):
the yellow bars correspond to hair cell losses at 400ppm ; the dark bars to a simultaneous exposure to
noise and styrene (respectively OBN: 8kHz; 86.2 dB SPL + 400ppm). Exposure schedule : 6h/d, 5d/w,
4w. Abscissa – upper trace :length (mm) of the entire spiral course of the organ of Corti from the
bottom to the hook.-lower trace :frequency-map. Ordinate: hair cell loss in percent. IHC: inner hair
cells; OHC1: first row of outer hair cells, OHC2: second row; OHC3: Third row.
Contrary to electrophysiological data, histological data show that there is a clear increase in outer hair
cell losses, particularly at the level of the second and third row. The difference of outer hair cell losses
was not sufficient to cause a significant increase of the PTS amplitudes measured with our
electrophysiological technique.
Discussion
It is clear that the uptake, via the lungs, of a gaseous solvent with a given concentration in ambient air
depends on the activity of the animals. If people in charge of the establishment want to compare the
concentration targeted in the animal investigations with that in the working environment, they have to
consider whether or not the animals are in activity. Most of the time, animals are sedentary when they
are exposed. As a result, a right estimation of the risks encountered by the people in their working
environment would be to decrease by approximately 200ppm the concentration mentioned by the
authors in the experimental protocol.
During the exposure with both agents (noise and styrene), the animals were exposed to a maximum
concentration which was not superior to 10 times the French recommended values. Indeed, the
recommended value in France is 50 ppm, and the concentration in the inhalation chamber was 400
ppm, [ 40ppm x 10 (safety factor)].
During the period of inhalation at a specific concentration (10 times the standard value maximum), the
animals have been exposed simultaneously to a low level noise exposure (86.2dB) which corresponds
to a Lex8h=85dB.
The results and more specifically the histological data show a risk of potentiation of noise-induced
hearing loss by styrene, even at a concentration as low as (40 X 10 = 400 ppm).
How might such exposure conditions relate to human exposure and human health?
When modified to account for possible greater susceptibility of people compared with rats, we
obtained a reference dose of 40 ppm. Thus, using currently accepted standards for assessing risk,
French workers could be exposed at or above the reference dose required to maintain safety from
potential of NIHL by styrene.
Policy related benefits
In France, the threshold limit value of styrene authorized in working environments is 50 ppm averaged
over a 8 hours work day. 30 ppm should be safer.
Publications in progress
Lataye R;, Campo P., Morel G. 2004 Fund. Appl. Toxicol.
See publications Chapter 4
Publications
Chapter 1
(1) Campo P., Lataye R., Cossec B., Placidi V. (1997) Toluene-induced hearing loss: a midfrequency location of the cochlear lesions. Neurotox. & Teratology 19(2) 129-140.
(2) Pouyatos B., Campo P., Lataye R. (2002) Use of DPOAEs for assessing hearing loss
caused by styrene in the rat. Hear. Res. 165 : 165-164.
Chapter 2
(1) Campo P., Pouyatos B., Lataye R., Morel G. (2003) Is the aged rat ear more susceptible to
noise or styrene damage than the young ear ? Noise & Health 5 :1-19
(2) Morata T, Campo P (2001) Auditory function after single or combined exposure to
styrene : a review. Noise-induced hearing loss/ Basic mechanisms, prevention and control.
Prasher D., Henderson D., Kopke R., Salvi R., Hamernik E., Eds nRn publications,
London, 293-304.
(3) Schuknecht H., Gacek M. (1993) Cochlear pathology in presbycusis. Ann. Otol. Rhinol.
Laryngol. 102 : 1-16.
Chapter 3
1. Abbott S., Hugues L.F., Bauer C.A., Salvi R., Caspary D. (1999) Detection of
glutamate decarboxylase isoforms in rat inferior colliculus following acoustic
exposure. Neurosciences, 93(4): 1375-1381
2. Bledsoe SC., Nagase S., Miller JM., Altschuler RA. (1995) Deafness-induced
plasticity in the mature central auditory system. Neuroreport. 7(1), 225-229.
3. Milbrandt JC., Holde TM., Wilson MC., Salvi RJ., Caspary DM. (2000) GAD levels
and muscimol binding in rat inferior colliculus following acoustic trauma. Hear Res.
Sep;147(1-2):251-60.
4. Feldblum S., Dumoulin A., Anoal M., Sandillon F., Privat A. (1995) Comparative
distribution of GAD65 and GAD67 mRNAs and proteins in the rat spinal cord supports
a differential regulation of these two glutamate decarboxylases in vivo. J. Neurosci.
Res. 42(6), 742-757.
5. Szczepaniak WS., Moller AR. (1996) Evidence of neuronal plasticity within the
inferior colliculus after noise exposure: a study of evoked potentials in the rat.
Electroencephalogr Clin Neurophysiol. 100(2):158-64.
6. Salvi RJ., Wang J. (1997) Evidence for rapid functional reorganization in inferior
colliculus and cochlear nucleus. In Acoustical Signal processing in the central
auditory system. ed. Syka, J., Plenum, NY pp 477-488.
7. Lataye R., Campo P., Barthelemy C., Loquet G., Bonnet P. (2001). Cochlear
pathology induced by styrene. Neurotoxicol Teratol. 23(1), 71-79.
8. Pouyatos B., Morel G., Lambert AM., Maguin K., Campo P. (2004) Consequences of
noise- or styrene-induced cochlear damages on glutamate decarboxylase levels in the
rat inferior colliculus. Hear. Res. 189 : 83-91.
June 2004
NoiseChem 54
Chapter 4
1. Lataye R., Campo P., Loquet G. (2000) Combined effects of noise and styrene exposure on
hearing function in the rat. Hearing Research 139 : 86-96.
2. Mäkitie AA., Pyykkö I., Sakakibara H., Riihimäki V., Ylikoski J. (2003) The ototoxic
interaction of styrene and noise. Hearing Research 179 : 9-20.
3. Loquet G., Campo P., Lataye R. (1999) Comparison of toluene-induced and styrene-induced
heraing loss. Neurotox.& Teratol. 21:6, 689-697.
4. Slikker W., Crump KS., Anderson ME., Bellinger D. (1996) Biologically based, quantitative
risk assessment of neurotoxicants. Fund Apll. Toxicol. 29, 18-30.
54
June 2004
NoiseChem 55
Animal Lab 2: Søren P. Lund
National Institute of Occupational Health, Denmark
55
June 2004
NoiseChem 56
Studies on the auditory effects of combined exposures to noise, toluene, and carbon
monoxide in rats.
Søren P. Lund and Gitte B. Kristiansen
National Institute of Occupational Health, Lersø Parkallé 105, DK-2100 Copenhagen, Denmark
Abstract
The present study has focused on the possible consequences of long-term exposure, and the
qualities of the noise being among the most important factors in causing the effects of interaction
of combined exposures to noise and organic solvents like toluene. Further investigation has been
directed towards the inclusion of additional risk factors in order to evaluate possibilities of
potentiation of noise induced hearing loss from combined exposures. All the studies were
performed with toluene as the ototoxic organic solvent. Prolonged exposures to steady state, wide
band noise and toluene did not increase the risk of synergetic interactions, unless the exposure
concentration of toluene was close to the low observed adverse effect level (LOAEL). An
exposure schedule including short periods of higher noise levels did not increase the interactions
between noise and toluene exposures. Impulse noise induces caused considerable more
impairment to hearing than wide band noise, and the effects of interaction was found to
proportional to the greater auditory impairment of the impulse noise without exposure to toluene.
Considerable hearing loss was found from exposure to impulsive type of noise at 84 dB SPL,
with 75% of the energy as noise impulses. No direct synergetic interaction with toluene exposures
was noted at this level of noise exposure, but a potentiation of the effects of exposure to carbon
monoxide (CO) was evident, even at the lowest exposure levels of toluene. Combined exposure
above the LOAEL of CO potentiation of the noise exposure increased LOAEL of toluene by
50%. Extrapolation of this result to the human working environment would imply that combined
exposure to noise and organic solvents could increase the risk of auditory hearing impairment of
tobacco smokers significantly. Further, if these findings can be generalized to other risk factors
for hearing impairment, a major part of the hearing impairment by occupational exposures may
be caused by interactions of low to medium levels noise exposure in combination with individual
risk factors of hearing loss.
Keywords
Hearing; Wide band noise; Impulse noise, Interaction; Organic solvents, Toluene, Rat; ABR,
DPOAE.
Introduction
The possible ototoxic effect of toluene was first reported in rats by Pryor et al. (1983), and
subsequent studies demonstrated hearing loss in rats after exposure to toluene (Rebert et al.,
1983; Pryor et al., 1984a; Pryor et al., 1984b; Johnson et al., 1988; Sullivan et al., 1989; Pryor et
al., 1991; Crofton et al., 1994; Rebert et al., 1995; Campo et al., 1997; Lataye & Campo, 1997),
styrene (Pryor et al., 1987; Yano et al., 1992; Crofton et al., 1994, 2000, Loquet et al., 1999,
Lataye et al., 2000, Campo et al., 2001, Pouyatos et al., 2002, Mäkitie et al., 2002 ), xylene
(Pryor et al., 1987; Crofton et al., 1994; Rebert et al., 1995), ethyl benzene (Cappaert et al., 1999,
Cappaert et al., 2000), trichloroethylene (Rebert et al., 1991; Crofton and Zhao, 1993; Jaspers et
al., 1993; Crofton et al., 1994; Rebert et al., 1995, Muijser et al., 2000), chlorobenzene (Rebert et
al., 1995) , and n-heptane (Simonsen and Lund, 1995). The ototoxic potency of the different
solvents varies significantly (Rebert et al., 1995, Loquet et al., 1999), but when rats were exposed
to combinations of two different organic solvents, the auditory impairment after combined
exposure was additive with respect to equal potency of the solvents under study (Rebert et al.,
1995).
In rats, the general auditory impairment from exposures to organic solvents exposure is a midfrequency hearing loss at 8-20 kHz following a loss of outer hair cells (OHC) in the middle and
56
June 2004
NoiseChem 57
basal turns of the cochlea (Pryor et al., 1984a; Sullivan et al., 1989; Yano et al., 1992; Jaspers et
al., 1993; Crofton et al., 1994; Johnson and Canlon, 1994a; Johnson and Canlon, 1994b; Campo
et al., 1997; Lataye and Campo, 1997). At low exposure concentration there is only smaller
changes at frequencies close to16 kHz, but with increasing exposure levels, the changes spreads
upwards as well as do downwards in frequency domains. However, a certain threshold level of
toluene has to be exceeded before even prolonged exposure induces signs of ototoxicity (Pryor et
al., 1984b; Nylén et al., 1987; Pryor et al., 1991; Jaspers et al., 1993). Corresponding to threshold
in exposure concentration, there seems to be threshold concentration of toluene in blood of 40-60
µg/ml before any auditory impairment is evident, and it seems to be toluene itself rather than a
metabolite that is responsible for the loss of OHC (Pryor et al.; 1991).
At present, the risk of human auditory impairment due to organic solvent exposure seems
primarily to be a problem concerning the effect of interaction of combined exposure to solvents
and noise. A growing number of studies have examined the nature of the interaction between
exposure to noise and organic solvents on hearing in animal experiments. A potentiation of the
auditory impairment was found in rats exposed sequentially to toluene (1000 ppm 16 hours/day, 5
days/week for 2 weeks) and noise (100 dB Leq 10 hours/day, 7 days/week for 4 weeks) when the
toluene exposure preceded the noise exposure (Johnson et al., 1988). The reverse exposure order
resulted only in an additive effect (Johnson et al., 1990). Synergistic interaction was
demonstrated in rats following simultaneous exposure to 2000 ppm toluene and noise (92 dB
SPL) for 6 hours a day, 5 days a week for 4 weeks (Lataye & Campo, 1997), as well as so to 1500
ppm toluene and noise (96 dB SPL), 6 hours/day for 10 days (Brandt-Lassen et al., 2000).
However, the synergistic interaction in both these studies were found only when the exposure to
toluene exposure caused considerable auditory impairment without concomitant exposure to
noise. A similar pattern of synergetic interaction has also been demonstrated on rats in combined
exposure to ethyl benzene and noise (Cappaert et al., 2001) as well as styrene and noise (Lataye
et al., 2000; Mäkitie et al., 2003). Of particular interest are studies in animal models of the
possible interaction of low-level exposure to organic solvents and noise, where the exposure to
each factor alone is without any effect. At present, no studies has shown synergetic effects
interaction from combined exposures to organic solvents and noise in animal studies, unless the
level of exposure was above or at least very close to the low adverse effects levels (LOAEL) of
the organic solvent without concomitant noise exposure. A possible reason for this is that the
length of the combined exposures has been to short to provide the data for the evaluation of the
hazards of human long-term exposures in the work environment (Cappaert et al., 2001).
The susceptibility to the ototoxic properties of organic solvents is species specific, and rats and
mice are appears to be very sensitive, while chinchillas and guinea pigs are not (McWilliams et
al., 2000; Davis et al., 2002; Cappeart et al., 2002; Lataye et al., 2003). So far, the rat has been
the preferred animal model for studying the risk from combined exposure to noise and organic
solvents. Other factors have been included in order to provide insight into the human risk of
hearing impairment from the combination of the chemical, physical and intrinsic risk factors.
Differences in ototoxicity of styrene between rats of 14 and 21 weeks of age has been found, the
young being the most sensitive to styrene ototoxicity, and possible due a changes metabolic rate
of styrene with age (Campo et al., 2003). Ethanol has been found to potentiate the ototoxicity of
styrene, either by inducing intrinsic changes in the cochlea, or by modifying the styrene
metabolism (Campo et al. 1998; Loquet et al., 2000).
The present study has focused on the possible consequences of long-term exposure, and the
qualities of the noise being among the most important factors in causing the effects of interaction
of combined exposures to noise and toluene. Further investigation has been directed towards the
57
June 2004
NoiseChem 58
inclusion of additional risk factors in order to evaluate possibilities of potentiation of noise
induced hearing loss from combined exposures.
Materials and methods
The study schedule has three main parts, each covering different aspects of the interaction
between the exposure to organic solvents and noise. Along with the change in topics within the
study, the methods have been changed in order to adapt to the specific requirements associated
with each topic. The methods of audiological testing of the animals have also been improved
during the course of the study, and especially the testing of the complete frequency range of
hearing in the rats is considered to be a major improvement. However, the description of methods
used will be divided into sections only whenever needed.
The animal welfare committee, appointed by the Danish Ministry of Justice, has granted ethical
permission for the studies. All procedures were carried out in compliance with the EC Directive
86/609/EEC and with the Danish law regulating experiments on animals.
Schedule
In all the three studies toluene was chosen as an example of an ototoxic organic solvent. Except
for the 90-day study, all rats were exposed 6 hours/day for a period of 10 days toluene, carbon
monoxide (CO), and noise, whether the exposures were performed for one or more of the factors.
The 90 day study
This part of the study was made to elucidate the consequences of long-term, low-level exposure
to both toluene and noise. A 90 day exposure schedule was chosen in order to confirm with the
OECD test guidelines for a sub-chronic inhalation toxicity study. Five groups of 12 rats were
exposed to 0 ppm, 100 ppm, 200 ppm, or 500 ppm toluene 6 hours/day, 5 days/week and to
steady state 90 dB SPL, 4-20 kHz wide band noise (WBN) 4 hours/day, 5 days/week for 90 days,
and one group was exposed to neither toluene nor noise.
Toluene, WBN and impulse noise
This part of the study was made to investigate the potential difference in interaction between
organic solvents and either wide band noise (WBN) or impulse noise. Eight groups of 12 rats
were exposed to 0 ppm, 500 ppm, 1000 ppm, or 1500 ppm toluene and either 92 dB SPL WBN or
impulse noise with a frequency distribution of 4-24 kHz. The noise exposure schedule and the
energy of the noise were the same for both types of noise, but the noise level was changed
throughout the daily exposures in the same way for both types of noise. Further, on group was
exposed to 1500 ppm toluene without exposure to noise.
Toluene, CO and noise
The topic for this part of the study was performed in order to investigate simultaneous exposure
to toluene, CO and noise. The noise chosen was a mixture of impulse and WBN, with the main
energy (75%) as impulsive noise. In order to find the LOAEL between exposure to noise and CO,
two groups of 12 rats were exposed 0 ppm, and 500 ppm carbon monoxide (CO) and 82.3 dB
SPL impulsive noise with a frequency distribution of 4-20 kHz, and further 6 groups of 12 rats
was exposed to 0 ppm, 300 ppm, and 500 ppm CO and either 85.3 or 88.3 dB SPL impulsive
noise. Finally 6 groups of rats were exposed to 0 ppm, 300 ppm, or 500 ppm CO, 85.3 dB SPL
impulsive noise and either 500 ppm, 1000 ppm toluene.
Animals
Male Wistar rats (MOL:Wist Han) were purchased from a local breeder (M & B, Ltd.) to have a
weight 175-200 g at arrival. The rats were housed two by two in polypropylene cages (425 × 266
× 150 mm) with steam cleaned pinewood bedding (Lignocel S 8). Municipal water and rodent
58
June 2004
NoiseChem 59
chow (Altromin 1324) was accessible ad libitum. In the animal quarters, the temperature was
maintained at 21 ± 1 °C and humidity at 55 ± 10 %. Lights were on from 7.00 PM to 7.00 AM.
To be exposed the rats were transferred daily from their home cages to closed climate chambers
and kept two by two in wire mesh cages without access to food and drinking water. In order to
secure the most equal exposure of the rats, the location the individual rats within the chambers
were rotated one position every day.
Toluene exposure
The exposures were performed in dedicated inhalation chambers with walls of stainless steel and
glass. The air exchange rate was 12 per hour with an air temperature of 20 ± 2°C and a humidity
of 55 ± 10 % RH. During the exposures, which were performed between 8.30 AM and 2.30 PM
with the animals in their normal waking state, the rats were housed in wire mesh cages without
access to food and water. Toluene (purity >99,5% GC; CAS-No. [108-88-3]) was evaporated in
the air-inlet of each exposure chamber by means of an HPLC-pump feeding toluene to the top of
a glass spiral, which was slightly heated by circulating water (36°C). The toluene concentration
was measured with an infrared gas cell spectrophotometer (Foxboro MIRAN-1A) on one
chamber every 5 minutes automatically changing from chamber to chamber. The system was
calibrated directly in units of concentration (ppm) and all exposure data were collected on a
computer for later analysis. For control, the daily quantity of toluene used for each chamber was
measured and checked against the toluene concentration. The means and standard deviations of
the toluene exposure concentrations at steady state, which was reached approximately 20 min.
after the start of the dosage pumps, are given below.
The 90 day study
The rats were exposed to toluene, 6 hours/day, 5 days/week, for a period of 90 days. The mean
toluene concentrations measured for the groups were 102 ±1 ppm, 202 ±1 ppm, and 500 ±1 ppm
toluene, respectively.
Toluene, WBN and impulse noise
The mean toluene concentrations measured for the groups were 491 ±15 ppm, 507 ±28 ppm,
1007 ±14 ppm toluene, 1009 ±14 ppm, 1505 ±44 ppm, 1501 ±48 ppm, and 1505 ±15 pm
respectively.
Toluene, CO and noise
The mean toluene concentrations measured for the groups were 503 9 ppm, 503
10 ppm, 1008 ±45 ppm, 1008 ±101 ppm, and 1018 ±94 ppm respectively
16 ppm, 504
CO exposure
The general exposure conditions were the same as described for the toluene exposures. CO was
feed to the air inlet of the chamber, under the control of simple flow meters (Porter). The CO
concentration was measured with an infrared gas cell spectrophotometer (Foxboro MIRAN-1A)
on one chamber every 5 minutes, automatically changing from chamber to chamber. The rats
were exposed to either 0 ppm, 300 ppm or 500 ppm CO, 6 hours/day for a period of 10 days. The
concentrations measured were 302 ±34 ppm, 303 ±22 ppm, 308 ±57pm, 299 ±50 ppm, 504 ±34
ppm, 493 ±74 ppm, 506 ±74 ppm, 530 ±90 ppm, and 504 ±20 ppm respectively.
Noise exposure
The noise were generated on a computer with a 16-bit D/A-converter board, amplified by audio
amplifiers (NAD 216), and delivered by dome tweeters (Vifa D26TG-05-06) situated above each
cage. The sound field was measured at various points at the level of the floor of the cages with a
59
June 2004
NoiseChem 60
½” condenser microphone (B&K4133) and a spectrum analyser (HP35670A). The noise level
within the exposure chambers was measured to be 35 dB SPL in the 2 - 48 kHz frequency range,
with the highest single frequency contribution being 25 dB at 4.8 kHz. Except for the 90 day
study, all noise exposures were started 20 minutes after the onset of the of the toluene and CO,
and correspondingly ended 20 minutes later than chemical exposures.
The 90 day study
The rats were exposed to 90 dB SPL, 4-20 kHz WBN 4 hours/day, which corresponds to
Leq8hours= 87 dB SPL. The daily noise exposure started 2 hours later than the toluene exposure.
The frequency distribution of the measured WBN was quite uniform with in the pass band, and
the level within the sound field varied less than ± 1 dB between measuring points.
Figure 1. A noise impulse, sampled inside the inhalation chambers at the level with the bottom of
the wire mesh cages. The sampling was performed with a B&K 1/2“ condenser microphone (type
4133), a B&K preamplifier (type 2669), a B&K Nexus amplifier, and a computer with a DATA
TRANSLATION data acquisition module (DT9803) at a rate of 100 kHz.
Toluene, WBN and impulse noise
The levels of the noise was varied between 82 dB, 92 dB and 102 dB SPL by a fixed schedule of
5 min. periods, so that the rats was exposed for 5 hours (60 periods of 5 min.) to a noise level of
to 82 dB, 30 min. (6 periods) of 92 dB, and 30 min. (6 periods) of 102 dB. The resulting 4-20
kHz noise exposures had a level of 92 dB SPL, either as WBN by varying the sound level or as
impulse noise by interval between impulses with a peak level just above 130 dB. A noise impulse
is shown in Figure 1. The 6 hours noise exposure corresponds to a Leq8hours= 90.8 dB SPL. The
frequency distribution of the WBN was quite uniform between 4-24 kHz (se fig. 2), and level
within the sound field varied less than ± 1 dB between measuring points. The impulse noise had
somewhat less equal frequency distribution due to a higher energy level towards the lower
frequencies (se Figure 2), and the level varied up to ± 1.5dB between measuring points within the
sound field.
60
June 2004
NoiseChem 61
Toluene, CO and impulse noise
The rats were exposed to a mixture of impulse and WBN, consisting mainly of (75%) of the
energy as noise impulses with a peak level just above 130 dB, and the rest of the energy (25%) as
WBN in the periods between the impulses. The frequency composition of the noise in were
changed from 4-24 kHz back to the same composition as in the 90 day study, i.e. 4-20 kHz. The
noise exposure levels were 82.3 dB, 85.3 dB and 88.3 dB SPL 6 hours per day, with an interval
between the impulses of 42.8 sec, 21.5 and 10.8 sec., respectively. The 6 hours of noise exposure
corresponds to Leq8hours = 81 dB, 84 dB, and 87 dB SPL, respectively.
Figure 2. The frequency distributions of the impulse and wide band noise (WBN), used to
investigate the potential difference of the effects of interaction between exposures to either WBN
and impulse noise and simultaneous exposure to toluene. The sound levels of the two type of noise
shown in the figure are in both cases 102 dB SPL. The sampling was performed with a B&K 1/2“
condenser microphone (type 4133), a B&K preamplifier (type 2669), and a HP35670A spectrum
analyzer.
Tests of hearing
Overall, the test of hearing where made both by determination of hearing thresholds by the
auditory brain stem response (ABR) as well as by measurements of distortion products
otoacoustic emissions (DPOAE) by the same stimulus source (probe assembly) with the animals
in anaesthesia (65 mg/kg pentobarbital sodium i.p).
The ABRs were recorded with a silver wire inserted subcutaneously at the back of the head as
active electrode, a small roll of silver wire in the mouth as reference electrode and a stainless steel
needle in the tail as ground electrode. The pure tone stimuli were generated with a repetition rate
of 19.9 per sec. by a programmable function generator (Hameg 8130) as symmetrical tapered 1.4
msec tone-pips. The response was amplified 50,000 times, filtered through analogue band pass
filter (10 Hz to 10 kHz), and 15 msec were sampled at a rate of 51.2 kHz by a 16-bit data
acquisition board. The ABR of each stimulus level consisted of 256 artefact free recordings that
was averaged and stored on hard disk for later analysis. After further digitally FIR-filtering (2
kHz cut of frequency and 4 kHz stop band) of the stored ABRs, the hearing thresholds were
61
June 2004
NoiseChem 62
determined as the lowest stimulus level, where both the first wave and the first trough of the ABR
could be clearly identified.
The only DPOAE measured was the amplitude of cubic distortion product (CDP, i.e. 2f1-f2 ),
having a fixed ratio of the primary tones (f1 and f2) of f2/f1 = 16/13 = 1.23, and always with the
level of f1 (L1) 10 dB higher than the level of f2 (L2 = L1 – 10 dB). The primary tones were
generated with a two-channel tone generator with phase control (HP 8904), and the output from
the probe microphone was feed to a FFT spectrum analyser (HP 35670A) under computer
control. The DP-grams across frequencies were obtained by measuring the cubic distortion
product (CDP) with fixed level of the primary tones (L1= 60 db and L2 = 50 dB SPL), and each
spectrum was made on 64 time-averaged recordings. DPOAE input/output-curves (I/O-curves)
were made by measuring the amplitude of the CDP to varying levels of the primary tones in 5 dB
steps. The number of time averaged recordings for each point of I/O-curves were based on
calculation of signal to noise ratio (S/N; N = mean of 8 bins, 4 on either side of the CDP), but was
not allowed to exceed 512, and the obtained amplitudes on the I/O-curves always had S/N better
than 3 dB. All test procedures and equipment was controlled by a computer and was programmed
in the visual programming language Agilent VEE (formerly HP VEE).
The 90 day study
The auditory measurements were made with an Etymotic Research ER-10B+ low noise
microphone system coupled to ER-2 tube phones by standard front tubes. The microphone probe
was placed in the left external ear canal by means of a probe holder, i.e. a small brass funnel fixed
to a micromanipulator. Before inserting the microphone probe, the position of the funnel was
adjusted to allow a direct view of the eardrum with an otoscope. The level of the stimulus from
the individual ER-2 tube phones were adjusted in accordance with the frequency response curves,
attained from the average response (84  1.5 dB SPL) from 1 to 17.5 kHz by an ER-7C probe
tube microphone at the ear drum of 6 animals (12 ears). The outputs of the ER-10B and ER-7C
microphones were adjusted in accordance with the frequency response curves supplied by the
manufacturer.
The hearing thresholds (4096, 8192 Hz, 12800 Hz, and 16384 Hz) were tested before and 8
weeks after exposure, while measurements of DPOAE where performed along the testing of the
hearing thresholds only after exposure. DP-grams were obtained by measuring the CDP with
fixed level and of the primary tones (L2 = 50 dB SPL), but varying f2 from 4096 to 17408 Hz in
steps of 512 Hz. DPOAE input/output-curves (I/O-curves) were made 4 frequencies of f2 (4096
Hz, 8192 Hz, 12800 Hz, and 16384 Hz) by measuring the amplitude of the CDP to varying levels
of the primary tones (L2= 20 - 80 dB SPL) in 5 dB steps. The rectal temperature of the rats was
throughout the measurement kept at 38.0 ± 0.5 ˚C. All auditory measurements were performed
randomly cage by cage, and without any knowledge of the exposure status of the rats.
Toluene, WBN and impulse noise
The auditory measurements were made with a custom made system comprising a probe system,
and using condenser microphones (B&K ½”, type 4191) as stimulus transducers. The distortion
products were picked up by a Knowles microphone (type FG3629/3452 electret microphone)
placed in the probe and pre-amplified with gain of 20 dB and before it was feed to the FFT
analyser (HP 35670A). The probe assembly consists of two parts: a probe body containing the
transducers and a fixture part to be placed on a micromanipulator for accurate placement of the
probe in the right external ear canal. By means of an otoscope, the correct position of the fixture
part was obtained, before the probe body itself was inserted. The levels of the output stimulus
signals were calibrated in an acoustic coupler with a cylindrical cavity of 2.6x3.9 mm towards a
B&K 1/8” condenser microphone (type 4138). The hearing thresholds of all animals were
measured at 16384 Hz before and 2 weeks days after exposure along measurements of DPOAE,
but for certain groups the tests of hearing was continued for up to 8 weeks after exposure. DP-
62
June 2004
NoiseChem 63
grams were obtained by measuring the CDP with fixed level of the primary tones (L2 = 50 dB
SPL), but varying f2 from 2048 to 73728 Hz in steps of 1024 Hz at 2048 - 10240 Hz, 2048 Hz at
12288 - 32768 Hz, and 4096 Hz at 36864 - 73728 Hz. DPOAE I/O-curves were made at 5
frequencies of f2 (4096 Hz, 8192 Hz, 16384 Hz, 32768 Hz, and 65536 Hz) by measuring the
amplitude of the CDP to varying levels of the primary tones (L2= 20 - 75 dB SPL) in 5 dB steps.
The rectal temperature of the rats was throughout the measurements kept at 37.5 ± 0.5 ˚C.
Toluene, CO and impulse noise
All procedures were identical to the description in the latter section.
Figure 3. A. Hearing thresholds in groups of rats (group mean and SEM) exposed to 0 ppm, 100
ppm, 200 ppm or 500 ppm toluene for 6 hours/day and 90 dB SPL wide band noise for 4
hours/day (Leq8hours = 87 dB SPL), 5 days/week for 90 days. Further, a control group exposed
to neither noise nor toluene. The hearings thresholds at 12.8 kHz are increased in the 0 ppm and
500 ppm groups when compared to the control group, but not in the 100 ppm and 200 ppm
groups. B. DP-grams from the same groups (f2/f1= 1.23; L1= 60 dB and L2= 50dB SPL). The
cubic distortion products (group mean and SEM) in the high frequency part of the DP-grams (f2
in the 10.2 - 17.4 kHz range) from the control group differs statistically significant from all the
other groups, while 100 ppm group differs from the control group as well as both the 0 ppm and
the 500 ppm group.
Statistics
Comparison of the measured effects in different dosage groups was performed using t-test for two
samples with different variance. All serial comparisons (DP-grams or DPOAE IO-curves between
individual groups were performed with analysis of variance (ANOVA) for repeated measures.
Unless otherwise stated, the term significant is used in the following when the statistical tests are
rejected at a 5 % level.
Results
The 90 day study
Figure 3A shows the hearing thresholds in the 5 groups of rats exposed to 0 ppm, 100 ppm, 200
ppm or 500 ppm toluene for 6 hours/day and 90 dB SPL wide band noise for 4 hours/day
(Leq8hours = 87 dB SPL), 5 days/week for 90 days. In all the groups exposed to noise there is
increase of hearing thresholds in the frequencies above 10 kHz when compared to the control
group. In 3B, the losses in auditory sensitivity of the groups exposed to noise are followed by a
63
June 2004
NoiseChem 64
loss in CDP in the DP-grams at comparable input frequencies. However, the changes are greatest
for the group exposed to noise only as well as the group exposed to noise simultaneous with
exposure to 500 ppm toluene, and both groups differs statistically significant from the control
group. The changes are somewhat less in the other two groups exposed to 100 toluene and noise,
where the changes do not reach statistically significant levels.
Figure 4. DPOAE I/O-curves at f2= 4096 Hz, 8096 Hz 12800 Hz, and 16384Hz in groups of rats
exposed to 0 ppm, 100 ppm, 200 ppm or 500 ppm toluene for 6 hours/day and 90 dB SPL wide
band noise for 4 hours/day (Leq8hours = 87 dB SPL), 5 days/week for 90 days, as well as a
control group exposed to neither noise nor toluene. The IO-curves (group mean and SEM) of the
cubic distortion products (L2= 30 - 60 dB) from the control group at 12 kHz and 16 kHz differs
statistically significant from all the other groups, while the I/O-curves at the same frequencies of
100 ppm group differs from the control group as well as both the 0 ppm and the 500 ppm group..
The same pattern is even more clearly seen the DPOAE IO-curves at 12 and 16 kHz (se Figure 4),
where the group exposed to the highest toluene concentration of 500 ppm closely follows the
group exposed to noise only, while both groups exposed noise and to the intermediary toluene
concentrations of 100 ppm and 200 ppm lies in between the latter groups and the control group.
Although there seems to be a dose-dependent effect of interaction to the low level exposure to
toluene and noise, the effects of interaction between these levels of exposure to toluene and noise
in rats seems to be antagonistic rather than additive or synergistic.
64
June 2004
NoiseChem 65
Figure 5. DP-grams (f2/f1= 1.23; L1= 60 dB and L2= 50dB SPL) of groups of rats (group mean
and 95% CI) exposed to a noise level of 92 dB SPL, as either wide band noise (WBN) or impulse
noise, 6 hours/day (Leq8hours = 90.8 dB SPL) for 10 days, simultaneous with exposure to either 0
ppm, 500 ppm, 1000 ppm or 1500 ppm toluene. The DP-grams of the groups marked Control in
these figures are the common means of all the animals in the other four groups in each of the
figures, measured before the rats were exposed to noise and toluene. NF denotes the noise floor,
computed as the mean of 8 bins, 4 on either side of the CDP. No effects of interaction is seem
between simultaneous noise exposures and toluene exposure at 500 ppm or 1000 ppm exposure
levels, but effects interaction is clear at the 1500 ppm level for both types of noise exposures.
Toluene, WBN and impulse noise
Figure 5 and Figure 6 shows the DP-grams from the all the 9 groups of rats in this part of the
study, when they were measured 14 days after the end of the exposures. Figure 5 displays the
differences between in effects of interaction of exposure to different levels of toluene and either
WBN or impulse noise. Figure 6 shows the main differences found between the two types of
noise exposures, where left section show the differences between the exposures to the to types of
noise exposure without concomitant toluene exposure, while right section shows the same with
simultaneous exposure to 1500 ppm toluene.
65
June 2004
NoiseChem 66
Figure 6. Top section shows the DP-grams (mean and 95% CI) of groups exposed to impulse or
wide band noise (WBN), comparable the ones shown in figure 5. Two groups of rats were
exposed to either 92 dB SPL WBN or impulse noise, 6 hours/day (Leq8hours = 90.8 dB SPL) for
10 days. The group marked Control are the common means of all the animals in the two groups,
before they were exposed to noise. The middle section shows the DP-grams for groups of rats
(mean and 95% CI) exposed to either 1500 ppm toluene only, 1500 ppm toluene and 92 dB SPL
WBN, or 1500 ppm toluene and 92 dB SPL impulse noise, 6 hours/day (Leq8hours = 90.8 dB SPL)
for 10 days. The DP-grams of the group marked Control in these figures are the common means
of all the animals in the three groups, before they were exposed to noise and toluene. NF denotes
the noise floor, computed as the mean of 8 bins, 4 on either side of the CDP. Comparison of the
two figures demonstrates the effects of interaction between exposure to 1500 ppm toluene and
either type of noise exposures. Impulsive noise is clearly more disruptive to hearing than WBN.
Likewise, the effects of interaction of exposures to impulse noise and toluene are proportionally
the impairment, induced by each type of noise without concomitant exposure to toluene.
Overall, the exposure to impulse noise induces clearly more loss of CDP than the WBN, and the
effects of interaction between the toluene and noise exposure seems also proportionally greater
for exposure to impulse noise at exposures to 1500 ppm toluene. However, there are no signs of
effects of interaction in the loss of CDP between the neither 500 ppm or 1000 ppm toluene
exposure and any of the to types of noise exposures displayed in figure 5.
Figure 7 shows the DPOAE IO-curves from all the groups also shown in Figure 6, and although
the mentioned differences are even more clearly demonstrated in this figure, the IO-curves does
contribute with any differences that are not clearly shown in the DP-grams in Figure 6 as well.
Toluene, CO and impulse noise
As shown above, impulsive noise disrupts hearing far more than WBN. Therefore, an impulsive
type of noise, where the major part of energy was contributed by noise impulses (se Figure 1),
was chosen for the final studies of the interaction between exposure to noise, CO, and toluene.
Initially, 3 experiments were performed with combined exposures to CO and to different levels of
the impulsive noise.
66
June 2004
NoiseChem 67
Figure 7. DPOAE I/O-curves at f2= 4096 Hz, 8096 Hz, 16384Hz and 32768 Hz in the same
groups of rats, measured at the same time at the DP-grams in Figure 6. The groups with open
markings are exposed to noise only, while the groups with closed markings are exposed
simultaneous to 1500 ppm toluene and either 92 dB SPL (Leq8hours = 90.8 dB SPL) wide band
noise (WBN) or impulse noise. The Control group is the common mean of all the animals in the
other groups in the figure, before they were exposure to toluene and noise. NF denotes the noise
floor, computed as the mean of 8 bins, 4 on either side of the CDP. Although the differences in the
CDP (mean and 95% CI) between the exposed groups in the figure seems rather clear, the IOcurves displays no further differences between the groups, than what are shown already in the
DP-grams in Figure 6.
In the first experiment, the combined exposure to 87 dB SPL (Leq8hours) impulsive and 0 ppm,
300 ppm or 500 ppm CO was investigated. The results of the measurements of the DP-grams
from these three groups of rats, ranging from the time before exposure and up to eight weeks after
the end of exposure, are shown in Figure 8. The auditory impairment from these exposures was
somewhat greater than expected, but the consistency of the measurements of the high-frequency
DPOAE is striking, and there seems to be little or no recovery in the measurements following the
first two weeks. The changes in hearing thresholds 16384 Hz (se Table 1) were highly correlated
with the losses in CDP in the DPOAE IO-curves at f2=16384 Hz (data not shown), i.e. that
r=0.93 (n=128) in the correlation of all measured threshold shifts with all measured losses in
CDP, and r=0.87 (n=33) in the correlation of the permanent threshold shift (PTS) with the final
measured losses of CDP.
67
NoiseChem 68
June 2004
Figure 8. DP-grams of groups of rats (group mean and 95% CI) exposed to 88.3 dB SPL
impulsive noise (Leq8hours = 87 dB SPL) and either 0 ppm, 300 ppm, or 500 ppm carbon
monoxide (CO) 6 hours/day for 10 days, when measured before exposure, the first day after
exposure, as well as 2 weeks, and 8 weeks after the end of exposure. NF denotes the noise floor,
computed as the mean of 8 bins, 4 on either side of the CDP. The consistency between the
measurements both before after exposure seems striking, and there appear to be little recovery of
the CDP in the measurements made later than 2 weeks after the end of the CO exposures.
Table 1. Compound threshold shifts (CTS) measured at 16384 Hz on the first day after exposure
(CTS 1), 2 weeks after exposure (CTS 2), and permanent threshold shifts eight weeks after
exposure (PTS) in the three groups of rats exposed to 0 ppm, 300 ppm, and 500 ppm CO and
impulse noise of 88.3 dB SPL, 6 hours/day for 10 days. *) Marks statistical significant difference
from the group without exposure to CO (0 ppm) at the 5% level.
ppm
0
300
500
CTS 1
dB
20.0 ± 7.1
29.0 ± 8.4
25.0 ± 3.0
CTS 2
dB
14.6 ± 6.6
29.0 ± 8.9 *)
18.5 ± 6.2
PTS
dB
11.7 ± 6.0
28.5 ± 10.3 *)
20.0 ± 5.6 *)
Figure 9 shows the effects of exposure to either 0 ppm, 300 ppm or 500 ppm CO and impulse
noise of different levels, i.e. Leq8hours= 81 dB, 84 dB SPL, or no noise exposure. A control
group without exposure to noise or CO is shown also in Figure 9. At exposure to 81 dB impulsive
noise there is only insignificant loss of auditory sensitivity (data on hearing thresholds are not
shown) and CDP, and no effects of interaction is notable with exposures to this noise level and
500 ppm CO. At exposure 84 dB impulsive noise there is loss of CDP at f2 ranging from 8 to 14
kHz (se also Figure 10, top section), as well as a clear synergistic interaction with the
simultaneous exposure to 500 ppm CO. With exposure to 84 dB impulsive noise, the group
exposed to 300 ppm lies in between the 0 ppm and the 500 ppm groups. Although the CDP is
68
June 2004
NoiseChem 69
statistically significant from the 0 ppm group at f2 ranging from 9 to 12 kHz, care should be taken
not to overestimate the effects of interaction, as the hearing impairment in the rats following
impulsive noise exposure is subjected to is a considerable variation within the groups.
Figure 9. DP-grams of groups of rats (group mean and 95% CI) exposed to 82.3 dB SPL
(Leq8hours = 81 dB SPL) and 85.3 dB SPL (Leq8hours = 84 dB SPL) impulse noise, and either 0
ppm, 300 ppm, or 500 ppm carbon monoxide (CO) 6 hours/day for 10 days. NF denotes the noise
floor, computed as the mean of 8 bins, 4 on either side of the CDP. The Control group in the top
section was exposed to neither noise nor CO, but was measured before and 2 weeks after sham
exposure. In the middle section, there is little effect of either 500 ppm CO exposure or the low
level noise exposure (Leq8hours= 81 dB SPL), and there seems to be no effects of interaction. In
the bottom section there is no statistical significant difference in the DP-grams between the
groups exposed to 84 dB SPL and either 0 ppm or 300 ppm CO, while there is clearly a loss of
CDP in the group exposed to 500 ppm CO in comparison to the group without exposure to CO.
Thus, the lowest observed adverse effect level (LOAEL) of CO exposure on the auditory
sensitivity in rats found in this study was 500 ppm, and the only effect of the CO exposure
demonstrated was a potentiation of the effects of the noise exposure.
69
June 2004
NoiseChem 70
Figure 10 shows the effects of exposure to 84 dB impulse noise with simultaneous exposure to
both CO (0 ppm, 300 ppm, and 500 ppm) and toluene (0 ppm, 500 ppm and 1000 ppm). The top
part of the figure shows the effects of simultaneous noise and toluene exposure, without exposure
to CO, and there seems to be no effects of interaction at these levels of noise and toluene
exposure. However, in the middle section, both 500 ppm and 1000 ppm toluene exposure seems
to induce a comparable synergistic interaction at the same frequencies as demonstrated in the
group exposed to 500 ppm CO and impulsive noise and without concomitant toluene exposure (se
bottom sections Figure 9 and Figure 10). In the bottom section of Figure 10, the synergistic
interaction between exposure to noise, 500 ppm CO and toluene is further increased in the group
exposed to 1000 ppm toluene, but only at the frequencies of f2 ranging from 12 to 24 kHz, while
the effects of interaction below 12 kHz seems to have saturated at the same level as in the two
groups exposed to noise, toluene and 300 ppm CO in the middle section of Figure 10.
Figure 10. DP-grams of groups of rats
(group mean and 95% CI) exposed
simultaneous to 85.3 dB SPL
(Leq8hours = 84 dB SPL) impulsive
noise, as well as either 0 ppm, 500 ppm
and 1000 ppm toluene, and 0 ppm or
300 ppm, or 500 ppm carbon monoxide
(CO), 6 hours/day for 10 days. NF
denotes the noise floor, computed as
the mean of 8 bins, 4 on either side of
the CDP. In the top section, no effects
of interaction were found with impulse
noise exposure simultaneous with the
exposures to either 500 ppm or 1000
ppm toluene. However, at the 300 ppm
CO exposure level in the middle
section, with combined exposure also to
either 500 ppm or 1000 ppm toluene,
the effects of interaction is notable at
the frequencies below 16 kHz. Further,
the loss of CDP seems to be of equal
size, i.e. the effect was more or less
independent of the toluene exposure
level. In the bottom section, with the
combined both noise, 500 ppm CO, and
to 500 ppm toluene there seems to be
no further effects of interaction than
was evident by exposure to CO and
impulse noise, while the exposure to
1000 ppm toluene clearly increases the
loss of CDP in the 12-24 kHz frequency
range. Thus, below the LOAEL of 500
ppm CO exposure (se Figure 9),
exposure to 500 ppm toluene seems to
have saturated the effects of
interaction, as an increase of the
toluene exposure level to 1000 ppm
seems to be without further loss in CDP at the frequencies below 16 kHz. However, above the
70
June 2004
NoiseChem 71
LOAEL of CO exposure there seems to be dose dependent effects of interaction of the exposure to
toluene with concomitant exposure to CO and
impulse noise on the loss of CDP at the frequencies between 12-24 kHz.
Discussion
The first objective in the present series of studies was to investigate whether long-term, low-level
exposure to both toluene and steady state noise may cause auditory impairment in rats. This
would imply, that the general used short-term exposure scheme of 5-14 days for testing the
ototoxic potential of organic solvents like toluene would be inadvertent to study the effects of
interaction between the exposure to organic solvents and noise. However, as shown in fig. 3 and
4, no signs of synergistic or additive interactions was found from the 90 day sub-chronic
exposures to 100 ppm, 200 ppm, and 500 ppm toluene. Contrary to what may have been
expected, exposure to 500 ppm toluene and noise was followed by the same changes in hearing
thresholds and loss of cubic distortion products (CDP) as exposure to noise only, while exposure
to 100 ppm and 200 ppm toluene and noise was followed by les reduction in auditory sensitivity
than exposure to either noise alone and 500 ppm toluene and noise (se figure 10). The effects of
the toluene exposures seems to display a dose-response relationship, but of antagonistic nature at
the low levels of the toluene exposures. The same pattern is seen clearly in all the measurements
of the hearing the in mid-frequency region of 12-16 kHz, but it is most clearly shown in the
input/output-curves (IO-curves) at f2=12800 Hz and f2=16384 Hz. Dose-dependant effects are
regarded as a significant sign of causal coherence in toxicological studies, and could be seen as a
clear indication of an antagonistic interaction from combined exposure to toluene and steady state
noise. However, the differences between the groups are small (~5 dB) and coincidence may have
ruled the distribution of the outcome between the groups. The overall conclusion from this study
is that even after long-term of combined exposure to ototoxic organic solvents like toluene and
noise, no signs of synergistic or additive interaction have was found below the LOAEL from
short-term studies
The next issue for investigation was whether the interaction from combined exposure to organic
solvents and noise and may have been underestimated in the rat model due to the general use of
steady state, un-impulsive band type of noise, with little similarity to human noise exposure in the
working environment. As the differences in hearing impairment between exposures to the
different types of noise may be shown only at higher frequencies than tested previously, the
techniques for the tests of hearing had to be improved. From the earlier findings it was clear, that
although the measurements of DPOAE did not increase the sensitivity of the auditory testing, the
overall variation between the measurements from groups of rats was consistently very small. It
was therefore decided, the main part of the relatively time consuming measurements of hearing
thresholds by sampling of ABR could preferentially be compensated by measurements of the DPgrams as well as the DPOAE IO-curves, and this approach would also allow tests of hearing over
the hole frequency range, and with the rats in anaesthesia for only a relative short periods of time.
Two types of noise exposure were investigated for potential effects of interaction with combined
exposure to toluene. The schedules of the noise exposures was changed in order to allow for short
periods of high levels noise exposure, either by varying the level of the WBN or by varying the
time between the noise impulses, each with a peak level of slightly more than 130 dB (se Figure
1). To include the effects in the high frequency range of hearing, the frequency distribution of the
two types of noise was extended from 4-20 kHz to 4-24 kHz. The frequency distribution of the
noise was fairly equal regarding the WBN, but the impulse noise did have a somewhat less equal
frequency distribution due to limitations of the loud speakers in use (se Figure 2). However, the
frequency distributions of the resulting hearing loss from the exposures to the two types of noise
exposure were clearly different. The rats exposed to WBN had the main loss of sensitivity at 12-
71
June 2004
NoiseChem 72
24 kHz, while the rats exposed the impulse noise had a hearing loss in the 4-24 kHz frequency
band, although it was most pronounced at frequencies around 9 kHz, i.e. at the frequencies with
the maximum amplification by the formation of standing waves within the external ear canal of
the rats (se Figure 5 and Figure 6). In general, the loss of auditory sensitivity found in the rats
from the impulse noise exposures has a larger variation, than what is seen from exposures to
WBN, but the auditory impairment induced by impulse noise is considerably greater, which can
be seen in Figure 5, Figure 6, and Figure 7. No effects of interaction were found from combined
exposures to either type of noise exposures and toluene exposure levels of 500 ppm or 1000 ppm.
Exposure to 1500 ppm toluene without simultaneous noise exposure did not induce notable
auditory impairment, while the combined exposure including impulse noise induced a greater loss
of auditory sensitivity over a broader range of frequencies than exposure to WBN. While the
exposure to the impulse noise alone was considerably more disruptive to hearing than WBN, the
effects of interaction from combined exposure to toluene and impulse noise only seems only to be
proportionally greater than the effects of interaction of combined exposure to toluene and WBN.
The effects of toluene exposures are initially seen in the rat in a narrow mid-frequency band,
which broadens out to include all frequencies from 4 to 32 kHz (Campo et al. 1997; Muijser et
al., 2000). With combined exposures to noise and organic solvents, synergistic effects of
interaction are observed, depending on the level and the frequency band of the concomitant noise
exposure (Brandt-Lassen et al 2000; Campo et al, 2000; Cappeart et al., 2001). Altogether,
impulsive noise seems to have the potential to cause hearing impairment even at relative low
levels of noise exposure (Starck et al., 2003), and the possible effects of interaction of impulsive
noise and exposure to organic solvents my in fact increase the risk of auditory impairment
significantly at very realistic scenarios of human exposure in the working environment.
The intention with the last study was to investigate the potential interaction of two risk factors of
auditory impairment in conjunction with the exposure to an organic solvent like toluene. Studies
in rats have shown that CO exposure potentiates the effects of noise induced hearing loss, while
exposure to CO alone has no effect (Chen and Fechter, 1999; Fechter et al., 2000). Several studies
have also shown that tobacco smoking is a risk factor for hearing loss in humans (Starck et al.,
1999), which may be partly caused by the concomitant exposure to CO. It was therefore decided
to test the potential synergetic effects of simultaneous exposure to toluene, CO, and impulsive
noise. The frequency range of the impulsive noise was changed back again to the 4-20 kHz
frequency band, because the main impairment of the impulse noise was found in the low
frequency range and not in high frequency range.
In three experiments the interaction of CO and varying levels of impulse noise was investigated,
and the main results of these studies can be seen in Figure 8 and Figure 9. The 87 dB SPL
impulsive noise exposure had a more dramatic effect on the hearing of the rats than expected
from the previous study of impulse noise, which may be caused by the change in the frequency
distribution of the noise as well as the time interval between the impulses. Basically, the same no
adverse effect level (NOAEL) of 300 ppm CO and LOAEL of 500 ppm CO found previously by
Fechter at al. (2000) in rats, was also found in the present study in conjunction with exposure to
both 87 dB and 84 dB impulsive noise, although the exposure to 300 ppm CO did seem to show a
potentiation of noise exposure close to the level of statistically significance.
The effects of combined exposure to toluene, CO, and impulsive noise were investigated in two
experiments, and the main results from these two studies are shown in figure 10. Without
exposure to CO, combined exposure to impulsive noise and 500 ppm or 1000 ppm toluene did not
show any effects of interaction. However, exposure also to CO at the NOAEL of 300 ppm,
concomitant exposure to either 500 ppm or 1000 ppm toluene did induce a potentiation of the
effects of the noise exposure, and primarily in the frequency range around 9 kHz. Thus, it seems
72
June 2004
NoiseChem 73
as the exposure to toluene potentiated the effects of the CO exposure up to a level, where the
detrimental effects of CO exposure saturates at the LOAEL. Further increase in the level of the
CO exposure did not significantly increase the potentiation in frequency region around 9 kHz, but
a new frequency region showed clearly an increased loss of auditory sensitivity in the group
exposed to 1000 ppm toluene, i.e. the 12-24 kHz region, where toluene normally would normally
have a synergetic effect on noise exposure at the 1500 ppm exposure level (se Figure 5 and
Figure 6).
Chen and Fechter (1999) has demonstrated, that CO potentiates the effects of octave band noise at
any frequency region, depending on the noise band, and further that the auditory impairment
induced by combined exposure to noise and CO seemed to lack the partially recovery normally
seen after noise exposure. The potentiation found in the present study did not increase in the lowfrequency region when the level of toluene exposure is increased from 500 ppm to 1000 ppm, and
exposure to toluene levels below these concentrations was not performed, potentiation below 500
ppm is a realistic possibility. Thus, if CO potentiates the noised induced hearing loss by reducing
the repair mechanisms inside the cochlea, the toluene exposures levels in the present study seems
potentiate the effects the CO exposure, i.e. the effects of the noise exposure more or less saturates
in frequency band, where the energy of the noise exposure is concentrated (Chen and Fechter,
1999; Rao and Fechter, 2000). These findings suggests that synergistic interactions in relation to
combined exposures noise and one ototoxic substance are found primarily at the LOAEL, but the
inclusion of exposure to other agents at low exposure levels may seriously increase the auditory
impairment. The potentiation of ototoxic chemicals on auditory impairment seems to arise close
to the LOAEL of the noise exposure, and the potentiation increases further to reach to maximum
at medium noise levels (Rao and Fechter, 2000). At high levels of noise exposure, however, the
effect of combined exposure to CO seems to saturate without notable potentiation from the
concomitant chemical exposure (Rao and Fechter, 2000).
In the present study, the noise level of the noise exposure are fairly low, but impulsive noise may
pose a threat to hearing even at noise exposure levels well below the occupational exposure limits
(Starck et al., 2003). The risk of human hearing loss from impulse noise has been studied in
analysis of data from noise-exposed workers (NoiseScan; Toppila et al., 2000), and
impulsiveness of the noise had greater impact than whether the level of noise exposures are above
or below median level of exposure. With inclusion of more risk factors in the analysis, high blood
cholesterol levels and intake of painkillers potentiated the effects of noise exposure at hearing
levels of 4 kHz, while increase of systolic blood pressure was also a risk of loss of hearing but did
not interact with an the exposure to noise. However, the noise levels of the exposed workers in
the analysis were well above 85 dB for most of the workers, even when the levels of the noise
exposures were measured below the hearing protection devices used. This would imply that the
effects of combined interaction of several factors could have even greater impact on the hearing
loss in workers exposed to more moderate levels of noise exposure, as shown in the present study
of rats exposed impulsive noise, toluene, and CO. Therefore, exposure to organic solvents like
toluene may be a risk factor of major human concern, especially in combination with exposure to
noise and other risk factors of hearing impairment like exposure to other ototoxic chemical
substances, tobacco smoking, intake of medicine and alcohol, etc. (Starck et al, 1999, Toppila et
al, 2000).
Conclusion
The effects of interactions of different patterns of combined exposures of toluene and noise have
been investigated in the rat model. All the studies were performed with toluene as the ototoxic
organic solvent. Prolonged exposures to steady state, wide band noise and toluene did not
increase the risk of synergetic interactions, unless the exposure concentration of toluene was close
73
June 2004
NoiseChem 74
to the low observed adverse effect level (LOAEL). An exposure schedule including short periods
of higher noise levels of did not increase the interactions between noise and toluene exposures.
Impulse noise induces caused considerable more impairment to hearing than wide band noise, and
the effects of interaction was found to proportional to the greater auditory impairment of the
impulse noise without exposure to toluene. Considerable hearing loss was found from exposure to
impulsive type of noise at 84 dB SPL, with 75% of the energy as noise impulses. No direct
synergetic interaction with toluene exposures was noted at this level of noise exposure, but a
potentiation of the effects of exposure to carbon monoxide (CO) was evident, even at the lowest
exposure levels of toluene. Combined exposure above the LOAEL of CO potentiation of noise
exposure increased LOAEL of toluene by 50%. Extrapolation from these studies to the human
working environment would imply, that combined exposure to both noise and organic solvents
may increase the risk of auditory hearing impairment of tobacco smokers significantly. If the
results can be generalized to other risk factors for hearing impairment, a major part of the hearing
impairment by occupational exposures may be caused by interactions of low to medium levels
noise exposure in combination with the individual risk factors of hearing loss.
References
Brandt-Lassen R., Lund S.P., Jepsen G.B. (2000) Rats exposed to Toluene and Noise may
develop Loss of Auditory Sensitivity due to Synergistic Interaction. Noise Health. 3(9): 33-44.
Campo P., Pouyatos B., Lataye R., Morel G. (2003) Is the aged rat ear more susceptible to noise
or styrene damage than the young ear? Noise Health. 5(19): 1-18.
Campo P., Lataye R., Loquet G., Bonnet P. (2001) Styrene-induced hearing loss: a membrane
insult. Hear. Res. 154:170-180.
Campo P., Lataye R., Cossec B., Villette V., Roure M., Barthelemy C. (1998) Combined effects
of simultaneous exposure to toluene and ethanol on auditory function in rats. Neurotoxicol.
Teratol. 20: 321-332.
Campo P., Lataye R., Cossec B., Placidi V. (1997) Toluene-induced hearing loss: a midfrequency location of the cochlear lesions. Neurotoxicol. Teratol. 19:129-140.
Cappaert N.L., Klis S.F., Muijser H., Kulig B.M., Ravensberg L.C., Smoorenburg G.F. (2002)
Differential susceptibility of rats and guinea pigs to the ototoxic effects of ethyl benzene. :
Neurotoxicol. Teratol. 24: 503-510.
Cappaert N.L., Klis S.F., Muijser H., Kulig B.M., Smoorenburg G.F. (2001) Simultateous
exposure to ethyl benzene and noise: synergistic effects on the outer hair cells. Hear. Res. 162:
67-79.
Cappaert N.L., Klis S.F., Baretta A.B., Muijser H., Smoorenburg G.F. (2000) Ethyl benzeneinduced ototoxicity in rats: a dose-dependent mid-frequency hearing loss. J. Assoc. Res.
Otolaryngol. 1: 292-299.
Chen G.D., Fechter L.D. (1999) Potentiation of octave-band noise induced auditory impairment
by carbon monoxide. Hear. Res. 132: 149-59.
Crofton K.M., Lassiter T.L., Rebert C.S. (1994) Solvent-induced ototoxicity in rats: An atypical
selective mid-frequency hearing deficit. Hear. Res. 80:25-30.
Crofton, K.M., Zhao, X. (1993) Mid-frequency hearing loss in rats following inhalation exposure
to trichloroethylene: Evidence from reflex modification audiometry. Neurotoxicol. Teratol.
15:413-423.
Davis R.R., Murphy W.J., Snawder J.E., Striley C.A., Henderson D., Khan A., Krieg E.F. (2002)
Susceptibility to the ototoxic properties of toluene is species specific. Hear. Res. 166: 24-32.
Fechter L.D., Chen G.D., Rao D., Larabee J. (2000) Predicting exposure conditions that facilitate
the potentiation of noise-induced hearing loss by carbon monoxide. Toxicol. Sci. 58: 315-323.
Jaspers, B.M.A., Muijser, H., Lammers, J.H.C.M., Kulig, B.M. (1993) Mid-frequency hearing
loss and reduction of acoustic startle responding in rats following trichloroethylene exposure.
Neurotoxicol. Teratol. 15:407-412.
74
June 2004
NoiseChem 75
Johnson, A.-C., Canlon, B. (1994a) Toluene exposure affects the functional activity of the outer
hair cells. Hear. Res. 72:189-196.
Johnson, A.-C., Canlon, B. (1994b) Progressive hair cell loss induced by toluene exposure. Hear.
Res. 75:201-208.
Johnson, A.-C., Juntunen, L., Nylén, P., Borg, E., Höglund, G. (1988) Effect of interaction
between noise and toluene on auditory function in the rat. Acta Otolaryngol.(Stockh) 105:56-63.
Johnson A.-C., Nylén P., Borg E., Hoglund G. (1990) Sequence of exposure to noise and toluene
can determine loss of auditory sensitivity in the rat. Acta. Otolaryngol. (Stockh) 109:34-40.
Lataye R., Campo P., Pouyatos B., Cossec B., Blachere V., Morel G. (2003) Solvent ototoxicity
in the rat and guinea pig. Neurotoxicol. Teratol. 25: 39-50.
Lataye R., Campo P., Loquet G. (2000) Combined effects of noise and styrene exposure on
hearing function in the rat. Hear. Res. 139: 86-96.
Lataye, R., Campo, P. (1997) Combined effects of a simultaneous exposure to noise and toluene
on hearing function. Neurotoxicol. Teratol. 19:373-382.
Loquet G., Campo P., Lataye R., Cossec B., Bonnet P. (2000) Combined effects of exposure to
styrene and ethanol on the auditory function in the rat. Hear. Res. 148:173-180.
Loquet G., Campo P., Lataye R. (1999) Comparison of toluene-induced and styrene-induced
hearing losses. Neurotoxicol. Teratol. 21: 689-897.
Makitie A.A., Pirvola U., Pyykko I., Sakakibara H., Riihimaki V., Ylikoski J. (2002) Functional
and morphological effects of styrene on the auditory system of the rat. Arch. Toxicol. 76:40-47.
Makitie A.A., Pirvola U., Pyykko I., Sakakibara H., Riihimaki V., Ylikoski J. (2003) The
ototoxic interaction of styrene and noise. Hear. Res. 179: 9-20.
McWilliams M.L., Chen G.D., Fechter L.D. (2000) Low-level toluene disrupts auditory function
in guinea pigs. Toxicol Appl Pharmacol. 167:18-29.
Muijser H., Lammers J.H., Kulig B.M. (2000) Effects of exposure to trichloroethylene and noise
on hearing in rats. Noise Health. 2(6): 57-66.
Nylén, P., Hansson, A.-C., Hagman, M., Höglund, G., Linder, G. (1987) Inhalation exposure to
toluene, effect of daily dose on auditory sensitivity and nerve conduction velocity in rats.
Proceedings of the XXII International Congress on Occupational Health. Sydney , September 27october 2, 1987.
Pouyatos B., Campo P., Lataye R. (2002) Use of DPOAEs for assessing hearing loss caused by
styrene in the rat. Hear. Res. 165: 156-64.
Pryor G.T., Dickinson J., Howd R.A., Rebert C.S. (1983) Neurobehavioral effects of subchronic
exposure of weanling rats to toluene or hexane. Neurobehav. Toxicol. Teratol. 5:47-52.
Pryor, G.T., Dickinson, J., Feeney, E.M, Rebert, C.S. (1984a) Hearing loss in rats first exposed to
toluene as weanlings or as young adults. Neurobehav. Toxicol. Teratol. 6:111-119.
Pryor, G.T., Rebert, C.S., Dickinson, J., Feeney, E.M (1984b) Factors affecting toluene-induced
ototoxicity in rats. Neurobehav. Toxicol. Teratol. 6:223-228.
Pryor, G.T., Rebert, C.S., Howd, R.A. (1987) Hearing loss in rats caused by inhalation of mixed
xylenes and styrene. J. Appl. Toxicol. 7:55-61.
Pryor, G.T., Rebert, C.S., Kassay, K., Kuiper, H., Gordon, R. (1991) The hearing loss associated
with exposure to toluene is not caused by a metabolite. Brain. Res. Bull. 27:109-113.
Rao D.B., Fechter L.D. Increased noise severity limits potentiation of noise induced hearing loss
by carbon monoxide. Hear. Res. 150: 206-214
Rebert C.S., Sorenson S.S., Howd R.A., Pryor G.T. (1983) Toluene-induced hearing loss in rats
evidenced by the brainstem auditory-evoked response. Neurobehav. Toxicol. Teratol. 5:59-62.
Rebert,C.S., Day,V.L., Matteucci M.J., Pryor G.T. (1991) Sensory-evoked potentials in rats
chronically exposed to trichloroethylene: Predominant auditory dysfunction. . Neurobehav.
Toxicol. Teratol. 13:83-90.
Rebert, C.S., Schwartz, R.W., Svensgaard, D.J., Pryor, G.T, Boyes, W.K. (1995) Combined
effects of paired solvents on the rat's auditory system. Toxicol. 105:345-354.
75
June 2004
NoiseChem 76
Simonsen, L., Lund, S.P. (1995) Four weeks inhalation exposure to n-heptane causes loss of
auditory sensitivity in rats. Pharmacol. Toxicol. 76:41-46.
Starck J., Toppila E., Pyykko I. (2003) Impulse noise and risk criteria. Noise Health. 5(20): 6373.
Starck J., Toppila E., Pyykko I. (1999) Smoking as a risk factor in sensory neural hearing loss
among workers exposed to occupational noise. Acta Otolaryngol. 119: 302-305.
Sullivan M.J., Rarey K.E., Conolly R.B. (1989) Ototoxicity of toluene in rats. Neurotoxicol.
Teratol. 10:525-530.
Toppila E., Pyykko I., Starck J., Kaksonen R., Ishizaki H. (2000) Individual Risk Factors in the
Development of Noise-Induced Hearing Loss. Noise Health. 2(8): 59-70.
Yano, B.L., Dittenber, D.A., Albee, R.R., Mattsson, J.L. (1992) Abnormal auditory brainstem
response and cochlear pathology in rats induced by an exaggerated styrene exposure regimen.
Toxic. Path. 20:1-6.
76
NoiseChem 77
June 2004
NoiseChem
Human Studies Lab Reports
77
June 2004
NoiseChem 78
Lab 1: Ann-Christin Johnson
National Institute for Working Life North,
Department of Work and the Physical Environment,
Sweden.
Karolinska Institutet, Unit of Audiology, Stockholm,
Sweden
78
June 2004
NoiseChem 79
Effects on the auditory system after exposure to styrene and noise in Swedish fiberglass
industry workers.
Ann-Christin Johnson 1,2 and Thais Morata 3
1. National Institute for Working Life North, Department of Work and the Physical Environment,
Umeå, Sweden
2. Karolinska Institutet, Unit of Audiology, Department of Clinical Science, Karolinska
University Hospital, Huddinge M41, S-141 76 Stockholm, Sweden.
3. National Institute for Occupational Safety and Health, Division of Applied Research
and Technology, Cincinnati, OH, 45226,United States.
Introduction
Styrene exposure causes a permanent and progressive damage to the auditory system of the rat
and several experiments revealed that noise interacts with styrene (Lataye et al., 2000; Makitie,
2003; Campo et al., 2003) in a synergistic manner. This has serious implications because noise is
the most likely agent to occur in settings where there is styrene exposure. Age also played a role
in the interaction between noise and styrene. Young rats had significant less hair cell loss than
older animals (Campo et al., 2003).
Styrene was shown to be a more potent ototoxicant than toluene (Campo et al., 1999).
Several field studies have been conducted with styrene exposed workers in boat or fiberglass
products manufacturing. Muijser et al., 1988 reported that workers exposed to low levels of
styrene did not appear to have increased hearing loss at high frequencies when compared to
controls. The comparison of the two extreme exposure groups, however, revealed a statistically
significant difference in hearing thresholds at high frequencies.
Styrene and noise exposures were assessed for 299 workers in the reinforced fiber industry (SassKortsak et al., 1995). Noise levels ranged between 85 to 90 dB(A), while styrene levels were
generally below the recommended level of 50 ppm. The association between noise exposure,
based on the developed lifetime noise dose estimate, and hearing loss (assessed by averaging pure
tone thresholds) was significant. That was not the case for styrene exposure. Styrene exposure
approached significance for hearing loss only at some specific frequencies (4 and 6 kHz, left ear;
Sass-Kortsak et al., 1995).
More recently, the effects of styrene were investigated in male workers exposed in factories that
produced plastic buttons or bathtubs (Morioka et al., 1999, 2000). Medical examinations,
audiological evaluations and exposure assessment to noise and solvents (in air and urine) were
conducted in both investigations. In the 1999 study, workers whose noise exposures exceeded 85
dBA were excluded from the study population. Participants were exposed to a mixture of solvents
containing mainly styrene and toluene. Of the 93 participants, only 6 were exposed to levels of
styrene that exceeded the 50 ppm Japanese exposure limit, and 2 were exposed to toluene levels
exceeding the Japanese limit of 50 ppm (Morioka et al., 1999). In the 2000 study, 48 study
participants were divided in 3 subgroups by their exposure condition: a control, unexposed group,
a group exposed to low levels of styrene (2.9 to 28.9 ppm) and noise (69 to 76 dBA), and group
exposed to noise levels that ranged from 82 to 86 dBA (Morioka et al., 2000). No effects of the
solvents were detected by conventional pure-tone audiometric testing, but the detection of high
frequency tones was reduced in workers exposed to styrene for 5 years or more. This effect was
associated to styrene concentrations in air and mandelic acid concentrations in urine. No effects
of other solvent exposure were detected.
The association of the biological determinant of styrene and auditory dysfunction was also
observed in the first report of the present cross sectional study conducted in Sweden, which aimed
to investigate the effects of occupational exposure to low levels of styrene and noise (Morata et
al., 2002). The study protocol included a questionnaire, assessment of styrene and noise
79
June 2004
NoiseChem 80
exposures, and an audiologic battery. The questionnaire gathered information on work history,
non-occupational solvent and noise exposure, and medical history. Exposure assessment included
gathering data from interviews and company records, and site measurements of noise levels for
different work tasks. Styrene measurements were conducted on all exposed workers by air
samples and biological monitoring of mandelic acid in urine. About 60% of the participants in
both groups exposed to noise (styrene/noise and noise alone) were exposed to noise levels above
the Swedish threshold limit value (85 dBA/3 dB exchange rate) and the range of exposure was
also similar in these groups (75-116 dBA). Styrene exposures were low, averaging 3.5 ppm with a
maximum level of 22 ppm, (8 h values, TLV (8h) in Sweden is 20 ppm). Workers exposed to
noise and styrene had significantly worse pure-tone thresholds at 2, 3, 4, and 6 kHz when
compared with noise-exposed or nonexposed workers. From the numerous variables that were
analyzed for their contribution to the development of hearing loss, age, noise exposure (past and
current) and mandelic acid levels (one of the biologic markers for styrene in urine), were the only
variables that met the significance level criterion in the final multiple logistic regression model.
The odds ratio estimates for hearing loss were 2.44 times greater for each mmol of mandelic acid
per gram of creatinine (95% CI: 1.01-5.88), 1.18 times greater for each dB of current noise
exposure (cumulative exposure index, 95% CI : 1.01-1.38), and 1.19 greater for each year of age
(95% CI : 1.11-1.28).
As part of a large investigation conducted in Poland, also within NoiseChem, (SliwiñskaKowalska et al. 2003), a group of styrene exposed workers (n= 194) was compared with groups
of workers exposed to styrene and noise (n= 56), styrene and toluene (n=26), styrene, toluene and
noise (n=14), noise (66) or unexposed (n=157). The study protocol included a questionnaire,
assessment of styrene and noise exposures, and an audiologic test battery. The questionnaire, an
adapted and translated version of the protocol used by Morata et al. (2002), included questions on
work story, non-occupational solvent and noise exposure, and medical history. The participants
from the styrene group were exposed to noise levels ranging from 78-86 dBA. Styrene average
exposures (8 h values) in the styrene groups ranged from 11 to 38 ppm with a maximum level of
120 ppm. Hearing loss was observed in 76% of the workers exposed to styrene and noise or
styrene and toluene, in 57% of the styrene only group, 56% in the noise exposed group, and 33%
in the unexposed group. Significantly higher mean audiometric thresholds (p<0.05) were
observed in the styrene exposed workers at 2, 4 and 6 kHz, when compared to the noise only and
the unexposed groups. When compared to the solvent exposed groups, mean thresholds were also
significantly higher at the frequencies of 4 and 8 kHz. The odds ratio estimates for hearing loss
were also the highest among all groups exposed to styrene.
In a clinical study, 18 workers underwent not only pure-tone audiometry, but a more complete
audiological and otoneurological test battery (Möller et al, 1990). Routine audiometric results of
workers exposed to styrene in a plastic boat industry did not indicate hearing losses resulting
from causes other than exposure to noise (Möller et al, 1990). Seven of eighteen workers,
however, displayed abnormal results in the distorted speech tests and cortical response
audiometry, as well as in some of the otoneurological tests performed.
The objective of this report is to present the results of an extensive audiological test-battery used
in workers exposed to styrene alone or in combination with noise, in Sweden. Pure-tone
audiometric data from this population was reported in an earlier publication (Morata et al., 2002).
We are now presenting the results from complementary audiological tests, including: distortion
product otoacoustic emission (DPOE), psychoacoustical modulation transfer function (PMTF),
distorted speech, speech recognition in noise, and cortical response audiometry. These results are
compared to the pure tone audiometry measurements.
80
June 2004
NoiseChem 81
Material & Methods
Subjects
The selection of styrene and the fiberglass products (FGP) industry took into consideration
styrene's potential neurotoxicity, available evidence of ototoxicity, severity of the problem,
accessibility and reliability of industry exposure records, accessibility to workers, and magnitude
of an occupationally exposed population. Although exclusive chemical exposure to solvents is
rare, the FGP industry is one of the few occupational settings, which has an almost
mono-exposure to styrene (small amounts of a second chemical acetone, is used for cleaning tools
in FGP industries). Eleven FGP manufacturers agreed to participate in this study, varying in size
from 5 to 500 employees. Of those, 154 styrene-exposed workers participated, 65 who were not
exposed to excessive noise levels (above 85 dBA time-weighted-average [TWA]). Noise-exposed
controls (n = 78) were selected from three companies in the metal products manufacturing
industry, and the non-exposed controls (n=81) were selected from a mail distribution terminal.
Details of drop-outs (n= 13) has been described by Morata et. al (2002).
All styrene-exposed workers employed for a minimum of one year in the FGP companies were
invited to participate in the study. The workers from the metal industry were selected based on
noise exposure levels that were equivalent to those of the FGP workers. The unexposed controls
were randomly chosen from a large number of employees in a mail distribution terminal. The
noise levels in the terminal were below 85 dBA TWA, and none of those workers had a known
history of exposure to occupational noise.
The project was approved by the Karolinska Institutets Ethical Committee.
Re-testing part of the subjects exposed styrene
One FGP plant was re-visited at the end of the project and 30 of the workers were re-examined
using the same exposure measurements, the same questionnaire and a new pure-tone audiometry.
Styrene Exposure Assessment
Details of the exposure assessment and the analyses of the samples have been described earlier
(Morata et al 2002) and will only be summarized here. To determine the level of exposure to
styrene, TWA exposure evaluations were conducted on all subjects exposed to styrene and in five
of each of the other groups for control purposes only. Passive samplers were used and two
successive samples were collected for each worker. The adsorption tube samples were then
securely sealed and stored in a freezer for later gas chromatography (GC) analysis.
Total styrene exposure was assessed also by the biological monitoring of mandelic acid and
creatinine in the urine collected during a 24hour period. Collection started at the beginning of the
work shift. Samples of urine were then taken for analyses by high-pressure liquid
chromatography (LC).
Noise Exposure Assessment
Details of the exposure assessment have been described earlier (Morata et al 2002) and will only
be summarized here.
Noise exposure was assessed by personal exposure measurements using noise dosimeters (Brüel
& Kjær 4436). Exposure assessments were calculated individually, based on eight-hour level
equivalent dosimeter measurements, Leq 8 dB (A). At least, one full-shift noise dosimetry was
performed for all different work tasks. For the workers (n = 128) whose personal noise dosimetry
was not performed, a mean value of the noise levels obtained from workers performing the same
work tasks was used in the analyses.
Estimates of exposure during work at jobs held before the present one were calculated, based on
information given by the worker, for example, if the workplace was “quiet” or “very noisy.”
Information on typical noise levels in certain industries was found in the literature (NoiseScan
81
NoiseChem 82
June 2004
Database, Pykköö et al 2002) and included in the estimation of past exposure. For each subject,
the total cumulative noise exposure (CNE) was calculated.
The characteristics of the study population regarding their age, tenure, and current and previous
exposures to the studied agents are presented in Table 1.
Table 1. Characterization of the study population (n=313). Mean values and range (within
parenthesis) for the variables of age, tenure, previous noise exposure, current noise and styrene
exposures, and estimated lifetime noise and styrene exposures.
Non-exposed Noise
Styrene
Styrene & Noise
(n= 89)
Variable
(n=81)
(n=78)
(n=65)
Age (years)
45 (26–62)
Tenure* (years)
18 (2–38)
Previous
noise 7 (0–25)
exposure (≥ 85 dBA
TWA) (years)
42 (20–64)
12 (1–35)
12 (1–26)
43 (21–62)
17 (1–39)
5 (0–16)
43 (21–65)
15 (2–37)
6 (0–21)
85 (75–116)
82 (75–84)
89 (85–108)
Lifetime
noise 79
exposure (dBA)
86
84
89
Current
(mg/m3)
styrene -
-
16 (0.2–96)
12 (0.03–50)
Mandelic acid in urine (mmol/g createnine)
-
0.9 (<MDC–2.9)
0.9 (<MDC–3.0)
Lifetime
styrene (mg-years/m3)
22.04
1303
884
Exposures
Current noise
(dBA)
level 77 (69–86)
* Asterisks indicate variables that met the significance level criterion (p<0.0001) for differences
between groups. MDC= Minimum Detectable Concentration.
Testing the Auditory System
A bus, converted into a mobile laboratory, was equipped and had two soundproof booths installed
that met the requirements of ANSI S 3.1, 1991 for audiometric testing environments. Otoscopy
was performed to screen for conditions that would exclude a person from the study, i.e., excessive
cerumen, external otitis or perforated tympanic membrane.
The test battery was chosen to include tests of the different parts of the auditory system. Damage
to the peripheral auditory system was tested with pure tone audiogram (PTA). PTA is sensitive to
cochlear impairments and it is also the classic measure of the status of hearing system needed for
comparison with other investigations. The outer hair cells of the peripheral auditory system was
also investigated with otoacoustic emissions (OAEs) and with psycho-acoustical modulation
transfer function (PMTF) a test that shows the ability of the ear to follow intensity modulations of
various frequencies.
The central auditory pathways were investigated through speech discrimination tests and cortical
auditory evoked responses.
The results of all tests in the exposed groups were compared with the control group included in
this study and also with normal reference values, if these were available, as specified.
82
June 2004
NoiseChem 83
Equipment
The equipment used for pure-tone, speech audiometry and psycho-acoustical modulation transfer
function is a Technical Audiological Measurement Processor (TAMP3) which is constructed at
the unit of Technical Audiology of the Karolinska Institutet. It was controlled by a personal
computer. TAMP3 was based on the signal processor Texas TMS32010 with a 96 kB memory,
A/D- and D/A-converters, a real time clock, antialiasing filters, controllable attenuators,
amplifiers with controllable gain and an output amplifier suited for the headphone type TDH-39
with MX41AR cushion. Software was developed for calibration, pure tone audiogram (ascending
manual, ascending automatic, fixed frequency Békésy), psycho-acoustical modulation transfer
function (PMTF).
The equipment used to measure the distortion product otoacoustic emissions (DPOEs) was based
on units from Tucker-Davis Technologies, controlled by a personal computer with software
developed at the unit of Technical Audiology. The system included a signal processing board,
microphone probe, signal sources and preamplifier. The microphone (Etymotic Research ER-10)
incorporated the signal delivery and microphone into a small package that was inserted in the
subject's ear canal using a soft foam ear tip.
Basically the same equipment from Tucker & Davis is used for measurement of the DPOEs is
used also for cortical response audiometry (CRA), with the addition of a preamplifier Entomed
510. A separate software program was used for the CRA.
The speech materials were stored on the computer hard disc that was connected, through a
controllable attenuator, to the output amplifier. The latter two were parts of the TAMP3
equipment. The output level 70 dB SPL was set by a computer program (calibrated in a 6 cc
coupler). The speech lists were presented through a TDH-39 headphone. For speech recognition
in noise an extra controllable attenuator and a mixer adding the noise to the speech signal was
added to the output amplifier.
Pure-Tone Audiometry (PTA)
Pure-tone thresholds were measured with the fixed frequency Békésy method for both ears at the
frequencies 1, 1, 2, 3, 4, 6 and 8 kHz. The first threshold at 1 kHz served as training and was not used for
the analysis. Most of the following measurements were performed only on the best ear, chosen by
comparison between the ears of the mean values of the thresholds between 1 and 8 kHz. Mean thresholds
were compared between groups as reported in Morata et.al. (2002). To evaluate the effects of age further
the individual thresholds were also compared to a new validated material of an otological unscreened, nonoccupational noise-exposed population (n=603; age 20-79 years) from Sweden (Johansson & Arlinger
2002).
Psycho-acoustical modulation transfer function (PMTF)
The psycho-acoustical modulation transfer function (PMTF) shows the ability of the ear to follow
intensity modulations of various frequencies, which is essential for speech recognition.
The threshold of a 4 ms brief tone, centered in a fluctuating octave filtered noise was measured.
The threshold of the probe tone was measured also without noise (i.e. a brief-tone threshold).
Separate thresholds were measured with the tone at the peaks and in the valleys of the modulated
noise. The signal-to-noise ratio at threshold was used as the resulting measure. The thresholds
were determined with an adaptive method (Békésy-technique). Threshold for the brief tones was
measured at noise levels from 25 to 95 dB SPL in steps of 10 dB. The noise consisted of the
octave-band around the test tone of 4000 Hz, with a sinusoidal 100 % intensity-modulation of 10
Hz. To prevent listening to sounds outside the octave-band, a masking noise of a faint, periodic
broadband noise was added. The degree of accuracy has been found to be about 2 dB for peak
and valley thresholds (Lindblad et al. 1992). An example of the outcome is shown in Figure 1. A
level dependence of the PMTF has been shown such that the best ability (i.e. the lowest
83
NoiseChem 84
June 2004
threshold) normally is found at noise levels in the range of normal speech levels, i.e. 55-65 dB
SPL. For hearing impaired subjects the best ability occurs at higher levels.
The brief-tone threshold, the top value of the peak threshold curve and the top value of the valley
threshold curve were used as outcome measures for the PMTF (see Figure 1).
Octave band frequency 4 kHz, modulation noise 10 Hz
Threshold
dB S/N
Top value of the
peak threshold curve
22
20
18
16
Top value of the
valley threshold curve
14
12
10
8
Thresholds measured;
at the peak of the noise the peak threshold curve
6
4
2
in the valley of the noise the valley threshold curve
0
-2
-4
-6
-8
-10
24 dB
25
35
45
55
65
75
85
95 105
Noise level dB SPL
Figure 1. An example of outcome measures from the psycho-acoustical modulation transfer
function (PMTF) recording. The test signal was a 4ms tone of 4000 Hz, centered in a
fluctuating octave filtered noise (100 % intensity-modulation, 10 Hz). Separate thresholds
(signal-to-noise ratio) were measured with the tone at the peaks and in the valleys of the
modulated noise, using an adaptive method. The level of the noise varied from 25 to 95 dB SPL
in steps of 10 dB.
Distortion product otoacoustic emissions, (DPOAEs)
Otoacoustic Emissions (OAEs) are a naturally occurring function of a healthy cochlea. OAEs are
considered as a test of the function of the outer hair cells in the cochlea (Shaffer et.al 2003). Two
types of evoked OAEs are commonly used in clinical settings, transient-evoked otoacoustic
emissions (TEOAEs) and distortion product otoacoustic emissions (DPOAEs). DPOAEs are
produced through the presentation of two tonal signals to the ear and measuring the distortion
product that is produced by the cochlea. In a healthy cochlea, the DPOAE can easily be
measured in the frequency range 1 - 8 kHz. When damage occurs due to, for instance, noise
exposure, the ability of the cochlea to produce DPOAEs is reduced.
Two continuous tones were presented to the ear at frequencies f1 and f2 where the ratio of the
frequencies is maintained at f2/f1=1.225. The level of f2 was 10 dB lower than the level of f1.
The tones were presented for 4.3 seconds and the frequency spectra of the responses were
averaged to produce a resultant spectrum of the ear canal signal. The computer controlled the
84
June 2004
NoiseChem 85
generation and sampling from the respective sources and microphone. A spectrum was calculated
and the software measured the energy at the distortion product frequency, 2f1-f2, and reported the
DPOE amplitude associated with the geometric mean (GM) frequency (f1 * f2)1/2 of the two
primary tones. Several averages were measured until a stable value was obtained. A response was
considered to be present when the DPOE amplitude was determined to be at least 3 dB above the
background noise floor.
The subject was instructed to sit quietly for the duration of each test sequence. The microphone
was fitted in the subject's ear canal and tested for an acceptable placement. The input-output
function of DPOAE was collected with f1 at the frequency 4 kHz for overall input levels 35 to 80
dB SPL in steps of 5 dB.
Cortical response audiometry (CRA)
Cortical response audiometry, CRA, tests the central pathways of the auditory system. CRA has
been shown to yield abnormal results in subjects with cerebello-pontine angle tumours but may
also be sensitive to lesions on higher levels of the auditory pathways. Ödqvist et al. (1987)
showed significantly abnormal results of CRA with frequency glides used as stimuli for a group
of subjects exposed to various solvents.
The stimulus was based on a 1000 Hz continuous pure tone at 60 dB HL. The actual stimulus is
the frequency glide of this tone. Its frequency is linearly increased 50 Hz in 20 ms. Then it stays
at 1050 Hz for 480 ms and finally returns linearly to 1000 Hz during 600 ms. The inter-stimulus
interval is varied randomly from 2 s and up with a mean of 4 s. Fifty 500 ms sweeps are summed
in the computer to obtain an average response. If the hearing threshold at 1000 Hz is raised to 20
dB HL or more the pure tone is amplified according to the "half gain rule", e.g. for a hearing
threshold of 40 dB HL the level 60 + 40/2 = 80 dB HL will be used. However, the highest level
permitted is set to 100 dB HL. The latency of the cortical response was used as outcome
measures.
Interrupted Speech (IS)
Various methods of distorted speech have been used to test the central hearing pathways and the
auditory cortex. Korsan-Bengtsen (1973) has shown that the most efficient distortion for this
purpose is to interrupt the speech signal with 7 interruptions/s.
Korsan-Bengtsens (1973) lists with 25 sentences containing 100 key words each were used. Five
sentences from one separate list were presented at 70 dB SPL for familiarization to the test. If the
subject asked for a higher presentation level, further sentences were presented at higher levels
until an appropriate level was found. Then one test list was presented. The key words correctly
repeated were counted and a score was given as the test result.
The results were considered below mean if the discrimination score was below 93 % and
abnormal when the score was below 78 % (mean minus 3 standard deviations). The normal
reference group used for comparison was adapted after Korsan-Bengtsen (1973) who gives the
mean of 93 % correct responses for older controls (mean age 55 years).
Speech recognition in Noise (SiN)
Speech recognition in noise is a very difficult test, with great demands on the whole hearing
system. It also mimics the difficult listening situation often met in real life for a person with
possible hearing difficulties.
An adaptive procedure has been developed for measuring the speech reception threshold in noise
(Hagerman, 1995). A fixed speech level of 70 dB SPL was used. The noise level was changed a
specified number of decibels after each sentence depending on the number of correct answers at
the sentence. A computer program executed the changes when the number of correct answers was
entered into the computer. The computer also calculated the final result as the signal to noise ratio
(S/N) when 40% correct answers were obtained.
85
June 2004
NoiseChem 86
Three lists of Hagerman's sentences in noise (Hagerman, 1982, 1984) were used. One list was
used as training to familiarize the subject with the test. If the subject asked for a higher
presentation level than 70 dB SPL, further sentences were presented at higher levels until an
appropriate level was found. Finally two test lists were run. Each list consisted of 10 sentences
with 5 words, i.e. altogether 50 words. Each word was scored.
The results were considered abnormal if the S/N exceeded –7,8 dB when compared to a normal
reference group of 10 normal hearing subjects, median age 27, (range 23 – 30) (Hagerman, 1995).
Data Analyses
In order to investigate the principal hypothesis, the factor “exposure group” with four levels
(styrene & noise, styrene only, noise only and control) was tested against the dependent variables
using one-way between-subjects analysis of variance (ANOVA), followed by pair-wise contrasts.
The dependent variables of interest were: Pure-tone audiometry thresholds, outcome measures in
the PMTF test, DPOEs, CRA latency, interrupted speech and speech in noise results.
The pure tone audiometry thresholds were also compared to a Swedish database (Johansson &
Arlinger 2002) using the following method. The proportion of persons with thresholds greater
than the median (50th percentile) and 90th percentile reference values were calculated. Separate
calculations were done for best and worst ears at each frequency for each of the exposed groups
(noise, styrene and styrene & noise), t-tests were done to determine if the proportions were
significantly different from 0.50 for the median and 0.10 for the 90th percentile.
Correlation between questionnaire and exposure variables and the dependent test variables was
made with Pearson bivariate correlations.
The data were analyzed using the Statistical Analysis System (Pre-Production Version 9.00,
SAS®, SAS Institute Inc, Cary, NC) and the SPSS statistical system (version 11.5, SPSS Inc
Chicago, Ill.
Results
Questionnaire data
One-way ANOVAs compared the means of the response scores on the medical history,
occupational and non-occupational exposure, lifestyle factors, and present health and tested for
differences between the four groups. The questions included data on smoking, diabetes, prior ear
surgery, head injury, high fever, measles, high blood pressure, mumps, ear infections, history of
hearing loss in the family, ototoxic medication use, and tinnitus. The only variable on which
groups differed statistically was tenure (p<0001, see Table 1), and the noise-exposed workers had
the shortest tenure. Influence of military service was investigated by comparing 64 cases (no)
with randomly selected 64 cases (yes). Point biserial correlations between the categorical variable
military (1 = yes, 0 = no) and hearing thresholds at 2 and 4 kHz were not significant (two- tailed
p. > .1).
Exposure assessment
Noise exposures exceeded recommended limits for 130 of the 313 studied workers. Styrene
exposure levels never exceeded the Swedish recommended limits, which are among the world’s
lowest (90mg/m3 or 20 ppm).
PTA
There was a higher prevalence of high-frequency hearing loss in the groups exposed to noise and
styrene simultaneously (48%) and exposed to styrene alone (47%), compared to other groups:
33% in the non-exposed group and 42% in the noise-exposed group.
In each group, mean thresholds at each frequency were calculated for each ear. Results are shown
in Figure 2. Significantly higher thresholds at 2, 3, 4, and 6 kHz were observed in the styreneexposed workers in both ears, compared to both or one of the other groups.
86
NoiseChem 87
June 2004
Threshold (dB HL)
0
Control
-5
-10
Noise
**
*
Styrene
*+
-15
Styrene & Noise
*+
-20
-25
** ++
** ++
Best ear
-30
1
2
3
4
6
8
Frequency (kHz)
Figure 2. Group mean hearing levels (dB HL) for the better ear at frequencies between 1 and 8
kHz. Asterisks (*) indicate frequencies that met the significance level criterion (p<0.05) when
exposed groups were compared to non-exposed workers, and the plus sign (+) indicate
frequencies that met the significance level criterion (p<0.05) when styrene exposed groups
were compared to noise-exposed workers.
Age-related comparison with the 10th, the median and the 90th percentiles of the Swedish database
are shown in Figure 3 and 4 for 4000 and 6000 Hz respectively.
The comparison of the proportions of persons in each exposed group differing from the median
showed significantly greater proportions than expected in both the styrene exposed groups for the
worst ear at 4000 Hz (p< 0.001) and 8000 Hz (p< 0.01), for both ears at 6000 Hz (p< 0.001). In
the group exposed to noise alone significantly greater proportions compared to the median were
found for the worst ear at 6000 Hz (p< 0.001) and for both ears at 8000 Hz (p< 0.05), see Table 2.
Comparison with the 90th percentile showed significantly greater proportions for the styrene
exposed group for the worst ear at 6000 Hz (p< 0.05) and for the styrene & noise group and the
noise alone group for the worst ear at 8000 Hz (p< 0.05), data not shown.
87
NoiseChem 88
June 2004
4000 Hz - Worst Ear
Noise
40
40
0
0
20
20
Threshold (dB)
60
60
80
Control
30
40
50
60
20
40
50
60
Styrene & Noise (N > 85 dBA)
40
40
20
0
0
20
20
Threshold (dB)
60
60
80
Styrene (N < 85 dBA)
30
30
40
Age (years)
50
60
30
40
50
60
Age (years)
Figure 3. Age-related individual thresholds (circles, dB HL) of the worst ear at 4000 Hz are
shown in separate graphs for different groups. Lines show the 10th, the 50th
and the 90th percentiles of an otological unscreened, non-occupational noise-exposed population
(n=603; age 20-79 years) from Sweden (Johansson & Arlinger 2002).
88
NoiseChem 89
June 2004
6000 Hz - Worst Ear
Noise
40
20
40
0
0
20
Threshold (dB)
60
60
Control
30
40
50
60
20
50
60
Styrene & Noise (N > 85 dBA)
60
80
40
60
20
40
0
0
20
Threshold (dB)
20
40
80
Styrene (N < 85 dBA)
30
30
40
Age (years)
50
60
30
40
50
60
Age (years)
Figure 4. Age-related individual thresholds (circles, dB HL) of the worst ear at 6000 Hz are
shown in separate graphs for different groups. Lines show the 10th, the 50th and the 90th
percentiles of an otological unscreened, non-occupational noise-exposed population (n=603;
age 20-79 years) from Sweden (Johansson & Arlinger 2002).
89
NoiseChem 90
June 2004
Table 2. Proportion (%) of each group with hearing levels (dB HL) above the median level of
age-correlated hearing levels from a database of a normal population in Sweden.
3000 Hz
4000 Hz
6000 Hz
8000 Hz
worst
worst
best worst
best worst
best
best
Group
ear
ear
ear
ear
ear
ear
ear
ear
Control
none none
none 4.6%
14.0% * 18.4% ** none
14.0% **
24.0%
Noise
none none
none 3.8%
1.3%
***
11.8% * 15.3% **
Styrene
(N<85 dB)
none
8.5%
none
31.2%
17.7% * 35.6% ** ***
9.3%
15.6% *
Styrene & Noise
none
(N>85 dB)
9.5%
none
32.0%
19.7% * 21.9% *** ***
none
20.8% ***
Asterisks (*) indicate significant higher proportions than expected compared to the median of the
Swedish database (Johansson & Arlinger, 2002). * p<0.05, **p<0.01, ***p<0.001
Re-testing of 30 workers
Neither the styrene levels nor the noise measurements results show any remarkable differences
on the two different occasions. Six (20%) of the 30 workers re-tested after 3 years showed a
significant (more than 10 dB HL) worsening of their auditory thresholds in at least one test
frequency. In only one worker this occurred in more than one frequency.
PMTF
A difference in hearing threshold, measured with a short tone at 4 kHz as stimuli, was shown
between the group exposed to styrene and noise compared to controls and to the group exposed to
noise alone. In the comparison between groups on the threshold in modulated noise a significant
difference was shown on the top value of the peak threshold curve where the group exposed to
noise alone had a lower threshold compared to controls. Means and standard deviations are
presented in Table 3.
DPOAE
A one-within, one-between ANOVA with factors exposure group and signal level (35 – 70 dB in
5-dB steps) was performed on DPOAE scores. As expected, the main effect of signal level was
highly significant (F (2.02, 34.32) = 81.34, p < .001; MSE = 48.02). While the main effect of
exposure group failed to reach significance, the interaction between exposure group and signal
level was significant (F (21, 119) = 2.12, p = .006; MSE = 13.85; see Figure 5).
In figure 5 it is shown that for low signal levels (up to 50 dB), control and noise groups are
showing somewhat higher DP levels compared to the two groups exposed to styrene However,
past 50 dB, the control group levels off in contrast to all three “exposed” groups.
The mean DP at higher levels (55-70dB) was significantly correlated to the hearing thresholds at
4 kHz in the best ear of the controls and the group exposed to noise. The groups exposed to
styrene did not show correlation with the hearing thresholds.
90
NoiseChem 91
June 2004
Mean DP (dB)
20
10
0
Exposure group
-10
-20
control
noise
styrene
styrene & noise
35
40
45
50
55
60
65
70
Input signal level F1 (dB)
Figure 5. Mean Distortion Product (DP) scores for the four exposure groups, plotted against
input signal level (F1). Note that at low input signal levels (up to 50 dB), control and noise
groups have higher DP levels compared to the two groups exposed to styrene, while, above 50
dB, the control group levels off in contrast to all three “exposed” groups.
CRA
There was a highly significant effect of exposure group on the latency of the cortical evoked
response (F (3, 263) = 3.48, p = .016; MSE = 360.70). The observe d significant effect permitted
a post-hoc analysis to be performed on the data. Tukey’s HSD carried out on the means for the
four exposure group levels revealed a significant difference on the mean CR latency scores
between the control and the styrene group and the styrene and noise group (p <0.05; Figure 6,
Table 3). The latency of the CRA was also correlated to the current noise exposure (r = .146, p =
.018; N = 266).
Figure 6. Mean latency scores (+- 2
S.E.M.) of the cortical evoked response
are plotted for each of the four exposure
groups (see abscissa label). The mean
latencies for the group exposed to
styrene+noise (1) and the group exposed
to styrene (2) were significantly different
from the latency in the controls (4).
91
June 2004
NoiseChem 92
Figure 7. Mean S/N ratio (+- 2 S.E.M.) from the Speech in Noise test for the four exposure
groups (see abscissa label). The average S/N ratio for the control group (4) was significantly
lower compared to the three exposed groups.
Speech tests
Speech recognition in noise
There was a highly significant effect of exposure group on the signal to noise (S/N) ratio of the
Speech in noise test (F (3, 278) = 7.85, p < .001; MSE = 8.71). A post-hoc test was performed on
the S/N data. (Tukey’s HSD) revealed a significant difference between the control condition and
the three exposed conditions (p < .005; see Figure 7 and Table 3). The same significant results
were found for all exposed groups when comparing them to normal reference values of –7.8 S/N.
The results of the speech in noise test were correlated to the current noise exposure (r = .124, p =
.037; N = 281).
Interrupted speech
There was no significant difference between the groups on the mean scores of the interrupted
speech test. A tendency towards a lower score was seen in the group exposed to styrene alone
compared to the controls (p<0.06). The groups were also compared to normal reference values
using the classification criteria of normal values being above the median of 93 % correct
responses and of 78 % as an abnormal value. A significant lower percentage of both groups
exposed to styrene were found above 93 % (p < 0.05), and a significant higher percentage of both
groups exposed to styrene were found below 78 % (p < 0.05) as seen in Figure 8 and Table 3. A
significant correlation was seen in the groups exposed to styrene between the cumulative styrene
exposure in the current company and a lower score of the interrupted speech (r=-0.190, p<0.023;
N = 143). The result of interrupted speech test was also correlated to the speech in noise test (r = .378, p < .001).
92
NoiseChem 93
June 2004
Sheet #1
Layer #1
1
X: 6.76in
Y: 6.09in
X: 1.99in
Y:02.21in
%
4.55in
4 Group of 4 objects
Figure 8. Percentage (%) of each group that
showed scoring values above (white) and
below (grey) 93% correct responses. Black
segments show the percentage (%) of each
group that had scoring values below 78%. A
significantly higher percentage had scoring
values below both 93% and 78 % was found
in the styrene group and in the styrene &
noise group compared to the control group.
93% is the mean of correct responses and
78% (= mean minus 3 standard deviations) is
considered abnormal (reference group of
Korsan-Bengtsen, 1973).
93
NoiseChem 94
June 2004
Table 3. Means and standard deviations of outcome measures in differentauditory tests. Significant differences between controls and exposed groups are indicated.
GROUP
PMTF
CRA
top of peak top of valley
4kHz threshold threshold (dB threshold
Latency
(dB SPL)
S/N)
(dB S/N)
ear (ms)
Interrupted speech
Speech in noise
right Latency left % correct % of group % of group dB S/N %
above
ear (ms)
answers
above 93% below 78% ratio
-7.8 S/N
Controls
11.9±12.6
18.8±5.6
12.5±6.6
134.5±22.2
134.8±20.4
90±13,4
47
6
-7.5
92
Noise
13.4±17.1
17.0±5.2a
11.3±5.8
141.0±20.2
142.5±25.8a 87±18.3
51
14
-5.3a
42a
Styrene
(N<85 dB)
16.6±19.7
18.2±5.9
12.7±6.3
144.6±22,8a
136.5±20.2
32a
20a
-5.6a
70a
17.3±19.0a,b
18.3±5.3
12.4±6.6
138.2±24,2
147.7±25.5a 88±9.3
35b
17a
-5.3a
82
85±13.5
Styrene &
Noise
(N>85 dB)
a = p< 0.05 compared to control group, b = p< 0.05 compared to noise group, PMTF = Psycho-acoustical modulation transfer function, CRA = Cortical
Response Audiometry, S/N = Signal to noise ratio
94
June 2004
NoiseChem 95
Discussion
In the present investigation styrene has been shown to affect the auditory system of exposed workers, despite
the low levels of measured styrene at the studied workplaces. This confirms similar results observed in
previous publications (Morata et al., 2002, Morioka 1999, 2000, Sliwinska-Kowalska et al., 2003).. We
reported in 2002 (Morata et al., 2002) that there was a higher prevalence of high-frequency hearing loss in
the groups exposed to noise and styrene simultaneously (48%) and exposed to styrene alone (47%),
compared to other groups: 33% in the non-exposed group and 42% in the noise-exposed group. In each
group, mean thresholds at each frequency were calculated for each ear. Significantly poorer thresholds at 2,
3, 4, and 6 kHz were observed in the styrene-exposed workers in both ears, compared to one or both of the
two groups not exposed to styrene.
Since the publication of the 2002 report, a database containing the hearing threshold levels of an otologically
unscreened population in Sweden became available (Johansson & Arlinger, 2002), along with a new
mathematical model of hearing threshold levels as a function of age. We used this information to re-examine
the audiometric data we had analyzed before using different methods. The new analysis confirmed our
previous conclusions in indicating that styrene exposure was associated with significantly poorer thresholds
at several of the test frequencies.
In the present report we have observed not only the occurrence of audiometric effects in the styrene exposed
groups, but also information on the sites of the disorders and the potential contribution of different
audiological tests in the detection of the styrene effects.
Control group
The prevalence of hearing loss in the non-exposed control group was above 30 %, which might seem higher
than expected. However, even if this does not bias the results of the group comparisons in the direction of
finding greater significant results it still needs to be commented. The high prevalence of hearing loss may be
due to that some of the controls had had previous jobs where noise exposure was present. The mean tenure in
these jobs was, however, not different from any of the exposed groups. The calculated life-time noise
exposure was also lower among the controls compared to the exposed groups. The results from the control
group show the difficulty of finding a suitable control group among blue-collar workers. To compensate for
this we also used comparison to other control material and databases when they became available.
Re-testing part of the population
Twenty percent of the re-tested workers showed a significant threshold shift in at least one test frequency. In
only one of these the greater loss of HL was seen in more than one frequency. The exposure measurements
of styrene and noise in the work place did not differ between the occasions. Studies that have investigated the
possibility of using the percentage of significant threshold shifts as an evaluation criterion to the
effectiveness of hearing loss prevention programs have reported that 3% to 6% (Morrill and Sterrett, 1981)
or 5% significant threshold shifts (Franks et al. 1989; Simpson et al. 1994) are reasonable incidence rates that
can be met by effective programs. Significant threshold shift incidence rates exceeding these percentages
suggest that preventive practices in place are not being effective (NIOSH, 1998). The present findings
suggest that the studied agents are still having an effect in the workers’ hearing.
Location of damage
Peripheral auditory system effects were detected not only through pure-tone audiometry, but also on the
results of distortion product otoacoustic emissions. Exposure to styrene has been associated with disruption
in outer hair cell function as shown by DPOAEs in animals (Pouyatos, Campo, Lataye, 2002; Lataye et al.,
2003). The authors indicated that DPOAEs could be used to monitor the ototoxicity induced by styrene even
though they cannot be considered as the most sensitive index of styrene cochlear pathology, when compared
to evoked potentials. Reduced amplitudes of transient evoked and distortion-product otoacoustic emissions
have also been reported among solvent exposed workers (Sulkowski et al, 2002). The results of the inputoutput function in this study show different results for both the groups exposed to styrene compared to both
controls and noise exposed workers. At low input signal levels (35-50 dB) the styrene groups showed lower
DP responses compared to controls and noise exposed, whereas at higher input signal levels the control
showed lower DP responses compared to all exposed groups. Hypothetically this could mean that styrene
95
June 2004
NoiseChem 96
affects the outer hair cell (OHC) function. The function of OHCs and their role as the cochlear amplifier is
supposed to play a greater role in the formation of the DPOEs at lower stimulation levels whereas other
mechanisms such as linear coherent reflection influence the responses at higher stimulation levels more
(Shaffer et.al 2003). However, this hypothesis is very speculative since recent research shows that more finetuned measurements than used in this study are needed to detect differences in how the responses of the
DPOAEs are produced (Shaffer et.al 2003).
The sensitized speech tests used and the evoked cortical responses to frequency glides (CRA) are relatively
insensitive to peripheral auditory lesions, and are indicators of retrocochlear and central dysfunction. Clinical
studies with workers with a long history of solvent exposure, and in some case diagnosis of psycho-organic
syndrome have detected significant abnormality in these test results (Odkvist et al., 1982, 1987; Moller et al.,
1989, 1990). Also Laukli & Hansen (1995) found abnormal results only in the central tests when using an
extensive test battery for evaluating auditory effects after occupational exposure to solvents. In the present
investigation, abnormal results of the interrupted speech and speech in noise test have also shown to be
associated with styrene exposure. Sensitized speech tests are known to be sensitive to cortical lesions,
however brainstem lesions can also produce reduced scores (Korsan-Bengtsen, 1973).
Occupational studies conducted on another ototoxic solvent, toluene, suggested a retrocochlear or
central auditory pathway involvement in some of the hearing disorders observed, based on the
results of acoustic reflex decay test (Morata et al., 1993; 1997).
The combined results of the tests performed in the present study suggest a location of the styrene damage
both in the peripheral and the central portions of the auditory system.
Central auditory tests involving verbal responses require a number of neural functions such as attention,
intensity or pitch discrimination, recognition, immediate memory, and memory for sounds. The
alterations observed in the present study offer clues to the type of the damage incurred in the individuals
exposed to solvents. It is conceivable that the cortical area that supports language ability in the temporalparietal region of the left hemisphere may be affected, as has been previously suggested in the literature
(Efrom, 1963; Musiek et al., 1980; Phillips, 1993). However, these tests do not give information on exact
sites or which hemisphere was affected. Thus, the issue of lesion site remains open for future studies.
Usefulness of test battery to complement pure tone audiometry
Several solvents commonly used in the workplace have been shown to have ototoxic properties. Noise is
often present in occupational setting and it can be rather challenging to distinguish the effects of these
physical and chemical agents.
The majority of investigations on occupational hearing loss have relied on averaging pure tone thresholds to
assess noise effects on auditory function. To investigate the effects of chemical exposure, this traditional
approach may not be sufficient. Approaches that are more robust indicators of the risk posed by occupational
chemical exposures on hearing include: classification of audiometric results using specific criteria and
subsequent estimation of prevalence or incidence rates and relative risk from analyzing hearing as a binary
variable (normal vs. high frequency hearing loss). Some examples of these alternatives were discussed
previously by Morata and Lemasters (1995). The prevalence of hearing loss between groups with different
exposure conditions should be examined even if audiometric thresholds, by themselves, do not allow for easy
identification of the effect of chemicals on hearing, and especially when pure-tone audiometry is the only
available test.
Audiometric findings associated with exposure to solvents reveal mild to moderate hearing losses. However,
despite a mild audiometric effect, the hearing loss from solvents may significantly impact the individuals
ability to communicate, once we consider that the effects of the solvents are not only restricted to the
peripheral auditory system. To better understand and detect the effect of these chemicals and to elucidate
effects of chemicals from effects from noise it is necessary to identify the audiological tests that are sensitive
to these effects.
In the present study, even though some abnormalities were noted on the DPOAEs and CRA testing, the tests
that seemed the most sensitive to the effects of styrene were interrupted speech and speech recognition in
noise. These tests are of easy and quick administration, they are not invasive, they allow for a distinction
from the effects of noise, and therefore are recommended for use with solvent exposed populations.
96
June 2004
NoiseChem 97
Conclusion
The results of the present study show that occupational exposure to styrene affects the auditory system even
when the exposure levels are low. The extensive test-battery showed that both the peripheral and the central
auditory system were affected by these exposure conditions. These results show the need to make workers
and the occupational health community aware of the potential risk of chemically induced hearing loss and to
recommend the inclusion of chemical-exposed workers in hearing loss prevention programs, even when the
noise levels are below recommended exposure limits.
References
Campo P., Loquet G., Blachere V. and Roure M (1999). Toluene and styrene intoxication route in the rat
cochlea. Neurotoxicol Teratol 21(4), 427-434
Campo P., Pouyatos B., Lataye R. and Morel G. (2003). Is the aged rat ear more susceptible to noise or
styrene damage than the young ear? Noise Health 5(19), 1-18
Efrom R. (1963) Temporal perception, aphasia and deja vu. Brain, 86: 404-424.
Franks JR, Davis RR, Krieg EF Jr. (1989). Analysis of a hearing conservation program data base: factors
other than workplace noise. Ear Hear 10(5):273-280.
Hagerman B. (1982). Sentences for testing speech reception threshold in noise. Scand Audiol 11, 79-87.
Hagerman B. (1984). Clinical measurements of speech reception threshold in noise Scand Audiol 13, 57-63.
Hagerman B. (1995). Efficient adaptive methods for measurements of speech reception thresholds in quiet
and in noise Scand Audiol 24, 71-77.
Johansson M.S.K.and. Arlinger S.D., Hearing threshold levels for an otologically unscreened, nonoccupationally noise-exposed population in Sweden. Int. J. Audiol. 2002; 41,180–194
Korsan-Bengtsen M. (1973). Distorted speech audiometry. Acta Otolaryngol (Sthlm) Suppl.310.
Lataye R., Campo P. and Loquet G. (2000). Combined effects of noise and styrene exposure on hearing
function in the rat. Hear Res 139, 86–96.
Lataye R., Campo P., Pouyatos B., Cossec B., Blachere V. and Morel G. (2003). Solvent ototoxicity in the
rat and guinea pig. Neurotoxicol Teratol 25(1), 39-50.
Laukli E. and Hansen P,W. (1995). An audiometric test battery for the evaluation of occupational exposure
to industrial solvents. Acta Otolaryngol 115,162-164
Lindblad A.-C., Olofsson Å. and Hagerman B. (1992). Tone thresholds in modulated noise. I. Level
dependence and relation to SRT in noise for normal-hearing subjects. Karolinska Institutet.
Makitie A.A., Pirvola U., Pyykko I., Sakakibara H., Riihimaki V. and Ylikoski J. (2003). The ototoxic
interaction of styrene and noise. Hear Res 179(1-2), 9-20
Möller C., Ödkvist L., Larsby B., Tham R., Ledin T. and Bergholtz L.M. (1990). Otoneurological findings in
workers exposed to styrene. Scand J Work Environ Health 16, 189-194.
Möller C., Ödkvist L.M., Thell J., Larsby B., Hydén D., Bergholtz L.M. and Tham R. (1989).
Otoneurological findings in psycho-organic syndrome caused by industrial solvent exposure: Acta
Otolaryngol (Stockh) 107, 5-12.
Morata T.C., Dunn D.E., Kretschmer L.W., Lemasters G.K. and Keith R.W. (1993). Effects of occupational
exposure to organic solvents and noise on hearing. Scand J Work Environ Health 19, 245-254.
Morata T.C., Fiorini A.C., Fischer F.M., Colacioppo S., Wallingford K.W., Krieg,E.F., et al. (1997).
Toluene-induced hearing loss among rotogravure printing workers. Scand J Work Environ Health 23(4),
289–298.
Morata T.C., Johnson A.C., Nylen P., Svensson E.B., Cheng J., Krieg E.K., Lindblad A.C., Ernstgård L.,
Franks J. (2002). Audiometric Findings in Workers Exposed to Low Levels of Styrene and Noise. The
Journal of Occupational and Environmental Medicine. 44(9), 806-814
Morata, T. and Lemasters, G (1995). Epidemiologic considerations in the evaluation of occupational hearing
loss. in Occupational Medicine: State of the Art Reviews. Eds. T. Morata and D. Dunn. 10, 641-656.
Morrill JC, Sterrett ML (1981). Quality controls for audiometric testing. Occup Health Saf 50(8):26-33.
Morioka I., Kuroda M., Miyashita K. and Takeda S. (1999). Evaluation of organic solvent ototoxicity by the
upper limit of hearing. Arch Environ Health 54(5), 341–346.
97
June 2004
NoiseChem 98
Morioka I., Miyai N., Yamamoto H. and Miyashita K. (2000). Evaluation of combined effect of organic
solvents and noise by the upper limit of hearing. Ind Health 38 (2), 252–257.
Muijser H., Hoogendijk E.M. and Hooisma J. (1988). The effects of occupational exposure to styrene on
high-frequency hearing thresholds. Toxicology 49, 331-340.
Musiek F.E., Pinheiro M.L. and Wilson D. (1980) Auditory Pattern Perception in “split brain” patients. Arch
Otolaryngol, 106, 610-612.
National Institute for Occupational Safety and Health, NIOSH (1998). Criteria for a recommended
standard. Occupational exposure to noise. Revised Criteria. Cincinnati: USDHHS, PHS, CDC, NIOSH,
publication no.98-126.
Ödkvist L.M., Arlinger S.D., Edling C., Larsby B. and Bergholtz L.M. (1987). Audiological and vestibulooculomotor findings in workers exposed to solvents and jet fuel. Scand Audiol 16, 75-81.
Ödkvist L.M., Bergholtz L.M., Åhlfeldt H., Andersson,B., Edling C. and Strand E. (1982). Otoneurological
and audiological findings in workers exposed to industrial solvents. Acta Otolaryngol (Stockh) (Suppl. 386),
249-251
Phillips D. P. (1993) Central auditory processing a view from auditory neuroscience. Am J Otol, 16(3): 338352.
Pinheiro M.L. and Musiek F. E. (1985). Assessment of Central Auditory Dysfunction: Foundation and
Clinical Correlates. Baltimore: Williams & Wilkins, pp. 219-238.
Pouyatos B., Campo P. and Lataye R. (2002). Use of DPOAEs for assessing hearing loss caused by styrene
in the rat. Hear Res 165(1-2), 156-64.
Pyykkö I.V., Toppila E.M., Starck J.P., Juhola M. and Auramo Y. (2002). Database for a hearing
conservation program. Scand Audiol 29(1), 52–58.
Sass-Kortsak A.M., Corey P.N. and Robertson J.M. (1995). An investigation of the association between
exposure to styrene and hearing loss. Ann Epidemiol 5(1), 15–24.
Shaffer L.A., Withnell R.H., Dhar S., Lilly D.J., Goodman S.S. and Harmon K.M. (2003). Sources and
mechanisms of DPOAE generation: implications for the prediction of auditory sensitivity. Ear Hear. 24(5),
367-379.
Simpson TH, Stewart M, Kaltenback JA (1994). Early indicators of hearing conservation program
performance. J Am Acad Audiol 5:300-306.
Sliwinska-Kowalska M., Zamyslowska-Szmytke E., Szymczak W., Kotylo P., Fiszer M., Wesolowski W.,
Pawlaczyk-Luszczynska M. (2003). Ototoxic effects of occupational exposure to styrene and co-exposure to
styrene and noise. J Occup Environ Med 45(1), 15-24
SulkowskiW.J., Kowalska S., Matyja W., Guzek W., Wesolowski W., Szymczak W., Kostrzewski P. (2002).
Effects of occupational exposure to a mixture of solvents on the inner ear: a field study. Int J Occup Med
Environ Health. 15(3), 247-256.
98
June 2004
NoiseChem 99
Lab 2: Krystyna Pawlas
Instutite of Occupational Medicine and Environmental
Health, Poland
99
June 2004
NoiseChem 100
Hearing and balance of workers exposed to different scenarios of noise and organic solvents
Pawlas Krystyna 1), Pawlas Natalia 2), Brużewicz Szymon 3), Noga Leszek 3), Wojciech Mniszek 1) , Marzena
Zaciera 1), Ewa Smolik 1)
1) Institute of Occupational Medicine and Environmental Health Sosnowiec
2) Department of Pharmacology. Medical University of Silesia, Katowice
3) Wroclaw Medical University, Wroclaw
Introduction
Factors impairing inner ear are called ototoxic factors. Adverse effects could be manifested as hearing loss
or/and balance changes. Hearing and balance impairment can be caused by several factors, including age,
heredity, biological factors casing illness of ear, diseases of nervous system, usage of ototoxic drugs and
exposure to ototoxic substances. Occupational hearing loss/ disorders is usually an effect of occupational
noise exposure, while balance impairments are said to be a result of exposure to neurotoxic chemical factors.
Both systems peripheral receptors are localised in inner ear and share some common elements. In working
environment exposures to noise, solvents and combined exposure to both factors are very common Some
animals study have shown ototoxicity of many chemicals mainly solvents and the interaction between
organic solvents and noise. (Johnson et al., 1988; Johnson and Canlon, 1992; Crofton 1994 Lataye et al.,
1997; Fechter 2000 Makitie et al., 2003) In these study organic solvents concentration and noise level are
usually high or relatively high. Recently, many research concerning the influence of noise and chemicals on
human both hearing and balance, working separately or together, have been carried out (Rybak, 1992;
Morata et al.,1993a; Morata et al., 1994 a; Franks et al., 1996; Carry et al.,1997). Ototoxicity of some
chemicals appears only if there is a concomitant exposure to other factors, such as noise, other chemicals,
and its influence on subjects differs. Mechanisms and aim organ of impairments in unknown. Dependence of
ototoxicity on doses of noise and chemicals is still unexplained. Pathophysiology of ototoxicity and its
audiologic pattern is not recognised.
A great variety of organic solvents are currently used and their usage is increasing. They are used in both
industry and home, so there are many occupation and home activities in which humans can be exposed to
them. The term “organic solvents” includes 11 general of solvents that differ widely in structure classes:
alcohols, aldehydes, aliphatic hydrocarbons, aromatic hydrocarbons, cyclic hydrocarbons, halogenated
hydrocarbons, esters, ethers, glycols, ketones, and nitrohydrocarbons. NIOSH estimated that 9,8 million
workers were exposed to solvents only in the USA in 1974. Since this time production and usage of solvents
increased, (Gerr and Letz, 1998), about 2 million workers in the UK, and up to 25% of the adult population
having some exposure (Dick et al., 2002). Organic solvents are constituents of paints, glues and are used to
produce plastic materials.
Solvents affect the nervous system, liver, kidneys and skin. The toxic effects of an solvent depends on the
target dose, that is, the concentration-time integral or time course concentration of active agent at the target
side rather than the external dose. Toxicokinetics of an solvent and its health effects may change in presence
of other chemicals included other solvents. Some of solvents are recognised carcinogens or suspected of
possessing carcinogenic activity in human in human and animals as well. Their toxic effects on central
nervous system and organs of vision, balance and smell are widely recognised especially after exposure to
relatively high concentration. Limited papers are available on the effects of organic solvents and combined
exposure to solvents and noise at a relatively low level, more common in industry now. Numerous
experiments on animals and growing number of epidemiological research suggest their toxic effects are
exerted on auditory organ as well.
Ototoxicity of selected organic solvents.
Styrene
Styrene (ethenylbenzene) is a widely used chemical for commercial production of polymers, copolymers and
reinforced plastics. Its annual world-wide production capacity was estimated to be over 16, 5 million tones in
1992 (Pfaffi P and Saamanen, 1993). At the beginning of 90-ties about 5000 workers were exposed to
styrene in Poland, and since this time the number of exposed to styrene increased (Kostrzewski, 1999)
Industrial production of styrene consists of two steps. The first is the catalytic alkylation of benzene with
100
June 2004
NoiseChem 101
ethylene. Both raw materials are supplied primarily from the petroleum industry. Ethyl benzene obtained can
be further dehydrogenated to styrene or oxidised to the hydroperoxide, which is then reacted with propylene
to yield the propylene oxide and a co-product, methyl phenyl carbinol. The carbinol is then dehydrated to
styrene. The pattern of styrene consumption is essentially the same world-wide. Occupational exposure
levels are usually orders of magnitude higher than the levels measured for the general population.
Professional exposition to styrene varies considerably depending on the operations concerned. Although
styrene is present in detectable amounts in styrene/polystyrene manufacturing plants, the greatest exposures
occur in the operations and industries that use it as a solvent. It should be remembered that in all industrial
styrene applications, high levels of exposure occur also during the clean-up and maintenance procedures.
Studies on workers, long-exposed to styrene, proved that the exposure affects the function of nervous and
reproductive system and has some genetic and carcinogenic effects (Bond, 1989)
Several publications on ototoxicity related to styrene exposure, particularly combined with simultaneous
exposition to noise were published recently. The studies on styrene effect on hearing organ are however
unequivocal. The severity of ototoxic effect reflects the species-related sensitivity to the chemical. Lost of
external hair cells were observed in mice and rats after styrene exposition whereas the similar effect was not
noted in the hearing organ of cats and monkeys. Surveys in humans occupationally exposed to styrene
revealed predominantly the posturographic and vestibular abnormalities (Calabrese et al.,1996; Moller et
al.,1990). Studies on hearing loss in styrene-exposed workers gave the unequivocal results. Some authors
(Muiser et al.,1988) claimed on higher loss in group exposed to lower styrene levels, if compared to higher
level exposed population. The other studies gave no unequivocal answer due to predominant effect of such
factors as noise exposure and age on hearing efficiency (Sass-Korstak et al., 1995). The other investigators
proved that styrene-induced hearing abnormalities are manifested by the loss of high frequencies and
alterations of auditory evoked potentials (Morata et al.,1994 a, Johnson and Nylen, 1995). Exposition to
styrene-containing solvent mixtures results in significant increase of auditory threshold particularly in high
frequency band. (Moller et al., 1990)
Benzene
Although benzene, as definitely carcinogenic, is eliminated from the industrial use, it should be remembered
that the other solvents, replacing him, are predominantly benzene-homologues and may contain significant
amounts of that chemical. Toluene, xylene, ethyl benzene, aniline, nitrobenzene are benzene homologues.
Benzene is not recognised as ototoxic up to date, but literature indicates that its homologues, like: toluene,
xylene, are ototoxic.( Emmett et al., 1995)
Toluene
Toluene (methylbenzene) is a commercially important intermediate chemical produced throughout the world
in large quantities either in the isolated form or as a component of mixtures. As a component of a mixture it
is used to back-blend gasoline. Isolated toluene is applied for (1) the production of other chemicals, (2) as a
solvent carrier in paints, thinners, adhesives, inks, and pharmaceutical products, and (3) as an additive in
cosmetic products. As the homologue of benzene, purified toluene usually contains less than 0.01% of that
chemical, but the industrial grade may contain up to 25% of it ( EHS 52, 1985).
Toluene affects predominantly the central nervous system, reflecting, inter alia, in emotional and
psychomotor disorders, which severity depends on concentration and exposure duration. Toluene is also
known to impair the function of equilibrium organ (Abbate et al.,1993; Morata et al.,1995).
The most extensive studies on organic solvent ototoxicity were performed in case of toluene. It was proved
(Johnson et al.,1992; Johnson et al.,1988) that toluene-induced hearing loss results predominantly from the
injury of external sensory cells in organ of Corti, whereas the alterations of nervous auditory pathway are
less obvious. Exposure to toluene reflects in elongation of auditory waves in evoked potential record. The
toluene acts synergistically with noise (Morata et al.,1991; Morata et al., 1993b; Morata et al., 1997; Lataye
et al..,1997), whereas ethanol co-existence attenuates its adverse effects on hearing (Nylen et al.,1995).
Toluene-induced hearing loss is however progressive and irreversible, even if the exposure is ceased (Pryor
et al., 1983; Johnson,1994). In case of chronic exposition even the low concentrations of toluene may work
ototoxically (Vrca et al.,1996).
101
June 2004
NoiseChem 102
Xylene
Xylene is an aromatic hydrocarbon which existing in three isomeric forms: ortho, meta and para. Technical
grade xylene consists of the three isomers and contains some ethylbenzene. Approximately 92% of mixed
xylenes are blended into petrol. It is also used in a variety of solvent applications, particularly in the paint
and printing ink industries. Occupational exposure to xylenes alone is rare. The approximate world
production of xylenes in 1984 was 15,4 million tonnes ( EHC 190, 1997)
Isolated exposition to xylene was proved not to affect the hearing organ. However, if combined with other
solvents, xylene enhances their adverse effects (Nylen 1994). Exposure to xylene reflects in increase of wave
latency in auditory evoked potentials. Also disorders of equilibrium manifested by horizontal nystagmus
were noted after exposition to the chemical described (Nylen 1994, Odkvist cited after Bazydło-Golińska,
1993)
Ethyl benzene
Ethylbenzene is produced as an intermediate for styrene. However it is also used in connection with the use
of xylene, as around 15-20 % of technical xylene is ethyl benzene. Exposure to ethyl benzene may occur
through contact with paints and varnishes, synthetic rubbers, gasoline and tabacco smoke. The estimated
annual production for ethyl benzene was 6,8 million tones in the USA in 1993, and in western Europe was
around 3 millions tonnes in 1983, ( Pryror and Rebert, (1993; EHC 186, 1996; Fishbein, 1985). Cappaert et
al. (1999) claimed on ototoxic effect of ethyl benzene in rats but not so evident in guinea pig. Effects
depended on method used for examination e.g. evoked potentials underestimated effects in comparison to
behavioral method. They found also that under specific conditions synergistic interaction between noise and
ethylbenzene can occur. Hearing effects of exposure to ethyl benzene in humans were not explored as far.
Trichloroethylene
Trichloroethylene is used as a degreaser in metal industry and in chemical laundries. It is also applied as a
solvent for lubricants. Long-term exposure to trichloroethylene reflects in psychomotor disorders, problems
with concentration and many others. Other solvents, e.g. trichloroethane, toluene or styrene, are known to
enhance the neurotoxicity of trichloroethylene. Animal studies revealed that trichloroethylene, particularly in
high concentrations, apart from various neurological disorders, is the reason for hearing loss at medium and
high frequencies (Jaspers et al., 1993; Crofton and Zhao, 1993; Repert et al.,1991; Crofton and Zhao,1997).
Other solvents
There are more solvents to be recognised as ototoxics besides above reviewed. Carbon disulphide is one of
them .. It widely used in industry in the production of viscose rayon fibres. Adverse health effects of the
chemical are observed either by high or low concentrations. Chronic exposition to carbon disulphide may
reflect in severe neurological and psychiatric disorders. Animal studies revealed that exposure to carbon
disulphide reflects in destruction of myelin sheath in neural fibres and causes the axonal lesions (Reinhardt et
al.,1997a; Reinhardt et al., 1997b; Huang et al., 1996). Resulting degeneration was observed in brain cortex,
thalamus, brainstem and spinal cord. Carbon disulphide was also proved to be the reason for atheromatous
lesions in blood vessels. Odkvist et al. (1982) and Sułkowski (1979) revealed that carbon disulphide affects
the organ of hearing and equilibrium. The impact of carbon disulphide concentration on hearing loss is
higher than the one of exposition duration. Morata (1989) proved that combined exposition to carbon
disulphide and noise significantly increases the risk of hearing organ injury comparing to isolated noise
exposition. Chang et al. (2003) found that CS2 exposure enhanced hearing loss in lower frequencies mainly
in concerted with noise. Sułkowski, Odkvist and Morata suggest that otoneurological studies combined with
auditory tests may give the evidence not only for ototoxicity but also for general toxic effect of carbon
disulphide. The animal studies of Hirata et al. (1992) on auditory brain stem response proved that carbon
disulphide causes the permanent elongation of wave latency by 800 ppm, while by 200 ppm its effect is
transient. The author cited finds his studies useful for monitoring of carbon disulphide-induced dysfunction
of central nervous system.
There exist few references on ototoxicity, especially related to high frequencies, of n-buthanol (cited after
Rybak, 1992 and after Franks et al., 1996), n-hexane (Hirata et al.,1992) and n-heptane (Simonsen and Lund,
1995). Resulting abnormalities are manifested by latency prolongation of inter-wave intervals in auditory
102
June 2004
NoiseChem 103
evoked potentials. Some other solvents exhibit ototoxic activity only providing the co-existence of other
chemicals.
Depending on combination, some substances work synergistically, as styrene-xylene and styrene-toluene, or
exhibit antagonistic effects, like ethanol-trichloroethylene and ethanol-methanol. However, the risk of
hearing loss related to exposure at organic solvent mixture is usually higher, even without the simultaneous
noise exposition (Crofton et al.,1994; Odkvist et al., 1987; Jacobsen et al., 1993; Franks et al.,1995; Mehnert
et al.,1994; Alessio et al.,1994).
The literature provides an increasing number of proves that solvents known to act adversely on balance may
simultaneously be ototoxic due to close anatomical connection between the receptor structures for hearing
and equilibrium (Bazydło-Golińska, 1993; Morata et al., 1991; Morata et al., 1995; Calabrese et al., 1996;
Sułkowski, 1979; Murata et al., 1991, Alessio et al., 1994).
Mixture
Organic solvent mixture is not always the same and its final health outcome depends of any of particular
components. Studies conducted on auditory and vestibular effects of workers exposed to mixture show
increased hearing loss and balance disturbances (Jacobsen et al.,, 1993; Morioka et al., 2000; SliwinskaKowalska et al., 2000; Niklasson et al., 1997). Review of animal and human studies show that mixture
containing toluene, xylenes, styrene and n-hexane are ototoxic and effects on hearing and balance (Johnson
et al., 1995; Yokoyama et al., 1997).
Solvents and noise
There exist numerous papers on synergistic or additive effect of noise and different organic solvents (Franks
et al.,1995; Morata et al., 1994; Morata et al., 1997; Johnson et al., 1988; Johnson and Nylen, 1995; Lataye
et al., 1997; Lataye et al., 2000; Jacobsen et al., 1993; Muijser et al., 1994). Hearing loss is related to the
organ of Corti or neural auditory pathway injury and may be measured by means of conventional audiometry
at wide range of frequencies, changes in otoacoustic reflex and alterations in records of auditory brainstem
evoked potentials. The effect very often results from the dose accumulation.
Concluding, the majority of authors find the hygienic standards for work environment sufficient if refers to
isolated exposure to solvents. However cases of combined exposition and particularly the simultaneous
action of organic solvents and noise, the injury of hearing organ and balance may occur even if the
concentrations of single factors are below their highest allowable concentration or highest allowable intensity
(Odkvist et al., 1982; Franks et al., 1995; Morata et al., 1994 b; Mehnert et al., 1994; Bhattacharya, 1993,
Yokoyama et al., 1997). There are many animals studies concerning effects of exposure to solvents and
combined effects of noise and solvents, showing that susceptibility to ototoxic properties of solvents seems
to be species specific, genotype and age related. (Li et al., 1992; Johnson 1994; Johnson and Nylen, 1995;
Loquet et al., 1999; Davis et al., 2002; Campo et al., 2003). In spite of these differences it is accepted that
animals studies sufficiently showed that many solvents are ototoxic and combined exposures to noise and
solvents increase risk hearing loss. Knowledge of effects in human is still rather poor and inconsistent.
Aim of the study
The aim of the study was to examine the influence of noise and organic solvents on balance system in
population occupationally exposed to these factors and to determine the risk of such exposition.
Materials and methods
Examined groups:
The study group consisted of two subgroups: one - workers exposed to organic solvents and/or noise, the
second – persons non-exposed. Relatively young age of workers was their common feature. Consequently,
the adverse effects on hearing and equilibrium were related to the occupational exposure rather than the other
sources. The workers from the following companies underwent the studies: 1 chemical plant producing
styrene from ethyl benzene, 3 plants using styrene for production of reinforced polyester plastic and 1 plant
using paints for producing automobile items. The companies characteristics are as follows:
• Company producing hydromassage systems and related products, made of acrylic laminate. Small
company manufacturing products made of polyester-glass laminate manually or in closed system .
103
June 2004
•
•
•
•
•
NoiseChem 104
Private company manufacturing products made of thermoset resins, reinforced with glass
reinforcements. It also distributes materials and equipment for thermoset resins processing. The
production is carried out by the hand lay-up method.
One of the leading companies in Polish chemical sector. It produces, inter alia, styrene plastics
(polystyrene) and is in third position among European manufacturers of butadiene-styrene rubber.
Production is highly hermetized but there are a few open sources of styrene and ethyl benzene.
Modern company producing automobile items. Mixture solvents in paints are used for producing
automobile items. Efficient ventilation system is used for air control there.
Workers of steel foundry were selected to the group exposed to noise without chemicals.
Control group was selected from persons previously unexposed to noise or solvents. Altogether 299
persons were examined.
Selection criteria for exposed workers
The selection of exposed workers based on prior knowledge about the companies. The criteria for the
exposed workers to be included in the study was
1) employment at the company for at least 3 months
2) known previous exposure to noise or solvents
3) priorities for workers younger that 45 years old however 17 workers above 45 years old were included to
the study
Selection criteria for controls
Controls were selected from employees at the Institute of Occupational Medicine and Environmental Health
and from adult students (above 18 years old) whose age suit the range of the exposed subjects.
The criteria for control to be included in the study was:
no previous exposure to noise or solvents
no ear surgery in the past
The workers and the controls participated the same examine protocols and filled out the same questionnaire.
Equipment
For immitance audiometry, conventional tone- audiometry and high - frequency tone audiometry –
equipment manufactured by Madsen was used: Zodiac and Orbiter audiometers, respectively. ABR and P300
were recorded by equipment type Viking manufactured by Nicolette and posturograph SWAY 7.0 of Danish
Development Products was used for body sway records.
Testing Protocols
Before each test the subjects were explained the testing procedure.
• Questionnaire (Noisechem questionnaire) on potential factors impacting hearing. Information was
gathered on endogenous (individual characteristics and medical history) and exogenous factors
(occupational and non-occupational exposures and life style factors) that could be associated with the
study outcomes e.g. tinnitus and ringing in ears, etc. as well as noise and solvents exposure history.
Noisechem questionnaire was prepared by Thais Morata and distributed to members of Noisechem team.
The questionnaire was translated into Polish. Method: face -to- face was used.
• ENT examinations, especially precious otoscopy for excluding serious pathology in the ears, were made
• Immitance audiometry (tympanometry plus acoustic reflex) was performed to determined middle ear
status and to measure stapedius reflexes at 4 frequencies (500, 1000, 2000 and 4000 Hz) plus white noise
stimulus, contra- and ispilaterally
• Pure-tone audiometry at different frequencies. Hearing threshold was determined as the lowest intensity
of an audible pure tone. Examination includes:
• conventional audiometry evaluation of hearing thresholds at 250, 500, 1000, 2000, 4000, 6000 and 8000
Hz in 1 dB steps, ascending mode
• high frequency audiometry above 8000Hz (10000, 12500, and 16000Hz) in 1 dB steps, ascending mode
104
June 2004
NoiseChem 105
•
Transient Evoked Otoacoustic Emissions (TEOAEs) were assessed by using click at 80 dB and 40 µsek
duration
• Brain Stem Audiometry (ABR) by click at 0.125 msek, intensities of stimulus: from 80 dB nHL to decay
of V wave in 10 dB steps, alternative polarisation. 2000 clicks were presented binaurally through
headphones for each record. Electrodes were placed on both mastoides and on vertex and forehead.
• Cognitive event-related potentials P300 were recorded by using stimulus of 80 dB intensity, 100 µsek
duration, alternative polarisation, frequency of non-target stimulus– 500 Hz, frequency of target
stimulus– 2000 Hz. Stimulus tones were presented binaurally through headphones in a random series.
300 tones were presented consisting of target and non-target tones. The participant’s task was to count
target tones. Electrodes were placed on both mastoides and on vertex and forehead
• Testing protocols of body sway include procedures limiting /switching off one or more systems
responsible for balance control. In posturographic examination balance is assessed by recording
micromovements of the whole body which are necessary for keeping erect posture. Spontaneous sway is
registered by unsteady platform which reacts on the changes of position/movements of body weight
centre. Electric signals evoked by sway are registered and processed by computer software. Testing
protocols include procedures limiting /switching off one or more systems responsible for balance control.
So examination is carried out in four conditions:
- standing on the platform with opened eyes- all systems: vestibular, proprioreceptive and visual
control the posture,
- standing on the platform with closed eyes- visual system is switched off,
- standing on the foam pad put on the platform with opened eyes- information from proprioreceptors is
modified,
- standing on the foam pad put on the platform with closed eyes- visual system is switched off and
information from proprioreceptors is modified.
Two foam pads were used: one foam was 1,5 cm thick, and the thickness of the second one was 3 cm (two
foam pads- 1,5 cm thickness each). Duration of each test was 30 seconds.
Prior to posture test session the subjects were interviewed regarding sleep duration, amount of coffee,
alcohol, and cigarettes used in previous 24 hours.
Commonly used sway parameters are: measurements of total sway length, graphic presentation of sway, its
mean velocity, total area, Romberg index etc. Sway measurement examination is safe, does not evoke
side/adverse effects in subject and gives objective result.
Sway measurement examination is safe, does not evoke side/adverse effects in subject and gives objective
result.
Hearing and posture assessment
Worker’s hearing was classified as normal with hearing loss no greater than 20 dB at any audiometric
frequency based in conventional or high frequency ranges.
Normal TEOAE was classified if reproducibility was higher than 50 % and signal to noise ratio higher than 3
dB for the whole spectrum components.
Average of latencies of controls were cut-off values for evoked potentials to be classified as normal or
abnormal.
The same base was used for posture assessment.
Exposure assessment
Exposures were determined by combining environmental sampling and biological monitoring of solvent
exposure. Exposure to noise was determined on noise measurements and hearing protector usage was taking
into consideration as cofounder.
A). Assessment of Solvent Exposure
Questionnaire
All workers were interviewed and asked to fill in a questionnaire about their past and present exposure to
noise and / or solvents. The interview covering time spent with particular tasks was used
105
NoiseChem 106
June 2004
Air measurements
Personal air sampling was used as a routine method of evaluating workers' exposure to solvent vapours,
according to Polish monitoring criteria of occupational environment (PN-89/Z-04008/07). It involved the
collection of air-samples by sampling devices worn by the worker. The sampling device was positioned as
close as possible to the breathing zone of the worker so the sampler collected closely approximate the
concentration inhaled. Aspirators type PSP manufactured by SKC were used. Samplings were proceeded
with the normalised procedure before analysis. Gas- chromatograph type STAR 3400 CX of VARIAN firm
was used for analysis. Single measurements were made.
Whenever reliable retrospective exposure records existed, they were used for estimating past exposure. The
usage of protectors was taken into account.
For each exposed subject, total cumulative styrene, toluene, acetone, xylene ethyl benzene or/and benzene
exposures (cumulative dose- CD) in lifetime were calculated according to the formula shown below:
i
CDsolvent =
∑ t * exp level
n=0
n
tn = time in year of exposure
explevel = exposure level to particular solvent in mg/m3 in air
Workers were divided into two subgroups: subgroup exposed to single particular solvent or mixture of
solvents independently on concentration of solvents and unexposed subgroup.
Admissible value of dose exposure was calculated for each worker according to the formula
i
ACDsilvent =
∑ t * OEL
n=0
n
solvent
Workers were divided into two groups: as 0 - group workers with CDsolvent ≤ ACDsilvent, and as 1- group
workers with CDsolvent > ACDsilvent
Biomonitoring
Additionally biological monitoring was carried out. Biomonitoring involved measurements of concentration
of the metabolites of solvents in urine at the end of shift (mandelic acid - for styrene, hypuric acid for
toluene, phenol for benzene).
Urine samples were collected after at least 6 hours of shift fourth or fifths workday (i.e. on Thursday or
Friday). Single sampling was made. Urine samples were stored frozen to – 200C up to the day of analysis.
Determination of mandelic acid in urine
Reagents
Mandelic acid (MA), β-naphtol (as an internal standard), ethyl acetate, diethyl ether,
N,O-Bis(trimethylsilyl)trifluoroacetamide (BSTFA) were obtained from Aldrich.
Chromatographic analyses
The electron ionisation (EI) GC-MS analyses were carried out by using Hewlett-Packard 5890 II Plus gas
chromatograph interfaced with HP 5972A mass detector. The gas chromatograph was equipped with HP5MS column (30 m x 0,25 mm x 0,25 µm). The following chromatographic conditions were set:
¾ the oven temperature was programmed: 40°C (0,5 min.) – increase of temperature 5 °C/min to 85 °C (for
0,1 min.), increase of temperature 10°C/min. to 220°C (for 7 min.);
¾ injector temperature: 250 °C;
¾ detector temperature 280 °C.
The carrier gas was helium, the column flow: 0,817 ml/min
Preparation of standard solutions
Mandelic acid basic solution was prepared in ethyl acetate at concentration 40mg/100 ml. An internal
standard solution (β-naphtol) was prepared in ethyl acetate at concentration 80mg/100 ml. Standard solutions
were prepared by adding acetate solutions of mandelic acid and β-naphtol to test tubes in accordance with
data in the table below.
106
NoiseChem 107
June 2004
Preparing of standards solutions
Volumes and concentrations
Concentration of
1
standard solution
Mandelic
acid
0.005
basic solution
β-naphtol solution 0.5
Units
10
50
100
400
800
mg/l
0.05
0.25 0.50
2.0
4.0
ml
0.5
0.5
0.5
0.5
ml
0.5
The contents of test tube were concentrated to dryness under nitrogen stream and then 2 ml of physiological
urine, 0,3 ml of 6n H2SO4 and 0.4g of NaCl were added. MA in urine was extracted with 10 ml of diethyl
ether (5 min.). Five millilitres of etheral layer were transferred to another test tube and evaporated to dryness.
200 µl of BSTFA was adding, the test tubes were closed and heated (the water bath) at 70°C for 15 min. One
millilitre of diethyl ether was adding. 2 µl of the solution was injected into the GC-MS.
Sample preparation
Determination of mandelic acid in urine
0,5 ml of β-naphtol solution was added to a test tube and evaporated to dryness. 2 ml of urine and 2 ml of
11n NaOH were added and heated (the water bath) at 100°C for 2 h. After hydrolysis, 5 ml of 6n H2SO4 and
0.4g of NaCl were added, then MA in urine was extracted and the further sample preparation was the same
as the preparation of standard solution described above.
Determination limit value was 0,4 mg/l urine.
Concentration of urinary metabolites was adjusted/ corrected to creatinine clearence. Biological Exposure
Indices Concentration of urinary metabolites were adjusted to/ corrected for creatinine clearence. Workers
were divided into two subgroups in dependence on Biological Exposure Indices (up to BEI’s and above
BEI’s) for further analysis.
Phenol and hipuric acid determination
Phenol, the metabolite of benzene, was determined in benzene-exposed patients. Acidified urine sample was
distilled with steam. Phenol was measured colorimetrically by spectrophotocolorimeter (Spekol 11 Carl
Zeiss Jena) within the distillate with 2,6-dibromoquinone chloroimide ( Dutkiewicz, 1964) No sample
reached BEI for exposed (70 mg/g creatinine). BEI for phenol in urine for unexposed is equal 19 mg/g
creatinine. This last value was base for workers’ division into subgroups.
Urinary determination of benzoic acid, or its main product, hipuric acid, is routinely performed for tolueneexposition assessment. Photometer 1101 M Eppendorf was used for measurements. BEI for hipuric acid in
urine is equal 1500 mg/g creatinine( Dutkiewicz, 1968).
Hewllet Packard gas chromatograph HP 5890 seria II Plus with mass detector HP 5972 A was used for
determination of mandelic acid in urine - a metabolite of styrene. BEI for mandelic acid in urine at the end of
shift is equal 800 mg/g creatinine or releasing equal 16mg/h. Both criteria were used.
B). Noise Exposure Assessment
Sound level meter and personal dose meter were used to assess noise exposure. Noise dose-meters involved
the use of body-worn instruments to monitor a person's noise exposure over the work shift. In case of usage
of the sound level meter time pattern of noise was assessed and noise exposure was calculated.
Measurements were carried out according to Polish standard PN 94/N- 1307. If available, retrospective
noise measurements were used to characterise noise exposure. The usage of hearing protectors was taken
into account. Bruel and Kjaer equipment as well as SVAN 912 of Svantek were used
Life noise dose (ND) was calculated according to the formula:
i
ND =
∑t * L
n=0
n
exp, A
107
June 2004
NoiseChem 108
tn = time in year of exposure
Lexp8h,A= time weighted average exposure level to noise in dB A
Exposure evaluation
Current exposure was assessed by TWA levels and compared to TLVs. Lifetime doses of particular hazards
were calculated as average levels multiplied by duration of exposure. All workers were split into two
subgroups: one subgroup consisted of workers exposed to levels up to TLVs and the second one – higher
than current TLVs. In a case of noise three subgroups were built up: unexposed, it means up TWA to 79 dB
A, second one workers exposed to noise exposure levels between 80 – 85 dBA, and the third one – workers
exposed to noise levels above 85 dBA. TLV for noise is equal 85 dB A TWA/8h
Admissible lifetime dose (ALND) for noise for each worker was calculated according to the formula ALND
i
=
∑ t * 85dBA / 8h
n=0
n
tn = years of exposure
Workers were divided into two subgroups: as 0 - group workers with NDt ≤ ALND and as 1- group
workers with ND > ALND
All tests performed in examinees were non-invasive and no discomfort or risks were associated with the
selected methods. Participation in the studies was voluntary, and examinee could refuse continuing
participation in examination at any moment. Bioethics Commission for Research at the Institute of
Occupational Medicine and Environmental Health accepted the testing protocol. The testing protocol is in
accordance with to the Declaration of Helsinki in its latest version.
Statistical analyses
Checks and rechecks data and codes were made. After correction of errors data file was analysed with SPSS
and Statistica packages. All statistical tests were done with assumed significance level<0,05, while p-values
0,051 up 0,10 was considered statistically significant for inclusion in multivariate analysis. Pearson
correlation, Spearman rang correlation, Student t- test, Chi – square test, analysis of variance ANOVA for
assessment of interactions, binary logit analysis (only in a case of nominal variables e.g. exposed and
unexposed, subgroups: normal and abnormal, etc.) were used suitably. Quantitative data were coded to
obtain binary values to use binary logit analysis because unnormal distributions of some independent data.
Odds ratios were calculated. Numerical data were analysed by multivariate regression too. By using multiple
linear regression analysis and multivariate regression for categorised variables and numeric variables,
respectively, were controlled for age, length of tenure, smoking habits, health history, alcohol intake, nonoccupational exposure to noise or solvents.
The relationship between particular hearing tests was analysed by Pearson’s correlation coefficient (r). The
same analysis was used for relationships between hearing tests and body sway test. Spearman rang analysis
was made for relationships between exposure parameters and hearing or body tests.
Results
Demographic data
Table 1 presents the data from Noisechem questionnaire concerning medical history, hearing status, exposure
history, health habits (smoking, alcohol consumption, coffee consumption) and leisure activities connected
with exposure to noise and solvents. No difference regarding age was found between the controls and the
exposed. Differences regarding age, statistically significant, were found for following couples of subgroups:
(i) noise and noise + mixture; (ii) noise and controls; (iii) noise and styrene; (iiii) noise and noise + styrene.
More than 40% of the exposed subjects have military service history and only 4 persons (~5%) of the control
ones. Tinnitus was more frequent among the exposed ones than among the control. Among the control no
case of measles was found, less allergy but far more frequent kidney disorders. Regarding leisure time, noise
or solvents exposure seems to be slightly more frequent among the exposed than among the control ones but
the differences are not statistically significant.
Annex 1encloses tables with results of statistical analysis of significance of differences
108
June 2004
NoiseChem 109
Study subgroups
Examined population was divided into 8 following subgroups:
− I – controls(C) – no occupational exposure to solvents and/or noise
− II noise group (N) – workers exposed only to noise ≥ 80 dB A
− III styrene group (S) - workers exposed only to styrene (on the base of questionnaire) with exposure to
noise at workplace < 80 dBA
− IV styrene and noise group (N+S) - workers exposed both to styrene (on the base of questionnaire) and
to noise at workplace ≥ 80 dBA
− V styrene and ethyl benzene group (S+EB) - workers exposed to styrene and ethyl benzene (on the base
of questionnaire) with exposure to noise at workplace < 80 dBA
− VI noise, styrene and ethyl benzene group (N+SEB) workers exposed to styrene (on the base of
questionnaire) and to noise at workplace ≥80 dBA
− VII mixture group (M) – workers exposed to solvent mixture (on the base of questionnaire) with
exposure to noise at workplace < 80 dBA
− VIII noise and mixture group (N+M) - workers exposed to solvent mixture (on the base of
questionnaire) with exposure to noise at workplace ≥80 dBA.
Exposure
Exposure to solvent mixture was found in two factories, but the composition of mixtures was different for
each of them. One plant used paints and consequently the mixture consisted of toluene, xylene, ethyl
benzene, butylacatate, 2-buthanone, trimethyl and 1-buthanol. Mixture from the second plant contained
styrene, acetone and ethyl benzene. Some workers were exposed to extremely low levels of solvents - below
detectable level of measuring equipment and they were classified to the relevant subgroups basing on
interview only. Exposure data for the subgroups are given in Table 2.
Exposure to solvents was relatively low mostly. Only exposure of 51 workers exposed to styrene exceeded
TLV. Concentrations of other solvents were below current TLVs. 130 workers were exposed to noise above
L ex, 8h = 85 dB A ( Polish TLV). 98 workers were exposed to noise ranged 86-90 dB A and 32 workers were
exposed to noise level 91-99 dB A. Only 90 workers were exposed to levels lower then 85 dB A. Figures 1
and 2 show histograms of noise levels and styrene levels at workplaces respectively.
The following TLVs were assumed: noise 85 dB A/8h, styrene 50 mg/m3, acetone 600 mg/m3, benzene 10
mg/m3, ethyl benzene 100 mg/m3, xylene 100 mg/m3, toluene 100 mg/m3, butylacetate 200 mg/m3, 2buthanone 200 mg/m3, 1-buthanol 50 mg/m3.
Exposure assessment was based on single measurement. Since some inconsonance was found between
measurement results and interview data, for the purpose of binary logistic analysis the data on exposure were
taken from questionnaire and expressed by binominal values, where 0 = unexposed and 1 = exposed.
Group exposed only to noise had longer tenure comparing to other subgroups exposed to solvents or noise
and solvents. There were statistical differences between the subgroup exposed to styrene alone and the
subgroup exposed to noise and mixture comparing to other subgroups regarding years of exposure to
solvents.
Leisure shooting, motorcycling were found mostly among exposed workers, and exposure to discotheque
noise, and diving were found in both controls and exposed at the similar rate. Non– occupational exposure to
solvents was found only in 5 persons of whole examined group including both exposed and controls.
Biomonitoring
Mean urine levels and result ranges of solvent metabolites in workers expressed in mg per g of creatinine are
presented in Table 3. Distributions of metabolites in urine are shown at figures 3-5.
The level of mandelic acid in urine was highest in subgroups exposed to styrene and styrene+noise. The
differences between exposed to styrene, styrene + noise and styrene+ ethyl benzene and other subgroups
were statistically significant regarding mandelic acid in urine. There was no difference statistically
significant between subgroups regarding phenol and hipuric acid in urine. Figures 3-5 show histograms of
levels of metabolites of styrene and toluene in urine respectively. Levels of phenol in urine were under BEI
109
June 2004
NoiseChem 110
for the exposed. Distributions of examined parameters were not normal , so Spearman rang correlation test
was used. Table 4 shows correlation between exposure parameters and levels of metabolites in urine.
Hearing tests
The studies on hearing and equilibrium organ were performed as a blind trial, i.e. the investigators knew
neither the noise exposition parameters nor the levels of solvent metabolites.
Hearing data are summarised in Table 5. Figures 6-8 show hearing thresholds for particular subgroups
compared to controls and noise exposed: subgroup I – exposed to styrene and styrene plus noise; II –
exposed to solvent mixture and mixture plus noise; III - exposed to styrene plus ethyl benzene and styrene
plus ethyl benzene plus noise.
Although lack of the otoscopically determined abnormalities was the condition qualifying subjects to the
study, tympanometry revealed 6,6% and 2,5% of type B and C tympanograms for the exposed group and
controls, respectively.
Typanograms type B and C according to Jerger classification were found in 14 and 12 ears left and right
respectively. Differences of averages stapedius reflex up to 2000 Hz both sides were statistically significant
between B and C tympanes and others one. The threshold of ipsilaterally induced stapedius reflex was
slightly lower in the group exposed, comparing to the control, whereas for contralateral stimulation the
results were contrary (Figure 9) Ipsilateral decrease of stapedius reflex threshold at 2000 Hz for the left ear
in styrene and noise exposed subjects was statistically significant comparing to the controls (MANOVA)
(Tables in Annex 2). For styrene exposed workers the significant decrease of the threshold was noted for the
left ear by contralateral stimulation at 4000 HZ and for the right ear by contralateral signal at 2000 Hz, 4000
Hz and WN. The results described may suggest the retrocochlear lesions of hearing organ. Spearman rank
analysis indicated significant negative correlation with age and noise exposure parameters (Lex and
LCpeak). Positive correlation was noted however with styrene exposition parameters (TWA) for the
contralateral stimuli exceeding 500 Hz, with urine concentration of mandelic acid, and lifetime doses of
styrene or acetone. Logistic analysis failed to indicate the significant relationships among the thresholds of
stapedius reflex and independent variables studied (Tables in Annex2).
An average hearing threshold for conventional band amounted 11,1 dB and 10,2 dB for right and left ear,
respectively, in the groups exposed to noise or solvents, alone or in combination. Analogous values
determined for the controls were 5,8 and 4,8, respectively. If the loss was lower than 20 dB HL for the entire
band of frequency, the hearing was considered as a normal. If the loss was larger for at least one frequency
the hearing was ranked as impaired. In the group exposed to noise and solvents, alone or in combination,
hearing was diagnosed as a normal or impaired in 57 % and 43 % of subjects, respectively. Similar criterion
of hearing impairment was employed for high frequencies. Accordingly, the percentage of subjects with
normal or bilaterally impaired hearing in the exposed groups was as 46,5% and 53,5%, respectively. Sixty
seven percent of normal hearing and 33% of impairments were observed in the control group. The fraction
with hearing loss was higher at high frequencies comparing to the conventional band. Considering the certain
subgroups, the fraction with hearing loss at conventional band amounted: 10% in controls, 28% in exposed
to noise, 15% in exposed to solvents (styrene only - 15%) and 20% in exposed to solvents and noise (noise +
styrene – 28%). The analogous percentage for high frequencies amounted 14%, 58%, 34% (39%) and 27%
(29%), respectively. Consequently, the susceptibility to hearing loss seems to be larger at high frequencies
than at conventional band, whereas noise is more harmful to hearing organ than styrene.
Calculations of relative risk, related to isolated exposition to solvents or noise are presented in table 6. Table
7 presents odds ratio for high frequency range of hearing loss
Tone audiometry
The lowest hearing loss was noted within the control group. The highest loss was noted among the noise
exposed subjects, but it should be noted that the group discussed was 5 year older than the solvent mixture
exposed workers and 10 years older than the remaining groups. The differences described were statistically
significant. The hearing loss among noise + styrene exposed subjects was higher than among the workers
exposed to styrene only, but the difference was not significant. It should be claimed that the duration of
exposure to styrene alone was nearly 4 times shorter than to styrene combined with noise, whereas average
levels of exposition were similar. The analogous differences in duration and levels of exposition were noted
110
June 2004
NoiseChem 111
for the workers exposed to styrene and ethyl benzene, if compared with the group exposed to the same
solvents combined with noise. The logistic analysis was carried out, taking into account the aforementioned
differences in exposition parameters and other independent variables. Moreover, the average hearing
thresholds were compared among the subgroups and with the median values given by ISO 1999 standard,
with age and noise exposure taken into consideration. The results are presented in Figure 10. It is evident that
the least differences among the median values of hearing thresholds according to ISO 1999 standard and the
mean values observed were noted among the workers exposed to noise and solvent mixture together
(differences of 2 and 1,5 dB for right and left ear, respectively), in the group exposed to noise only (4,5 and
2,5 dB) and for the controls (5 and 4 dB). In case of the remaining groups, the differences among the hearing
thresholds obtained and median values corrected for age and noise exposure were significantly larger. The
differences amounted: 1) 9 and 9,5 dB for styrene, 2) 8 and 7,5 dB for styrene and noise, 3) 8 and 7,5 dB for
styrene and ethyl benzene, 4) 9 and 7 dB for styrene, ethyl benzene and noise, and 5) 8,5 and 6 dB for
solvent mixture exposition, for right and left ear, respectively. Many factors contribute to hearing loss
besides ageing process, extra occupational noise exposure, some diseases , ototoxic drugs etc. Many of them
are not under control, and people are not able to t remain some events of their life that could be ototoxic, so
hearing threshold of unscreened population is usually worse in comparison to standards.(Toppila et al.,
2000) This is one of reason of differences between ISO 1999.1990 hearing loss due to age and hearing loss
of controls at least.
The combined exposure to noise and solvent mixture reflected in decrease of hearing threshold, comparing to
the isolated exposition to one of that factors, which was revealed by means of MANOVA. The analysis
proved also that the noise exposure elevates the hearing threshold at some frequencies.
Spearman rank analysis revealed significant relationship between the hearing loss and age, cigarette
smoking, volumes of coffee drunk, ear surgeries or BMI at the entire band of frequency. Differences in
hearing thresholds between smokers and non- smokers are (figure11) statistically significant. Age of both
groups was the same. The amounts of cigarettes smoked or coffee drunk at the day of hearing examination
were also significantly correlated with the results of tone audiometry. Positive correlation was also observed
between cholesterol level and hearing loss. Statistically significant relationships were limited to the high
frequencies in case of hyperthyroidism or previous mumps episodes. Tinnitus and ear ringing were correlated
with the hearing loss. The hearing thresholds were positively correlated with parameters of styrene or
acetone exposure (TWA and MAX). Negative correlation was noted however between the hearing loss up to
2000 Hz and the parameters of toluene or xylene exposure. Significantly positive correlation was found
between the hearing loss up to 1000 Hz and lifetime dose of noise and styrene, as well as lifetime dose of
acetone exposure. The negative correlation was noted however for lifetime dose of xylene or butyl acetate.
Urine concentrations of mandelic or hipuric acid were significantly correlated with the hearing loss at a
whole band of frequency. Tables are enclosed in Annex 3
Transient Evoked Otoacoustic Emisions - TEOAE
Results of TEOAE in the same groups are presented in Figures 12-14. Hearing threshold is elevated due to
exposure to styrene while the TEOAEs reproducibility and amplitudes were suppressed if compared to
controls. The effects described were enhanced in case of combined exposure to noise end styrene. The
effects of noise alone seem more pronounced but it should be remembered that the relevant subgroup was
about 10 years older than the other subjects and the age is known as a very strong confounder. Consequently,
the average hearing loss at frequency range from 500 to 6000 Hz, presented in the figure 10, is compared
with the same parameter calculated of median values of the same age given in ISO 1999. 1990. No
difference was found between subgroups in the immitance measurements except acoustic reflex threshold at
stimuli 4000 Hz contralaterally and WN. ( Figure 9) The exposed groups seemed to have lower thresholds of
stapedius reflex at 0,5; 1; and 2 kHz ipsilaterally comparing to the controls, and higher thresholds of
stapedius reflex at 2 and 4 kHz and WN stimulus contralaterally. Hearing threshold of the controls is lower
comparing to the exposed. Individual’s hearing classified as normal was found more frequent among the
controls comparing to the exposed in both frequency ranges: conventional and high frequency. Right side
was slightly worse comparing to the left side in both groups (the controls and the exposed) but differences
were not statistically significant. 56% of controls and only 22% of exposed workers had normal TEOAEs.
TEOAEs were symmetrical.
111
June 2004
NoiseChem 112
The results of tone audiometry were significantly correlated with the parameters of TEOAE. TEOAEs
reproducibility and S/N ratio were the highest in the control group and the lowest – among the noise-exposed
workers. The latter group was however significantly older, which reflected in significantly longer duration of
noise exposure. The parameters of TEOAEs were slightly more positive in case of isolated solvent exposure,
comparing to the ones obtained for the combined, solvent and noise, exposition. Considering the particular
solvents however, the combined exposition to styrene and noise, or to styrene, ethyl benzene and noise
affected the results more adversely than the exposure to styrene, alone or combined with ethyl benzene.
TEOAEs reproducibility and S/N ratio, resulting from exposition to solvent mixture alone, were lower than
in case of combined exposure to mixture and noise. ( Figure 12-14)The odds ratio for styrene was high and
statistically significant at the whole band in such cases. (Table 8)The similar results were obtained for noise
exposition. The values of odds ratio for ethyl benzene were also high, though not significant. The odds ratio
values for either the acetone or toluene and benzene indicated however that the substances affect hearing
positively. Odds ratio was high for cigarette smoking, whereas its values were low values for alcohol
consumption. Differences between average value of TEOAEs of smokers comparing to non- smokers (Figure
15) are statistically significant only for emission strength, however differences for reproducibility and S/N
ration are not statistically significant. Low, but significant, risk is related to age, reflecting in odds ratio value
about 1,1. Spearman rank analysis revealed significant negative correlations with age, cigarette smoking,
BMI, urine mandelic acid concentration and lifetime dose of noise. The positive significant correlation was
however noted with the rate of mandelic acid elimination. ( Annex 4 includes Spearman rang correlation)
Evoked Potentials
Auditory Brainstem Response - ABR
The data obtained by means of evoked potentials are summarised in the following figures. Figure 16
presents ABR pattern of a control (right) and an exposed to noise and mixture (left) Both are at the same age.
. Mean values and standard deviations of the auditory brainstem response (ABR) for particular subgroups are
presented in figures 17-19. The average latencies and amplitudes of P300 for non-target and target stimuli
are given in figure 20. The values of N1 with two positive peaks are also presented in figure 21-22 .
Morphology of P300 plots for workers exposed to solvents and workers exposed to noise + solvents was
altered comparing to the control plots. Mean values ± SD of ABR and P300 parameters measured for the
controls were considered as reference for the exposed workers. That is because there are no universal
reference values for ABR or P300. Therefore each laboratory develops its own set of latency values for both
methods by measuring the ABR/P300 latencies for young adults (15-30 subjects). Latencies of both ABR
and P300 strongly depend on designs of used stimuli and their repetition rates. Latencies and amplitudes of
both methods determined for controls were used as references values. (Jacobson et al., 1985; Pilecki et al.,
2002; Polich, 1998; Polich and Herbst, 2000; Gonsalvez and Polich, 2002). Comparing to the controls, the
waveform pattern of ABR and P300 in workers exposed to solvents was normal very rarely, presenting
numerous artefacts. Consequently, some data which assessment was impossible were excluded from
analysis.
Besides the increase of wave latency and wave intervals, other morphological alterations were recorded
among the exposed workers, and particularly in solvent exposed ones. The evident wave peaks were lacking
and some waves were lost. The latency of wave I was similar for all the groups. More variability was noted
for wave III, whereas the records of wave V were the most differentiated among the groups. The ABR
records proved significant effect of noise and combined exposure to noise and styrene. The differences of
latencies and intervals were not significant for the other solvents. Changes in ABR records indicate that
solvents impair the central part of auditory pathway. Therefore they exhibit neurotoxical properties. Annex 5
encloses Spearman rang correlation between ABR parameters and investigated independents
Auditory Cognitive Potentials - P300
The similar alterations were noted in P300 records. Morphology of the potential records was changed
significantly. The latencies were not significantly different among the groups in case of N1 and P2 non-target
paradigm waves. The P3 wave was increased however and the extent of alteration was the most pronounced
in the group exposed to solvents. Solvents or noise alone usually increased latency, whereas in case of
combined exposition, the latencies were decreased, comparing to the controls or groups exposed to only one
112
June 2004
NoiseChem 113
of the aforementioned factors. Non-target wave amplitudes were decreased comparing to the controls.
Variance analysis revealed the significant differences in the groups exposed to solvents, alone and combined
with noise. Spearman rank analysis indicated the significant negative correlation of all P300 parameters with
age, rate of mandelic acid elimination, styrene exposure or maximal levels of noise. Annex 6 encloses
Spearman rang correlation between P300 parameters and investigated independents
Posturoraphy - Body sway
The stabilograms of one worker obtained during the body sway performed under the different test conditions
(eyes open- non foam (EONF), eyes closed – non foam (ECNF), eyes open – 1 foam (EOF), eyes closed
foam (ECF), eyes open – 2 foams (EO2F), eyes closed – 2 foams (EC2F)) are presented in Figure 23.
Commonly used sway parameters are: measurements of total sway length, graphic presentation of sway, its
mean velocity, total area, Romberg index etc. We have decide to used the most often parameters : mean
velocity, total area, and Romberg index. Furthermore we used frequency analysis of body. sway.
The comparison of sway area and sway velocity for the subgroups examined are given in Figures 24 and 25.
Body sway area was larger, and body sway velocity as well as sway frequency were higher for the exposed
comparing to the controls. Sway frequency of smokers was higher comparing to non-smokers. Sway area and
sway velocity in different test conditions were positively correlated with styrene and acetone exposure
parameters and with mandelic acid in urine. The same relationship was found for BMI. No age effect was
found. Differences were statistically significant. There was no regular association between the other
parameters of body sway (mean, transversal, sagittal and sway index) and exposure, and the significant
values of the correlation were accidental. Romberg ratio was calculated according to Sack et al. (1993). Also
the calculations of area and velocity ratios in different test conditions ( ECNF/EONF, ECF/EOF,
EOF/EONF, ECF/ECNF, ECF/EONF, EC2F/EO2F, EO2F/EONF, EC2F/ ECNF, EC2F.EONF) were
performed. The results provided the evidence for the relative effect of visual inputs with and without
proprioceptive modifications. The values of relevant calculations are presented in Figure 26. Mean values of
the controls ± SD were considered as the references in course of the analysis.
Body sway area and Romberg values are the most frequently considered parameters of body sway. The sway
area for the exposed groups was larger than for the controls. The most pronounced alterations of the sway
area were observed among the workers exposed to the solvents only. The changes noted in the group
exposed to noise and solvents together were less evident, while for the workers exposed to the noise only, the
increase of the parameter discussed was the lowest of all the experimental variants. The sway area was larger
for the thick foam than for the thin one, when the signals for proprioreceptors were modified. Turning off the
optic receptor reflected in the maximal increase of the sway area. While modifying inputs from vision
receptors and prioprioceptors and comparing changes of results show increasing reliance of higher centers.
The following comparisons concerning sway area were made: eyes opened standing on foam versus eyes
opened with no foam under feet (i), eyes closed on foam versus eyes closed with no foam (ii), eyes closed
with no foam versus eyes opened with no foam (iii), eyes closed on foam versus eyes opened with foam
(iiii), eyes closed on foam versus eyes opened with no foam. The same experiment was performed for both
thin and thick foam. . Figure 24 shows that reliance is increasing linearly in case of thin foam but with thick
foam effect of disturbance is more distinguish. Also the sway frequency was on the increase, when the
balance control was interrupted by modification of signal for proprioreceptors or eye closing. The lowest
increase was noted in the group exposed to noise alone, and the parameter was gradually increasing in
solvent exposed workers to reach the highest values among the group exposed to noise and solvents together.
If analysed for two subgroups, with normal (loss up to 20 dB) and impaired hearing (loss above 20 dB), the
changes of sway area slightly differed. In the group with normal hearing the increase of the sway area was
the largest among noise exposed subjects and the least intensive – among the solvent exposed workers. The
changes of sway area were moderate in the workers exposed to noise and solvents together. In the subgroup
with hearing loss however, the maximal enlargement of the sway area was noted in solvent exposed workers
and the parameter was gradually decreasing in noise and solvent exposed ones, to reach to lowest values
among the noise only exposed workers. As previously described, changing experimental conditions, i.e.
modification of proprioreceptive signals and/or eye closing, reflected in increase of the sway area. The
parameters of sway area and sway frequency were correlated with the variables of styrene and acetone
exposure, the level of exposition to noise, urine concentration of mandelic acid, BMI and lifetime doses of
113
June 2004
NoiseChem 114
noise, styrene and acetone. The sway frequency was correlated with the habits of alcohol or coffee drinking,
besides. The calculated value of Romberg trial (station test) was correlated with either the level of noise
exposition or with the lifetime dose of noise. Multivariate regression showed effects of maximal level of
styrene on sway area with eyes closed and TWA exposure to styrene effected on sway velocity. Negative
relation was found between smoking habits and sway area with eyes closed on thin foam pad. Results of
multivariable analysis are presented in Annex 7. Effects of noise and exposure to toluene on sway area and
sway velocity with open eyes on thick foam pad, and finally sway area with eyes closed and thick pads were
affected by alcohol and motorcycling and adversely with toluene. Summing up the results it could be
concluded that noise and/ or solvents affect body sway. The pattern of changes indicates that exposure
affects in the balance system following parts: vision, vestibular organ., and cereberallar part. Sway area and
sway velocity of controls and exposed don’t differ significantly in a case of modifying inputs of
prioprioceptors. So prioprioceptors seems to be intact by exposure. In conditions of closing eyes both sway
area and sway velocity increased but differences between controls and exposed were considerable,
especially in a case of sway velocity. It indicates vision effects. Comparing results obtained during stand
with eyes closed and foam under feed with results obtained with open eyes and no foam we can assess
vestibular effect. Sway velocity and sway are the most affected by noise and exposure to solvents and
combined exposure to solvents and noise seems to affect less vestibular system than exposure to noise alone.
Effect of noise exposure was reported recently ( Starck et al., 1994; Shupak et al., 1994; Goltz et al., 2001)
These results are consistent with earlier published results (Morata et al., 1991; Morata et al., 1995; Sack et
al., 1993; Smith et al., 1997; Yokoyama et al., 1997; Yokoyama et al., 2002)
Statistical analysis
Two hundred and forty five dependent variables were analysed in the present study (results of audiometry at
conventional and high frequencies, TEOAEs, body sway, ABR and cognitive evoked potentials). They were
mostly of normal or close to normal distribution. The analysis dealt with the relations of that variables to
occupational exposure, including noise and organic solvents. The data from questionnaire on medical and
exposure histories were also considered. Independent variables expressed high variability. Either the binary
qualitative ones (zero-one), describing whether the subject was exposed or not to some factors, or the
qualitative ones with several grades of exposition (noise), occurred among them. Also the continuous
variables (some with only several values) were considered. Some variables exhibited normal distribution
(age, BMI) but the right-sided distributions (index of styrene exposure, maximal styrene concentration, etc.)
were predominating. Due to high number of different type variables, various methods were employed for the
purpose of statistical analysis.
I.
Analysis of variance. Comparison of measurements depending on type of occupational exposure
(styrene, solvent mixture, noise or noise and solvents together)
Considering type of the exposition, the population studied was divided into 8 subgroups, with the subject
number ranging from 7 to 97 people for groups styr+EB and control, respectively. The amount of subjects in
the particular subgroups determined the kind of statistical test employed for the result analysis. Analysis of
the measurements for the group studied was performed by means of co-variance analysis (ANCOVA), with
age of subject as a co-existing (co-variance) variable. The age was analysed since it significantly affects
several dependent variables and the groups studied were age-differentiated. The p-value of probability given
in tables of Annex 8 was calculated from univariable variance analysis with age as a co-existing variable.
The choice of exposure scenarios enabled more detailed analysis of some factor effects. For three
experimental designs it was possible to perform the entire variance analysis with factor interactions
(bivariable analysis with co-existing variable). The effects of noise and solvent mixture, noise and styrene,
and noise, styrene and ethyl benzene were analysed in the first, second and third design, respectively. The pvalues presented in Table were obtained by means of bivariable analysis with age as co-existing variable.
II.
Analysis of predictive variable effects by multiple regression model and univariable analysis
Binary and continuous variables were considered as the predictive ones in the multiple regression model.
The objective of the model analysis was to establish the trend and statistical significance of predictive feature
effects on particular dependent variables. Moreover, by calculation of determination coefficient, R2, the
114
June 2004
NoiseChem 115
percentage of dependent variable variation related to the predictive one, and consequently the rate of model
fit were determined.
Univariable analysis was also performed for the particular predictive variables. One-way variance analysis
and Tukey’s HSD test of multiple comparisons (or t Student’s test for binary data) were employed for
qualitative parameters with two or more categories. The effects of continuous variables were analysed with
an aid of Spearman coefficients of correlation, due to evidently right-sided distribution of that features.
III.
Epidemiological approach. Logistic regression model.
In the model described, several dependent variables were transformed into the binary ones with 1 describing
the workers, whose parameters were different from the reference values (which suggest the adverse effects
existance). Binary variables dominated as the predictive ones in the model, but age of workers was
considered as well. Consequently, it was possible either to analyse the trend and statistical significance of
exposure variable effects on disease probability (deviation from the normal value), or to calculate the odds
ratio for the particular factors.
Logistic regression was employed only for analysis of categorised, or possible to categorise, data. The results
of hearing examination by means of impedance audiometry, as well as the results of body sway and P300,
were difficult to categorise, due to their high individual variability. Consequently, the effects of factors
studied on the aforementioned variables were analysed only with an aid of multivariate regression. Results of
logistic regression are enclosed in Annex 8
By means of bivariable ANOVA the interaction were revealed among the following exposures: (i) noise,
mixture, and noise + mixture; (ii) noise, styrene, and noise + styrene: (iii) noise, styrene+ ethyl benzene, and
noise + styrene+ ethyl benzene.
Distributions of hearing parameters were normal hence statistical tests were used for analysis. The
correlation of tone-audiometry values with the results obtained by transient otoacustic emissions for each
reproducibility and signal-to-noise ratio parameters (1, 2, 3, 4, 5 kHz and overall TEOAE’s) was
significantly negative. The relevant coefficients of correlation ranged form –12 to - 48 depending on couple
of parameters. Correlations between tone- audiometry parameters (right, left side) and acoustic reflex (right,
left side respectively) were statistically significant for 250, 500, 1000 kHz and white noise (WN). Values of
the correlation coefficients (r) ranged from 12 to 36. The latencies of waves I, III and V, as well as wave
intervals, were significantly correlated with the mean values of auditory threshold (average thresholds 250 –
8000 Hz, 500 – 2000 Hz, 4000 – 8000 Hz and high frequency range 6000 – 16000 Hz). The correlation
described was however irregular between particular audiometric frequencies and ABR parameters, and only
single parameters of P300 were correlated with pure tone audiometric results. It was impossible to find the
regular base for the relationship analysed. Finally, the significant correlations were found between pure tone
audiometry and some parameters (mean, transversal, sigittal, sway area and sway velocity) of the body sway
tests (eyes open, eyes closed, eyes open and thin or thick foams, eyes closed and thin or thick foams). The
correlation described was almost constant in case of sway area and sway velocity and incidental for the other
parameters.
Stapedius reflex at 500 – 2000 Hz and for WN stimulus is adversely associated with LEx and LC peak.
Exposure to styrene and cumulative dose of styrene showed positive association with contralateral stapedius
reflex at 1000-4000 Hz and WN. The same results were found for acetone. Ipsilateral stapedius reflex at 5002000 Hz was negatively associated with mandelic acid in urine and contralateral at 4000 Hz and WN
contrary.
By means of one-way ANOVA, the hearing threshold was proved to be significantly associated with age,
tobacco smoking, noise (LEx and LCpeak values and tenure), use of acoustic protectors and styrene or acetone
exposure at the whole extent of frequency. Therefore dependence on dose of above mentioned occupational
factors is obvious. The similar relationship of hearing threshold at the low frequencies up to 1000 Hz was
noted for toluene, xylene and ethyl benzene only. The hearing threshold is significantly dependent on age,
Lex and styrene and ethyl benzene levels at the whole extent of high frequency. Hearing threshold in whole
frequency range was positively associated with ear surgery, cholesterol, BMI (body mass index), coffee
consumption, and tinnitus. Measles occurred to affect high frequency range. Positive association with
madelic acid ( for whole frequency range) and hipuric acid (at 250- 4000 Hz right ear and 1000- 8000 Hz left
115
June 2004
NoiseChem 116
ear) in urine was found. Factors elevating hearing threshold were negatively associated with TEOAEs
parameters. Moreover, TEOAEs was positively associated with releasing of mandelic acid.
The analysis of association among ABR parameters and independent variables for 80 dB of signal level
revealed significant relationships in case of (1) LEx or styrene and wave III latency and (2) inter wave
intervals 103 and 1-5 or 103 only for LEx and styrene respectively. The latency of wave V was always
significantly correlated with Lex and styrene level for low signal levels (40-10 dB nHL). The correlation was
similar bilaterally. Prolongation of ABR I – V waves latencies correlated with age and smoking habits.
Positive association was found between latency of III wave of both sides and mandelic acid in urine.
Latencies of Vth wave at low intensities of stimulus (10-40 dB nHL) correlated with mandelic acid in urine
too. Significant association between LEx value and amplitude suppression for target potentials P300, N1 and
P1 was proved bilaterally by means of one-way ANOVA. In case of peak values of noise, LCpeak, the
aforementioned relationship co-existed with latency elongation of the waves described. The similar
associations were revealed for the ABR parameters and exposure to styrene.
.Software of posturograph SWAY 7.0 enabled to measure many factors of body sway: anterior – posterior
deviation, lateral deviation, mean deviation, sway area, sway velocity and so called sway index formulated
by producer. Moreover FFT analysis of body sway was available too. All parameters were measured but
following recommendations of Bhattacharya (1993) selected parameters of body sway to be assessed. They
are: sway area and sway velocity and additionally results of FFT analysis were assessed too. Results are
included in figures 27. Results of statistical analysis are enclosed in suitable annexes in dependence of used
test.
For body sway, significant correlations were noted between SV or SA increase or Romberg ratios and LEx
and styrene values as well as mandelic acid in urine. The other independent variables were not significantly
correlated with the dependent variables studied.
Interaction assessment was the next step of analysis. The results are given in Tables 7-9.
The odds ratios of hearing loss in the exposed workers versus controls obtained by binary logic analysis are
presented in Table 7. The results of MANOVA performed for the dependent variables related to the variables
of medical history and exposure parameters are presented in Table 8.
Only statistically significant relationships are included.
Discussion
Occupational exposure
An efficient program of decreasing the exposure to organic solvents was implemented in 70. and 80. of XX
century. Consequently, the highest admissible concentrations are presently exceeded at several working
stands only. According to Polish statistics, about 7% of workers is employed at stands for which cumulative
values of HAC (highest admissible concentration) for all the chemicals, including organic solvents, are
exceeded (MPiPS, 2003) Consequently, the serious sequelae of intoxication, like severe encephalopathies,
neuropathies and other disorders were reduced or even eliminated. It reflected also in exposition parameters
of the population studied.
Mean styrene concentration amounted 23 mg/m3 in the present study, while the TLV limit is 50 mg/m3.
Environmental studies revealed however significant variability of working conditions. The exposure to
styrene was significant in the plants manufacturing styrene containing products. These were small plants
with manual production. The excessive parameters were revealed for 51 of 134 workers exposed to styrene
and these were the only cases, that TLVs were exceeded in the plants studied. Besides the styrene, the
workers were also exposed to acetone. The highest measured levels of acetone corresponded to 0,5 of TLV
In styrene manufacturing plant, where the hermetic technological process were used, the workers could be
exposed to styrene at three stands only. Consequently, no exposure was detected from the single records of
aspirators in the subjects working at the other stands. The admissible limit of urine mandelic acid
concentration (BEI = 800 mg/g of creatinine) was not exceeded in that group of workers. The normal levels
of urine hipuric acid concentrations were however exceeded in 20% of that population, and sometimes the
upper limit was exceeded twice with the maximum value of 332 mg/g of creatinine measured. That excess
may be related to the mixture of benzene and toluene (bentol), released in course of styrene production,
though no detectable limits of bentol were measured in the air at working stands. The workers were however
exposed to the spectrum of other chemicals, like for instance the antioxidants applied for synthetic rubber
116
June 2004
NoiseChem 117
production, for which hipuric acid is not a metabolite. Efficient ventilation systems were used in the painting
plant of another company, where the expected exposure to solvent mixture could be significant.
Consequently, the environmental solvent level measured was evidently below the standards admissible, and
at some stands even below the detection level. The BEI values for phenol and hipuric acid were however
exceeded in 15-20% subjects. No working stands exposed to one solvent only were noted. Styrene exposure
was connected with simultaneous exposition to acetone or ethyl benzene. The workers from other stand were
exposed to the solvent mixture containing toluene, xylene, 2-buthanone, butyl acatate, trimethylbenzene and
1-buthanol, or to the co-existing combination of styrene, acetone and ethyl benzene. Similar situation
referred to the stands with combined noise and solvent exposure. Summarising exposure levels to solvents of
examined workers were low mostly.
The stands with noise level < 80 dB A/8h were considered as free from exposition.
The parameters of environmental exposition to organic solvents in the plants studied were typical for the
technologies applied and resembled the ones occurring for the same type of companies and technologies in
the other countries. It should however be remembered that the distribution of organic solvent concentration is
highly variable during the working day (Kumagi amd Matsunaga, 1999). Consequently, although the
measurements were performed at representative time points, and considered all the temporal changes of
concentration, the precise evaluation of exposure could be limited. It is particularly important, since the
measurement results were sometimes inconsonant with the data obtained from medical questionnaire or
interview with the employer. Although the precise measurement of exposition to organic solvents is crucial
for the assessment of dose-effect relationship, it should be remembered that the available neurobehavioral
tests are not of the equal sensitivity. The question, which of the exposure parameters (actual concentration,
metabolite level, lifetime dose) is correlated best with the adverse health effects is still not understood. The
equivocal answer on this problem was not obtained in the present study as well.
Demographic data
Besides the 15 subjects, workers subjected the study were young, with no data from history on other that
actual occupational exposure to solvents. Despite of relatively rigorous criterions of selection, the age of the
investigated groups was different: the noise-exposed subjects were the oldest and have the longest
professional experience. In case of exposure to solvent mixture its composition was however temporarily
variable. Consequently, the least variability occurred for the groups exposed to styrene, alone or combined
with noise, and for the control group. The size of remaining groups was lower, which, taking the co-existing
cofounders into account, may reflect in loss of differentiation of exposition effects. Interpretation of
exposure level was highly interrupted since it was based on single measurement of solvent concentrations
and noise levels only. The impossibility of multiple measurements of the factors discussed was related to the
worker reluctance and lack of employers’ interest with the study. The reluctance could be partly interpreted
by the reduction of plant staff, which occurred in Poland since the transformation. Consequently,
performance of non-obligatory studies on workers or occupational environment is not accepted since it
disturbs the working process in the plants with small staff. On the other hand, the worker hardly agree for
participation in the studies, being afraid of losing the job, due to potential health perturbations diagnosed in
course of the experiment. Accordingly, many workers subjected the present study intercepted the experiment
and precluded the performance of all the measurements expected.
Concentrations of solvent metabolites were additional parameter considered at exposure evaluation.
Spectrum of metabolites analysed was related to the solvents predominating in the environment of particular
groups. Consequently, mandelic acid was determined as a main metabolite of styrene, whereas phenol was
measured in case of benzene exposition and hipuric acid concentration was controlled in toluene exposed
workers. As expected, the exceeded BEI for mandelic acid in urine of worker reflected high environmental
levels of styrene. Surprisingly, also the urine concentrations of hipuric acid and/or phenol were noted in that
group, although the data on exposure to other than styrene or acetone solvents were lacking in the medical
history. Excessive concentrations of hipuric acid and/or phenol were detected in workers exposed to solvent
mixture, although the environmental level of its particular components was low. Analysis of data from
history, obtained either from workers or from the employer, failed to find out the reasons for elevated level
of the metabolites. Considering, that either toluene or xylene are benzene homologues, it is very likely that
some trace amounts of the latter were present in solvent mixture. The statement could be true also in case of
117
June 2004
NoiseChem 118
workers exposed to combinations of styrene with noise or with ethyl benzene and noise. Urine
concentrations of acetone were not measured for styrene and acetone exposed workers, since the BEI values
for acetone are not determined. The literature provides no equivocal interpretation for styrene and acetone
interactions (Apostoli et al., 1998; Lof and Johanson, 1998; Marhuenda et al., 1997; Alessio, 1996). Studying
the hearing effects of combined exposition, the authors considered only the styrene action, without coexistence of acetone taken into account. In contrary, researchers dealing with metabolism of organic solvents
do not agree if the action of the chemicals discussed is really independent. Some state that acetone has no
effect on styrene metabolism, whereas the others claim that acetone slows down the kinetics of styrene, or in contrary – that combined exposition to both the solvents results in increased elimination of mandelic acid.
Also the data on the effects of benzene on toluene metabolism or on co-existence of the latter with xylene
and other solvents, such as butyl acetate or ethylbenzene, are unequivocal (Lof and Johanson 1998; Alessio,
1996). In order to normalise the solvent metabolite levels, the urine samples for their determination were
collected from workers on 4th or 5th day of the week, following at least 5 hour occupational exposure.
Determination of metabolite level based on single urine collection and consequently, the potential individual
variability, for instance related to diet, was not considered. Many authors claim on significant individual
variability of solvent metabolism (Alessio, 1996; Symanski et al., 2001; Wenker et al., 2001).
The levels of noise exposition were similar in the groups studied, i.e. among the workers exposed to noise,
alone or in combination with solvents. Duration of noise exposure was however the longest among the noise
only exposed subjects (about 18 ± 10 years). The parameter discussed was nearly half shorter among the
noise + styrene + ethylbenzene exposed workers and more than half shorter for the remaining groups.
Consequently, the cumulated dose of noise differed significantly among the groups. The noise only exposed
workers were significantly older than subjects from the other groups. The duration of exposure in workers
exposed to solvents alone was however the shortest. Consequently, the cumulative doses of the chemicals
were low, and the workers selected, with single exceptions, have no previous episodes of occupational
exposure to solvents.
Recent studies revealed, that the low levels of exposure, even below the HAC, might also reflect in adverse
health effects (Edling and Lundberg, 2000; Morata, et al., 2002; Altmanet al., 1995; Reinhardt et al., 1997;
Vrca et al.,1996; Bhattacharya, 1993; Sack et al, 1993; Iregren, 1996). Such exposition most frequently
results in neurobehavioral disorders, including hearing or balance disorders. According to the authors cited,
the hearing, and particularly the equilibrium organ, are sensitive indicators of disorders of central, and for
some chemicals (ototoxic) also of peripheral nervous system. The present study, performed on population
exposed to relatively low solvent levels, proved that either the low concentrations or low lifetime doses of
that chemicals might reflect in disorders of both parts of the neural auditory pathway. It was not possible to
asses whether the disorders noted were reversible or not, since the present study was of the cross-sectional
character. Similarly to literature (Axellsson and Prasher, 2000) tinnitus reported by workers correlated with
noise exposure, leisure noise exposure and cholesterol . Dizziness correlated with history of head trauma,
cholesterol, and negative relation was found with coffee drinking, and exposure to acetone and noise.
Impedance audiometry
The studies on chemical effects on stapedius reflex are not dealt as extensively as tone audiometry in the
available literature, though numerous papers from the field of clinical diagnostics were published on the
problem. Impedance audiometry was employed as a complementary test rather in the present study, and
consequently, the examination protocol was relatively simple. Nevertheless, the results are worth
consideration. The thresholds of stapedius reflex were correlated negatively with the noise exposition levels,
with the significant relationships up to 2000 Hz. The correlation was significantly positive with styrene
exposure parameters above 2000 Hz by contralateral stimulation. That results indicate possible retrocochlear
location of hearing loss. All the diagnostic possibilities were not employed in the present study, since the
impedance audiometry was treated as an accessory procedure in hearing examination. According to many
authors the accuracy of that procedure is to low (Katz, 1985; Kowalska and Sulkowski, 1994).
Tone audiometry
As expected, the effect of age on results of tone audiometry was evident, though the odds ratio was not high
and most frequently ranged from 1,09 to 1,3. Similar values of the odds ratio are given in the available
118
June 2004
NoiseChem 119
literature (Śliwińska-Kowalska et al 2001b; Morata et al., 2002). Although the odds ratios were not
extremely high, the p values for all the hearing parameters were lower than 0.05, and frequently p<0,001.
Though the subjects younger than 45 years were preferred to the experimental group, 10% of older workers
was included, undoubtedly enhancing the relationships observed.
Analysis of the medical history revealed adverse effects of such factors as head trauma, otitis, arterial
hypertension, cigarette smoking and elevated cholesterol level. The hearing is adversely affected also by
alcohol consumption, though the relevant odds ratio values were not statistically significant. It should
however be noted, that the questionnaire answers on rate of alcohol drinking are usually little reliable.
Generally, the subjects conceal or lower the real level of consumption. Noise exposure is related to increased
risk of hearing loss (odds ratio >1,3), but the p-value is usually lower than 0,05. It is related to polish
legislation, which obligates the workers to use the hearing protectors in case of noise exposition.
Paradoxically, consideration of hearing protector in the present experimental model, reflected in unexpected
elevation of odds ratio for this factor, which always exceeded 1. This phenomenon is difficult to interpret.
On one hand, the frequency of protector use could be higher among the workers with already existing
hearing loss. On the other hand, the real efficiency of hearing protectors is relatively low and inconsonant
with the expected catalogue values given by the producer. Field studies revealed that the discipline of
hearing protector use is relatively low among the workers (Pawlas and Grzesik, 1990; 1995; Lundin, 1978;
Pekkarinen, 1987). Consequently, although the employers supply the workers with protectors, they are
inefficient and the hearing loss is progressing. Accordingly, the apparent increase of risk of hearing loss
related to hearing protector use, could be revealed by means of statistical analysis.
Smoking is more often recognised as risk factor to hearing (Toppila et al., 2000; Mizoue et al., 2003)
Analysis identified cigarette smoking as a risk factor of hearing loss (statistically significant for 2000 and
6000 Hz). Similar relationships were noted at the high frequencies, with the exception of acetone, which
affects hearing adversely at this band, but the relationship is not statistically significant.
The results proved that the exposure to organic solvents affects toxically the hearing and equilibrium organ,
either in its peripheral parts or at the higher levels, in the central part of neural auditory pathway. The value
of odds ratio for styrene exposition was high with the exception of 2000 Hz, for which it was lower than 1.
The results suggest adverse action of the factor discussed on hearing, but the significant values were noted
only for 4000 Hz. The studies revealed that although the styrene exposure is positively correlated with
hearing loss at the entire band of frequencies, it results with the lost at the high (above 4000 Hz) frequencies
only. Similar results were described by other authors (Muijser et al., 1988; Sass-Kortsak et al., 1995), the
experimental populations were however exposed to higher concentrations of solvents.
Odds ratio for acetone, co-existing with styrene was always lower than 1. Consequently, acetone could
positively affect the hearing threshold, though the values discussed were not statistically significant.
Exposition to benzene, co-existing with styrene at some working stands, was related to high risk of hearing
loss up to 500 Hz (high, statistically significant, odds ratio). Analysis revealed positive, but non statistically
significant, effect of exposure to ethyl benzene on the state of hearing at conventional frequencies. The odds
ratio values for toluene were very high though also non significant.
It should be noted, that although the levels of exposure and lifetime doses were usually low and did not
exceeded the admissible limits within the group studied, the subjects exposed were at risk of hearing loss.
The values of relative risk of different scenarios of exposure to solvents, alone or combined with noise,
indicate that the highest ototoxicity is related to styrene alone and than decreasingly, to styrene combined
with ethyl benzene and other mixtures. The term „solvent mixture” is however neither definable nor
equivocal. Depending on mixture composition and proportions, the toxicokinetics and subsequently the
health effects, are different (Alessio, 1996; Lof and Johanson 1998).
The results of present study suggest, that particularly the exposure may affect the hearing organ and the risk
of hearing loss is slightly higher if the exposition is combined with noise. Sass-Kortsak also didn’t find any
particular contribution of noise in combined exposed with styrene (Sass-Kortsak et al., 1995). The result is
worth consideration since the animal studies (Makite et al., 2003) revealed, that likewise for exposition to
solvent mixture and/or noise, the risk connected with mixtures is related rather to low and medium
frequencies, while the noise is particularly dangerous at medium-high range, i.e. for 2000 to 6000 Hz
moreover some inconsistency in published results could be based on differences in concentrations of solvents
used to researches. Makitie et al. (2003) and others (Lataye et al., 2003) . The results presented are
119
June 2004
NoiseChem 120
consonant with the available literature, i.e. the exposure to organic solvents affects the hearing organ and the
combined exposition to solvents and noise reflects in the more pronounced hearing loss. Styrene, as well as
the other solvents, may be ototoxic, even without the co-existence of noise (Kulig, 1990; Morata et al., 1994
b, Morata et al., 1995; Johnson, et al., 1998; Viaene, 1998; Morata et al., 2002; Sulkowski, 2002; SassKortsak et al., 1995). The hearing loss at high frequencies were positively correlated with exposition to
acetone and the relevant odds ratio for the right ear was close to 5 (4, 98) with p-value 0,015. Although
higher than 1, the odds ratio for styrene was statistically insignificant for the right ear, whereas the hearing
loss in the left ear was correlated with styrene exposure with odds ratio 2,67 and p-value 0,051. Hearing loss
was negatively correlated with the rate of mandelic acid elimination for this band. Surprisingly, statistically
significant effects of noise were found out by means of multifactor analysis. Similar results were obtained
also in the previous studies (Pawlas, 1996; Pawlas, 2000). Hearing loss was significantly correlated with age
at the band analysed (p<0,0000). Considering the environmental factors, at the band discussed the correlation
was significantly positive for noise and its all parameters, and taking chemicals into account – for mandelic
acid level and for reciprocal of its elimination rate. Also the duration of cigarette smoking affected the results
significantly. The band is however difficult to analyse, since besides the increase of auditory threshold, the
decrease of upper limit of sensitivity to the frequency should be taken into account. Unfortunately, the
equipment employed was potent to generate only three frequencies (10000, 12500 and 16000) outside the
conventional band. Morioka (Morioka et al., 1999; Morioka et al., 2000) revealed significant relationship
among the decrease of upper hearing threshold and the dose of organic solvent absorbed. The author
however used the audiometer generating signals up to 25 kHz, with 1kHz intervals. Regarding previous
studies (Pawlas, 1996; Pawlas, 2000) and other former papers (Fletcher, 1985; Dieroff, 1982; Morioka et al.,
1995; and Ahmed et al., 2001), high frequency audiometry seems to be useful for evaluation of early hearing
loss caused by the ototoxic agents, providing the examination was performed at high frequency range with 1
kHz intervals. Other authors used 1 kHz intervals, too (Morioka et al., 1995; Hallmo et al., 1994; and many
others, e.g. group of S.A. Fausti and Rappaport 1985, or Dreschler’s group, 1985).
Some other factors, besides the occupational environment parameters, were correlated with the hearing loss
and have the odds ratio values exceeding one. One of them was cigarette smoking, correlated significantly
with bilateral hearing loss at the entire band of frequency. Also logistic analysis revealed that smoking
constitutes the risk factor for the organ of hearing. Considering the entire band, the hearing loss in smokers
was on average 2-3 dB lower than for non-smokers. Hearing loss was correlated with either the number of
years the subject smoked or the number of cigarettes. The odds ratio was significant up to 6000 Hz, and up to
2000 Hz for right and left ear, respectively. The p-value of odds ratio was lower than 0,05 for the remaining
frequencies. Elevated cholesterol level was found to be another risk factor for hearing organ. The respective
p-value was lower than 0.03 and the odds ratio of average hearing loss amounted >2 and nearly 2,5 for left
and right ear, respectively. The relationships between arterial hypertension and hearing loss were however
difficult to interpret due to their ambiguous character. The values of respective odds ratio amounted either
less or more than 1. Data from history on otitis, head injuries and other than professional exposure to noise
were the other factors positively correlated with hearing loss and having the odds ratio significantly
exceeding 1. Other authors claimed on the aforementioned variables as risk factors of hearing loss (SassKortsak et al., 1995; Nakanishi et al., 2000; Stark et al., 1996). Relative risk (Table 9) for combined
exposure is slightly higher than for solvents alone. Tendency of the results is similar to the published by
Mehnert et al. (1994) but they examined other chemical (CO, PB and CS2 alone or in concerted with noise)
Relative risk due to exposure to noise alone is highest but calculation was not age adjusted. Subgroup
exposed to noise is oldest, and age is very strong cofounder. The parameters of TEOAE were correlated
with the results described.
TEOAE
Hearing examination by means of otoacoustic emission is relatively new technique in occupational medicine.
Majority of the articles published as far dealt with the noise effects on TEOAEs. Significant decrease of
reproducibility and S/N ratio related to noise exposure indicated strong effects of the factor studied on the
organ of Corti. The observation is consonant with the relevant data published by numerous authors
(Trybalska et al., 1999; Rershef et al., 1993, Hotz et al., 1993, Engdal, 1996; Attias et al., 2001). TEOAEs
were significantly decreasing with age, which is consonant with the available literature (Satoh et al., 1998;
120
June 2004
NoiseChem 121
Bonfils et al., 1988; O-Uchi et al., 1994). TEOAEs parameters were suppressed in cigarette smokers ( figure
15). Exposition to styrene and acetone however did not affect the routinely measured TEOAEs parameters,
reproducibility and signal-to-noise ratio, changing only the TEOAEs level. The results suggest that
micromechanic properties of ear decrease not only with age and increasing noise exposure (Avan et al.,
1993, Hoth and Weber, 2001; Engdal, 1996) but are also affected by styrene and ethyl benzene exposition.
Multivariate analysis revealed that the factors are significantly involved in hearing loss over 2000 Hz and
1000 Hz for right and left ear, respectively. Furthermore, such health states like otitis or arterial
hypertension, exhibit adverse effects on ear micromechanics at lower frequencies, up to 2000 Hz and 1000
Hz for right and left ear, respectively. Negative correlations of the values of reproducibility and S/N ratio
with parameters of noise and styrene concentration indicate the adverse affects of the aforementioned factors
on the state of sensory cells, and consequently prove their ototoxic properties. Positive correlation with the
rate of mandelic acid elimination indicates that the micromechanics of ear is intact, providing the quick
excretion of that metabolite. Cigarette smoking, since it decreases the level of reproducibility and S/N
values, may be considered as a risk factor with ototoxic properties. Papers of Fechter et al. (2000 and 2002)
revealed the adverse effects of carbon oxide, one of the main components of tobacco smoke, on the hearing
threshold. Moreover, the author claimed, that carbon oxide enhances harmful effects of noise even at low
concentrations. Taking into account that the carboxyhemoglobin level (CO metabolite) is elevated in
smokers, it is very likely, that hypoxia of sensory cells in hearing organ could be on of the mechanisms
responsible. Because the stimulus is click, TEOAEs are not so frequency specific as DPOAEs. Anyway, they
enable the evaluation of cochlea frequencies in significantly shorter time. Due to technical considerations,
frequency analysis is restricted to 5000 Hz. Unquestionable advantage of TEOAEs is fact, that they are
objective and, as it was proved in the present study, sensitive even to low intensities/concentrations of
environmental factors. The disadvantage is however that their application is restricted to the hearing loss up
to 35 dB HL and the reliable results are obtainable only if background noise is reduced as much as possible
in course of the examination. The question, whether TEOAEs would be useful for detection the subjects with
increased sensibility to noise and ototoxic solvents, is still not explained. The present study revealed that
TEOAEs results are correlated with the results of conventional audiometry. The procedure however seems to
be more sensitive. It outranks the distortion product acoustic otoemissions since it is much less time
consuming, although the latter enable examination at wider range of frequency (Attias et al., 2001; Engdal,
1996). Considering the results of the present study, the method described seems particularly useful for
diagnostic of early lesions resulting from solvent and noise exposition, since the relationships observed were
the most significant.
ABR
Many factors should be considered in course of correct analysis of ABR results. Parameters of stimuli
applied, body temperature, sex, age, medicines taken and alterations of hearing organ etc. affect the ABR
records (Sturzebecher et al., 1988). Wave latencies increase with the severity of hearing loss and the decrease
of stimuli level (Janczewski et al., 1991, Katz 1985, Kochanek et al., 1992, Attias and Pratt, 1985). The
aforementioned parameters however do not affect the wave intervals. Having in mind, that the hearing loss in
the exposed group was more severe than among the controls, the latencies of the exposed subjects should be
longer than the control ones. The intervals should be intact however, and consequently their alterations are
related to exposition and other factors. Besides the hearing loss, numerous factors could prolong the wave
latency. If individual factors, medical history and exposure parameters were considered in course of analysis,
age, history of otitis and leisure shooting were found out as the factors related to increase of wave latency,
which in fact is consonant with the actual knowledge. Surprisingly, either the latency or wave intervals were
correlated with use of hearing protectors. Consequently, the latter variable may be an indirect indicator of
hearing loss, since, as previously mentioned, the frequency of protector use could be higher among the
workers with already existing impairment. Increase of V wave latency at the levels of stimulation was
correlated positively with the noise exposition, particularly at its maximal values, correlation between
exposure to styrene prolongation of waves III and intervals I-III were found. Positive relationship between
mandelic acid /g creatynine and latency of wave III and interpeak latency I-III both sides was found too.
There was no relation dependent on toluene, what is inconsistent with the former study of Vrca et al. (1996).
ANOVA confirmed interaction of noise in combination with styrene on latencies as well as waves intervals.
121
June 2004
NoiseChem 122
Although results of both multivariate regression as well as multiple logistic regression regarding other
independent variables of health history and other occupational factors doesn’t confirm these relations.
Therefore these correlations seem to be rather accidental and difficult to interpretation. Our results confirm
the earlier paper of Lille et al.(1993). They reported that somatosensory evoked potential of workers exposed
to solvents exhibited changes with no effects on ABR parameters. Other investigated solvents including
toluene have not shown any effects on ABR. Concentration of hipuric acid in urine of examined workers was
higher than those one reported by Vrca et al. (1996), who found adverse influence of toluene on brain stem.
Therefore results are not univocal. Although poor wave morphology of ABR could be interpreted as
neurotoxic effects of exposure to solvents, even at low level even if workers didn’t report neurological health
problems
P300
Cognitive Event Related Potential (P300)– endogenous potentials, unlike exogenous ones such as ABR, are
dependent mainly on the psychological factors than on the parameters of stimulation applied. P300 illustrate
the changes of electric voltage, induced by the information processing. In the present study, P300 reflected
distinguishing between the infrequently (2000 Hz) and frequently (1000 Hz) appearing tones, i.e.
examination followed the classical protocol of so-called “oddball paradigm”. Several waves are
characteristic for the records of potentials described. The first one, N1, with a negative voltage, and the
following one, P300, of positive amplitude, belong to the most important. Sometimes P300 wave exhibits
two deflections, P3a and P3b. The wave P3a is not always present or the deflections are overlapping each
other. P3a is evident, if patient actively distinguishes among the stimuli (Polich, 1987). Body temperature,
seasonal variation, heart rate, exercise, medicines, nicotine, alcohol, as well as personality and intellect of the
subject, co-existence of certain neurological or metabolic (diabetes) disorders and age are considered as
factors affecting the results of P300 examination (Polich, 1998). Age-related differentiation of P300 records
amounts 1,3 ms among adults aged from 20 to 80. One-way analysis, performed in course of the present
study, revealed that either the non-target N1 wave or its amplitude decrease with age, whereas P3 is
increased either for target or non-target wave. The changes described are symmetric. Multivariate analysis
revealed relationships of certain P300 parameters with cigarette smoking, consumption of coffee and alcohol
drinking, but most of them were not statistically significant. Considering the non-target records of N1 in the
right ear, significant or near the level of significance, were positive relationships between the increase of the
wave latency and cigarette smoking or styrene and toluene exposure, whereas acetone exposition reflected in
decrease of the latency. Changes after acetone exposure are rather unexpected because usually after
exposure to solvents delay inp300 was observed (Morrow et al., 1992: Vrca et al., 1997; Steinhauer et al.,
1997) The wave amplitude decreased however with age and noise and styrene exposure. Non-target latency
of N1 in the left ear exhibited positive correlation with age and toluene and have the negative relationship
with the exposition to acetone. The alterations described were not observed for the target waves. The latency
of P3 was toluene-, benzene- and alcohol consumption-dependent, whereas practically no relationships were
found out for P3B in non-target records. Target P3b wave was altered by styrene or acetone exposure,
whereas, similar to non-target records, no significant effect of any factor studied was revealed for P3b. The
latter phenomenon may reflect low concentrations of solvents in the occupational environment and relatively
short exposure duration of the subjects studied. Consequently, the exposition occurring could be insufficient
to induce the evident alterations. The effects of isolated toluene exposure on P300 records observed during
the present study were similar to the ones described in few publications dealing with a problem (Vrca et al.,
1997, Steinhauer et al., 1997). In case of combined exposition however, the wave latencies were significantly
decreased, which was proved by means of variance analysis. Unfortunately, available literature dealing with
the effects of occupational environment on the evoked potentials is scantly. The papers published consider
rather the effects of neurotoxic factors on the cognitive potentials, while the relevant data on combined
exposure effects is lacking. More references deal with somatomotoric (SEP) or visual (VEP) evoked
potentials. It is likely, that low solvent concentrations may affect rather peripheral part of neural pathway,
working ototoxically, which in fact was proved by already discussed results of audiometry and acoustic
otoemission. Unfortunately, by means of experimental protocol described, it was not possible to conclude, if
longer exposure reflected in changes at higher level of auditory pathway and if the resulting changes were
persistent or transient.
122
June 2004
NoiseChem 123
The latencies in women are significantly shorter comparing to males, which should be considered during the
interpretation of present study results, since females were predominating among the controls and men –
among the exposed workers. Consequently, the decrease of latency among the exposed to noise and solvents
was virtually even more pronounced than it appears from a simple comparison of differences. The record
amplitude however increased with difficulties with stimulus distinguishing. The lack of relevant standards
extorted high number of controls examined. The results obtained in the control group were subsequently
considered as the reference values. Differences of latency and amplitude values, observed between the
subjects and reference group, were interpreted as the disorders in cognitive function of potentials induced by
acoustic stimulation, reflecting the aberrations in central nervous system. Undoubtedly, the changes in record
morphology, potential latencies and amplitude values, revealed by the present study, clearly indicate the
cognitive disorders in the subjects exposed.
Body Sway
The co-existence of hearing loss and equilibrium organ disorders, revealed at the present study, was also
described by other authors (Rebert et al., 1994; Franks et al., 1995; Calabrese et al., 1995). Maintenance of
upright position depends on precise processing of signals coming from a few systems. Vestibular organ is
situated in inner ear. Maintenance of upright position depends on precise processing of signals coming from
a few systems. Reflex reactions responsible for head position in reference to ground and trunk/corpus and for
adapting limbs and eyes position are triggered by afferent stimuli/impulses from receptors placed in:
vestibular organ (i), retina (ii), and muscles of neck, trunk and limbs (iii) Impulses are conducted to higher
levels of nervous system via connections with cerebellum, reticular formation and cerebral cortex. Central
nervous system co-ordinates precise collaboration of systems responsive for keeping balance and movement
control. Accordance of signals from various systems is necessary for keeping the correct posture, spatial
orientation and adapting reactions. This ability is conditioned by efficiency of central nervous processing.
Therefore balance system examination allows assessing of both the central nervous system and each single
system separately due to applying specific procedures. Balance examination depends both on physiologic
and psychological characteristics of subject (of examination). Electronystagmography (ENG) is a common
method for examining the balance system. ENG matter is to provoke nystagmus, using stimuli, which
irritate the labyrinth, and to assess nystagmus parameters. However this method is inconvenient for the
subject since it provokes vertigo. As a result of multidisciplinary collaboration an alternative method and
instrument- posturograph was built. It allows assessing the station test objectively. Postural stability
examinations vary due to different software used to process and calculate results. Frequency of sway
registration differs between various types of instruments and different times of recording are used by
researchers. Hence, raw results obtained by different sway meters cannot be compared. The result of
examinations confirmed the existing influence of noise and solvents exposure on balance system. The
highest effect was exerted by exposition to styrene and both styrene and ethylbenzene. Solvents which
composite was toluene resulted in a little bigger increase of body sway area during the concomitant
exposition to noise. Noise exposure of exposed workers was very high in contrary to exposure to solvents. So
dominating effect of noise should not be surprising. Using thick foam more differentiated effects of exposure
to particular factor. These conditions of examination indicated that vision and vestibular system of exposed
to solvents were affected more compare to exposed to noise and solvents, however noise affected both
system the most. Increasing of frequency of sway above 1 Hz could be interpreted as a results of changes in
central nervous system. Sway frequency increased the most in group exposed to combined noise and
solvents. After the combined exposition both to noise and solvents, the frequency of body sway increased,
what is common for toxic effects exerted on central nervous system. To sum up, exposition to organic
solvents, even in low concentrations, exerts influence on both central and peripheral nervous system and
impairs vision system, which susceptibility is widely known, balance system and hearing at all frequencies.
(Bhattacharya et al., 1987; Ledin et al., 1989, Nageris et al., 2000; Yokoyama et al., 1997) Multivariable
regression indicated that noise is important factor influencing body sway. It is consistent with literature
Results of the study are consistent with literature on the effects of solvents and noise on hearing and balance.
Hearing loss may be cochlear and retrocochlear (Moller et al., 1989, Morata et al., 1993), and balance is
impaired too (Bhattacharya et al., 1987; Bhattacharya, 1993; Morata et al., 1995; Calabrese et al., 1996;
123
June 2004
NoiseChem 124
Yokoyama et al., 1997; Aylott and Prasher, 2002). It should be emphasised that our results, which were
obtained even at low levels of solvents and relatively short lifetime exposure to them, showed induced
changes.
Although the accuracy of organic solvent exposure assessment is crucial for determination of the dose-effect
relationship, the latter was not useful in every test performed under industrial conditions, even in course of
the same experiment (Seeber et al., 1996). The question, which of the exposure parameters (actual
concentration, metabolite level, lifetime dose) is correlated best with the adverse health effects, is still not
understood. The equivocal answer on this problem was not obtained in the present study as well. By means
of logistic analysis, the most significant relationship was found with the actual solvent concentration, and
then, decreasingly, with the urine metabolite level and the lifetime dose of solvents. It could be related to
several reasons. The population studied was young and usually without previous occupational exposure,
which could be the explanation for simple relationship between the state of hearing and equilibrium organ
and the parameters of actual exposition. Weaker relationships with the metabolite level could be more less
related to individual variability. Also the other factors connected with the metabolite level, by their potential
effects on solvent toxicokinetics, should be considered. Spearman rank analysis revealed positive correlation
with the habits of cigarette smoking and coffee drinking, while the negative relationships were noted with
BMI, indicating both the state of nutrition or the level of adipose tissue, known depot for solvents. The diet
was not controlled in course of the study, though its effect on toxicokinetics of xenobiotics, and consequently
on individual variability, is indubitable. It is probably one of the reasons for the lack of unequivocal
relationship between the metabolite level and consequences of the exposition. If no changes occur in
technologies or equipment applied, the time-weighted air concentration of solvents is stable. Consequently,
the metabolite level reflects not only the exposition itself, but also the temporal state of the body and balance
between absorption and excretion of solvent. Correlations between the metabolite level and solvent
concentrations in occupational environment were however the strongest (p<0,000).
Single measurement of metabolites is not satisfactory because of exposure variability during a day as well as
during a longer time. Inter-subjects and inner-subjects variability could give false information on level of
exposure randomly positive as well as negative. Factors influencing metabolism of solvents like medicines,
nutrition, alcohol usage or sampling time influence on level of metabolites too. Other problem is connected
with validity of information obtained form employees and employers. In our opinion sometimes it is
intentional sometimes-not. Employers would like to be seen as caring about good occupational hygienic
environment. Workers try to hide other non-occupational exposure in a case of applying for being developed
occupational disease. Results of environmental measurements compared with biomonitoring results showed
some discrepancies between them. Answers related to smoking habits are usually valid but information on
alcohol consumption is very shaming in common opinion, so answers related to alcohol should be accepted
with very limited confidence.
Numerous difficulties and obstacles appeared during realisation of the project. Constant reorganisation of the
plants and resulting large rotation of the workers eliminated many of originally selected subjects at the first
stage of experiment. Proprietary transformation and health care reform extremely delayed and impeded
project realisation.
Conclusions
As has been expected, the effects of noise and age on hearing are obvious, even in as relatively young group
as examined in this study. Even low levels of solvents affects hearing and balance. Hearing thresholds of
workers exposed to solvents were elevated comparing to the unexposed persons Combined exposure to noise
and solvents reflected in greatest hearing loss, but contribution of noise seemed to be small. The results
suggest that balance is affected by exposure to solvents and noise and to noise alone, too. There are
differences in effects of exposure to styrene and exposure to solvent mixture. Styrene seems to be strong
ototoxic for human. Results of the present study are consonant in that matter with previuosly obtained by
animal experiments. Other parts of nervous system are affected too. Changes are located in vision as well as
in central part of nervous system. Our study shows that in population of workers exposed to low levels of
organic solvents without noise exposure, hearing threshold is significantly elevated comparing to nonexposed population. Hearing loss may be both of cochlear and retrocochlear origin. Hearing is susceptible to
solvents, and especially to styrene, even at low levels without any neurological symptoms co-existing.
124
June 2004
NoiseChem 125
Consequently, workers exposed to solvents should be included into periodical hearing examination focused
on hearing loss prevention. Styrene alone constitutes risk for hearing and the results obtained confirmed its
ototoxicity. Noise alone is a strong ototoxic but it affects higher auditory pathway, too. High- frequency
audiometry with 1 kHz step in frequencies should be considered as ta ool for early detection of noise induced
hearing loss. Method of measuring transient evoked otoacoustic emissions is very sensitive for early changes
in hearing but it is not specific for any particular factor. This method of hearing examination is not timeconsuming, and its sensitivity to factors impacting cochlea is the additional advantage. It gives however
limited frequency information up to 5 kHz, but simultaneously sensitive enough to detect early solventinduced changes. ABR, as well as P300, have almost no practical application in assessment of hearing status
in field study, but they are useful for clinical purposes. Both the methods are useful to differentiate between
hearing loss resulting from cochlear or retrocochlear damages. Accordingly, they are useful for diagnostic
goals but not for hearing conservation programme in occupational health services. Considering the
morphology of changes detected, however, both the methods are sensitive for early neurological changes in
auditory pathway. Either hearing or balance abnormalities were correlated rather with exposure parameters
than with lifetime doses, but design of the study does not allow to answer whether the changes observed are
permanent or reversible partly or completely. Although the results revealed interactions between noise and
solvents, but they were not so evident to postulate reduction in TLVs for combined exposure, in a case of
styrene and noise working together.
Acknowledgments
This investigation was support by European Commsion CONTRACT N°: QLK4-CT2000-00293 Noise and
Industrial Chemicals: Interaction Effects on Hearing and Balance –NOISECHEM and Polish State
Committee of Science No 673/E-225/SPUB-M/5.PR UE/DZ 227/2001-2003.
Authors express their special grateful and thanks to audiological technicians, Dorota Kuś and Violetta
Maksymowicz, for their skillful assistance.
We thank workers of Analytical Laboratory of the Institute for urine analysis and Barbara Socha and Danuta
Majewska for environmental factors measurements.
We also thank the companies and all the workers who agreed to participate in the study for their time and
cooperation.
Literature cited
Abbate C., Giorgiani C., Munao F.: (1993) Neurotoxicity induced exposure to toluene. An
electrophysiological study. Int. Arch. Occup. Environ. Hlth 64, 245 – 254
Ahmed HO, Dennis JH, Badran O, Ismail M, Ballal SG, Ashoor A, Jerwood D (2001) High-frequency (1018) kHz hearing thresholds: reliability, and effects of age and occupational noise. Occup. Mmed (London)
51, 245-258
Alessio L., Apostoli P., Crippa M.: (1994) Multiple exposure to solvents and metals. Occup. Hyg 1,127 151,
Alessio L (1996): Multiple exposure to solvents in the workplace. Int Occup. Environ Health 69, 1-4
Altman L, Neuhann F-F, Kramer U, Witten J, Jermann E, (1995) Neurobehavioral and neurophysiological
outcome of chronic low-level tetrachloroethene exposure measured in neighbourhoods of dry cleaning shops.
Environmental Research 69, 83-89
Apostoli P, Alessandro G, Alessio L, (1998) Metabolic interferences in subjects occupationally exposed to
binary styrene-acetone mixtures. Int. Arch. Occup. Environ. Health 71, 445-452
Attias J Pratt H (1985) Auditory-evoked potential correlates of susceptibility to noise-induced hearing loss.
Audiology 24, 149-156
Attias J, Horovitz G, El0Hatib N, Nageris B (2001) Detection and clinical diagnosis of noise-induced
hearing loss by otoacoustic emissions Noise and Health19-323, 12
AugustyńskaD, Pośniak M (eds) Hazardous factors in occupational environment TLVs (2001) ( in Polish)
ed. CIOP, Warszawa
Avan P., Bonfils P., Elbez M Erminy M (1993) Exploration of cochlear function by oto-acoustic emissions:
relationship to pure-tone audiometry. Progress in Brain Research 97, 67 – 75
Aylott S, Prasher D, (2002) Solvents impair balance in man. Noise & health 4,14,63-72
125
June 2004
NoiseChem 126
Axelsson A, Prasher D (2000) Tinnitus induced by occupational and leisure noise Noise and Health 8, 497 54
Bazydlo - Golinska G.: (1993), Wplyw rozpuszczalnikow organicznych na ucho wewnetrzne . Med. Pracy
44, 69 – 78
Bhattacharya A, Morgan R, Shukla R, Ramakrishanan HK, Wang L ( 1987)Non-invasive estimation of
afferent inputs for postural stability under low levels of alcohol Annal of Biomedical Eng. 15, 533-550
Bhattacharya A, (1993) Quantitative posturography as early tool for chemical toxicity In: Biological
Monitoring eds. S. Quelle van Nostrand Reinhold 421-435
Bond JA. (1989); Review of the toxicology of styrene. Crit Rev In Toxicol 19: 227-249.
Bonfils P., Bertrand Y., Uziel A.: (1988) Evoked Otoacoustic Emissions; Normative data and presbycusis.
Audiology, 27, 27-35
Calabrese G., Martini A., Sessa G., Cellini M., Bartolucci G.B., Marcuzzo G., De Rosa E.: (1996)
Otoneurological study in workers exposed to styrene in the fiberglass industry, Int. Arch. occup. Environ.
Health, 68, 219-223
Campo P, Pouyatos B, Lataye R, Morel G (2003) Is the aged rat ear more susceptible to noise or styrene
damage than the young ear? Noise and Health 5, 19, 1-18
Cappaert NL, Klis SF, Muijser H, et al. (1999) The ototoxic effects of ethyl benzene in rats. Hear Res; 137:
91 - 102.
Cary R, Clarke S, Delic J (1997) Effects of combined exposure to noise and toxic substances- critical review
of the literature Ann. Occup Hyg 41,4, 455-465
Chang S-J Shih TS, Chou TC Chen CJ Chang HY Sung FC (2003) Hearing loss in workers exposed to
carbon disulphide and noise Environmental health Perspect. 11,13,1620-1624
Crofton KM, Zhao X. (1993;) Mid-frequency hearing loss in rats following inhalation exposure to
trichloroethylene: evidence from reflex modification audiometry. Neurotoxicol Teratoi 15: 413-423.
Crofton KM, Lassiter TL, Rebert CS. (1994) Solvent-induced ototoxicity in rats: an atypical selective midfrequency hearing deficit. Hear Res; 80: 25-30.
Crofton KM, Zhao X. (1997) The ototoxicity of trichloroethylene: extrapolation and relevance of high
concentration short duration animals' exposure data. Appl Toxicol; 38: 101-106.
Cruickshanks KJ, Klein R, Klein BEK, Wiley TL, Nondahl DM, Tweed TS (1998) Cigarette smoking and
hearing loss: the epidemiology of hearing loss study JAMA 279 21, 1715-1719
Davis RR, Murphy Wj, Snawder JE, Striley CAF, Henderson D, Khan A, Krieg EF ( 2002) Susceptibility to
the ototoxic properties of toluene is species specific. Heairn Research 166, 24-32
Dick F, Semple S, Osborne A, Soutar A, Seaton A, Cherrie JW., Walker LG, Haites N, Organic solvent
exposure, genes, and risk of neuropsychological impairment. (2002) QJ Med. 95,6, 379-387
Dieroff H-G.: (1982) Behaviour of high-frequency hearing in noise., Audiology,21,83- 92,
Dreschler W.A., v.d.Hulst R.J.A.M., Tange R.A., Urbanus N.A.M.: (1985);The role of highfrequency audiometry in early detection of ototoxicity. Audiology,24,387-395,
T. Dutkiewicz, J. Piotrowski, J. Kęsy-Dąbrowska (1964). „Chemiczne badania materiału biologicznego w
toksykologii przemysłowej” PZWL, W-wa
Dutkiewicz T (1968). „Chemia toksykologiczna” PZWL, W-wa
E N 26189:191 ISO 6189:1983 Acoustics - Pure tone air conduction threshold audiometry for hearing
conservation purposes
Edling C, Lundberg P, (2000)The significance of neurobehavioral tests for occupational exposure limits: an
example form Sweden. NeuroToxicology 21, 653-658
Emmett E, Frank A, Gochfeld M, Hessl SM. (1995) Year Book of Occupational and Environmental
Medicine. Mosby- Year Book Inc., St. Louis, pp. 114-117.
Engdahl B., Arnesen A., Mair I., ( 1993) Reproducibility and short-term variability of TEOE. Scand Audiol,
24, 99-104
Engdahl B. (1996) ;Clinical Application of Otoacoustic Emissions ed. Ulleval University Hospital, Oslo
Environmetal Health Criteria (EHC) No186 Ethyl benzene, WHO , Geneva 1996
126
June 2004
NoiseChem 127
Environmental Health Criteria (EHC) No190 Xylenes, WHO, Geneva 1997
Environmental Health Criteria (EHC) No52 Toluene, WHO , Geneva 1985
Fechter LD Chen G-D, Rao D ( 2000) Characterising conditions that favour potentiation of noise induced
hearing loss by chemical asphyxiants Noise and Health 3,9, 11-21
Fechter LD Chen G-D, Rao D ( 2002) Chemical asphyxiants and noise. Noise and Health 4,1 4, 49-62
Fishbein L. (1985) An overview of environmental and toxicological aspects of aromatic hydrocarbons, IV
Ethylbenzene. Sci Total Eniron. 44, 269-,287
Fletcher J.L (1985).: A history of high-frequency hearing,research and application., Seminars in Hearing,
6.4, 325-329;:
Franks JR, Morata TC (1996). Ototoxic effects of chemicals alone or in concert with noise: a review of
human studies. In: Axelsson A, Borchgrevink HM, Hamernik RP, Hellstrom PA, Henderson D, Salvi R,
(Eds.), Scientific basis of noise-induced hearing loss. Thieme, New York, 472 pp.
Gerr F, Letz R (1998): Organic solvents in: Environmental and occupational medicine ed. Wlliams N Ropm
Lippincoll-Raven Publishers, Philadelphia, 1091-11-08
Goltz A, Westerman ST, Westerman LM, Goldenberg D, Netzer A, Wiedmyer T, Fradis M, Joachism Z (
2001) The effects of noise on the vestibular system Amer J Otolaryngol 22,3 190-198
Gonsalvez CL Polich J (2002) P300 amplitude is determined by target –to-target interval. Psychophysiology
39,3, 388-389
Hallmo P, Borchgrevink HM, Mair IWS (1994):Extended high-frequency thresholds in noise-indeuced
hearing loss Scand Audiol 24, 47-52
Hirata M., Ogawa Y., Okayama A., Goto S.: (1992) Changes in auditory brainstem response in rats
chronically exposed to carbon disulphide, Arch. Toxicol., , 66, 334-338
Hoth S, Weber FN, (2001) The latency evoked otoacoustic emissions : Its relation to hearing loss and
auditory evoked potentials. Scand Audiol 30.3 173-183
Hotz MA, Probst R, Harris FP, Hauser R M (1993) monitoring of the effects of noise exposure using
transiently evoked otoacoustic emissions. Acta Otolaryngol (Stockh) 113, 478-482
Iregren A, (1996) Behavioral methods and organic solvents: Questions and consequences. Environmental
Health Perspectives 104, Suppl2 361-366
ISO 1999:1990, Acoustics – Determination of occupational noise exposure and estimation of noise induced
hearing impairment.
Jacobsen P, Hein HO, Suadicani P, ParvingA., Gyntelerg F. (1993);Mixed solvent exposure and hearing
impairment: an epidemiological study of 3284 men. The Copenhagen male study Occup Med 43: 180-184.
Janczewski G Kochanek K, Dawidowicz J Tanzariello A, Dobrzynski P, Bardadin J ( 1991) The effect of
noise on the human auditory brainstem responses, Relations betweeb temporary and permanent threshold
shift ( in Polish) Medycyna Pracy 42,1, 37-42
Jaspers, R.M.A., Muijser, H., Lammers, J.H.C.M., Kulig, B.M. (1993) .Mid-frequency Hearing Loss and
Reduction of Acoustic Startle Responding in Rats Following Trichloroethylene Exposure. Neurotoxicol
Teratol, 15, 407-412,
Johnson A-C, Juntunen L, Nylen PR, Borg E, Hoglund G (1988). Effect of interaction between noise and
toluene on auditory function in the rat. Acta Otolaryngol, 105:56-63.
Johnson A-C, 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 (Stockh), 109:34-40.
Johnson AC, Canlon B. (1992) Auditory sensitivity in rats exposed to toluene and/or acetyl salicylic acid.
Neuroreport; 3: 1141-1144.
Johnson AC, ( 1994) The ototoxic effect of toluene and influence of noise, acetylic acid, or genotype. Scand
Audiol 23 Suppl 39,
Johnson A-C, Nylen PR (1995). Effects of industrial solvents on hearing. Occup Med: State of the Art
Reviews, 10(3): 623-640.
Johnson A-C, Morata TC, Andersson IM, Nylen PR , Hagerman B, Lindh T, Svensson EB (1998).Hearing
loss after exposure to styrene and noise: a pilot study in : Advances in Noise research Biological effects of
noise ed D. Prasher , L, Luxon,Vol1, ch 27, 280-294
Katz J (1985) Handbook of clinical audiology ed. Williams&Wilkins, Baltomore, 423-495
127
June 2004
NoiseChem 128
Kochanek K, Grzanka A Dawidowicz J Jaskiewicz M Zając J Mika U Śliwa L ( 1992) The effect of brief
tone envelopes on ABR and behavioral thresholds (Otolaryngol Pol 46, 3 296-301
Kostrzewski P, (1999) Safe working conditions in small production plants:assessment of exposure to styrene
(in Polish) Medycyna Pracy50, 5, 403- 408
Kowalska S, Sułkowski W (1994) Wartośc audiometrii impedancyjnej w audiologii i otoneurologii.
Medycyna Pracy 45, 4, 333-341
Kulig, B.M. 1990.Methods and issues in evaluating the neurotoxic effects of organic solvents. In: R.W.
Russell. P.E. Flattau and A.M. Pope (Eds). Behavioral Measures of Neurotoxicity, National Academy of
Sciences, 317-322
Kumagi S, Matsunaga I, (1999), Within –shift variability of short-term exposure to organic solvents in
indoor workplaces Am Ind Hyg, Assoc. J 60, 16-21
Lataye R, Campo P. (1997) Combined effects of a simultaneous exposure to noise and toluene on hearing
function. Neurotoxicol Teratol; 19: 373-82
Lataye R, Campo P, Loquet G (2000);. Combined effects of noise and styrene exposure on hearing function
in the rat Hear Res 139( 1-2): 86-96.
Lataye R, Campo P,Pouyatos B, Cossec B, Blachere V, Morel G (2003) Solvent ototoxicity in the rat and the
guinea pig. Neurotoxicology and Teratology 25, 39-50
Ledin T, Odkvist LM, Moller C ( 1989 ) Posturography findings in workers exposed to industrial solvents.
Acta Otolaryngol 107 5-6, 357-361
Li HS. Johnson AC, Borg E, Hoglund G. (1992) Auditory degeneration after exposure to toluene in two
genotypes of mice. Arch Toxicol; 66. 382-386.
Lille F, Margules S, Mallet A, Deschamps D, Garnier R, Dally S (1993) Evoked potentials in workers
exposed to organic solvents Electromyogr clin Neurophysio. 33, 279-283
Lof A, Johanson G: ( 1998) Toxicokinetics of organic solvents: a review of modifying factors Critical
Review in Toxicology 28(6) 571-650
Loquet G, Campo P, Lataye R. (1999) Comparison of toluene-induced and styrene-induced hearing losses.
Neurotoxicology and Teratology; 21(6): 689-697
Lundin R.: (1978) The effectiveness of hearing protectors in practical situations. Bilsom internal
Makitie A. Pirvola U. Pyykko I. Sakakibara H. Riihimaki V. Ylikoski J.( 2002) Functional and
morphological effects of styrene on the auditory system of the rat. Archives of Toxicology. 76(1):40-7,
Makitie AA, Pirvola U, Pyykko I, Sakakibara H, Riihimaki V, Ylikoski Y. (2003) The ototoxic interaction of
styrene and noise. Hearing Research 179, 9-20
Marhuena D, Prieto MJ, Periago JF, Marti J, Perbellini L, Cardona A, (1997) Biological monitoring of
styrene exposure and possible interference of acetone co-exposure. Int. Arch. Occup. Environ. Health 69,
455-460
Mehnert P., Fritz M., Griefahn B.: (1994) Noise - induced hearing loss and ototoxic agents. Arch Cof
Complex Environ Studies( ACES) ES 6, 1 –11
Mizoue T, Miyamoto T, Shimizu T ( 2003) Combined effect of smoking and occupational exposure to noise
on hearing loss in steel factory workers Occup Ernviron Med. 60, 56-59
Moller C, Odkvist LM, Thell J Larsby B, Hyden D., Bergholtz L (1989). Otoneurological findings in
psycho- organic syndrome caused by industrial solvents. Acta otolaryngologica 107, 5-12
Moller C, Odkvist LM, Larsby B, Tham R, Ledin T, Bergholtz L (1990). Otoneurological findings in
workers exposed to styrene. Scand J Work, Environ Health, 16:189-194.
Morata T.C.: (1989), Study of the effects of simultaneous exposure to noise and carbon disulphide on
workers’ hearing, Scand. Audiol, 18, 53-58
Morata TC, Dunn DE, Kretschmer LK, Lemasters GK, Sanstos G.K.. (1991). Effects of simultaneous
exposure to noise and toluene on workers’ hearing and balance. In Fechter L.D. Eds.), Proceed of IVth
Inetrn. Conf. on Combined Environ. Factors 81 – 86
Morata TC, Dunn DE, Kretschmer LK, Lemasters GK, Keith RW (1993 a). Occupational exposure to
organic solvents and noise: effects on hearing. Scandinavian Journal of work, Environment and Health 19,
245- 254.
Morata TC, Dunn DE, Kretschmer LW, et al. (1993b) Effects of occupational exposure to organie solvents
and noise on hearing. Scan J Work Environ Health; 19(4): 245-254.
128
June 2004
NoiseChem 129
Morata TC, Dunn DE, Sieber WK (1994 a) occupational exposure to noise and ototoxic organic solvents
Archiv Environ Health 49(5) 359-365
Morata TC, Franks JR, Dunn DE (1994 b). Unmet needs in occupational hearing conservation. The Lancet,
344(8920):479.
Morata TC, Nylen PR, Johnson A-C, Dunn, DE (1995). Auditory and vestibular functions after single or
combined exposure to toluene: a review. Archives of Toxicology, 69:413-443
Morata TC, Fiorini AC, Colacioppo S, et al. (1997) Toluene-induced hearing loss among rotogravure
printing workers. Scand J Work Environ Health; 23: 289-98.
Morata TC Johnson A-C, Nylen Psvensson E, Cheng J, Krieg EF, Lindblad A-, Ernstgard L, Franks JR,
(2002) Audiometric findings in workers exposed to low levels of styrene and noise JOEM 44 (9) 806-814
Morioka I, Miyashita K, Gowa Y, Takeda S. (1995); Evaluation of noise-induced hearing loss by reference
to the upper limit of hearing. Int. Arch Occup. Environ Health 67): 301-304
Morioka I, Kuroda M, Miyashita K, Takeda S. (1999) Evaluation of organic solvent ototoxicity by the upper
limit of hearing. Arch Environ Health; 54(5): 341-46.
Morioka l, Miyai N, Yamamoto H, Miyashita K. (2000); Evaluation of combined effect of organic solvents
and noise by the upper limit of hearing. Industrial Health 38(2): 252-257.
Morrow LA, Steinhauer SR, Hodgon MJ, (1992) Delay in P300 latency in pateints with organic solvent
exposure. Arch Neurol 49,3, 315-320
MPiPS ( Ministry of Labour and Social Policy) (2003) Assessment of occupational safety and health in 2002,
ed.: MpiPS, Warszawa 2003
Mujiser H. Hoogendijk E, Hooisma J. (1988),The effects of occupational exposure to styrene on highfrequency hearing thresholds. Toxicol 49: 331-340
Muijser H, Lammers JHCM, Kulig BM (1994). Synergistic effects of combined exposure to
trichloroethylene and noise on hearing in the rat. TNO Nutrition and Food Research Institute- Toxicology
Division. Annual Report, Toxicology 1993-1994:53.
Murata K., Araki S., Yokoyama K., Maeda K.: (1991) Autonomic and peripheral nervous system
dysfunction in workers exposed to mixed organic solvents, Int. Arch. Occup. Environ. Health, , 63, 335-340
Nageris BI, Attias J, Feinmesser R (2000) Noise –induced vestibular dysfunction Noise and Health 3,9, 4548
Nakanishi N, Okamoto M, Nakamura K, Suzuki K, Tatara K (2000) Cigarette smoking and risk for hearing
impairment: a longitudinal study in Japanese male office workers JOEM 32, 11 1045-1049
Niklasson M, Moller C, Odkvist LM, Ekberg K, Flodin U, Dige N, Sklodstig A ( 1997)Are deficits in the
equilibrium system relevant to the clinical investigation of solvent induced neurotoxicity? Scand J Work
Environ Health 23, 206-13
Nylen P, Hagman M. (1994a) Function of the auditory and visual systems, and of peripheral nerve, in rats
after long-term combined exposure to n-hexane and methylated benzene derivates. II. Xylene. Pharmacol
Toxicol; 74: 124-129.
Nylén P. (1994b) :Organic solvent toxicity in the rat; with emphasis on non additive effects in the nervous
system (Thesis). Arbete och Hälsa 3
Nylén P, M Hagman, A-C Johnson. (1995) Function of the auditory system, the visual system, and peripheral
nerve after long-term combined exposure to toluene and ethanol in rats. Pharmacology & Toxicology; 76,
107-111.
Odkvist LM, Bergholtz LM, Ahlfeldt H, Andersson B, Edling C, Strand E (1982). Otoneurological and
audiological findings in workers exposed to industrial solvents. Acta Otolaryngol, (Suppl. 386):249-251.
Odkvist LM, Arlinger SD, Edling C, Larsby B, Bergholtz LM (1987). Audiological and vestibulooculomotor findings in workers exposed to solvents and jet-fuel. Scand Audiol, 16(2):75-84.
O-Uchi T, Kanzaki J., Satoh Y, Uchi T., Yoshihara S .Ogata A, Inoue Y, Mashino H (1994) Age- related
changes in evoked otoacoustic emissions in normal-hearing ears. Acta Otolaryngol (Stockh) Suppl 514, 8994
Pawlas K., Grzesik J.: (1990) Efficiency of ear protector in laboratory and real life condition. Inter. Archiv.
Occup. Environ. Health, 62, 323-326.
Pawlas K.: . (1995) Skuteczność ochronników słuchu w praktyce (Efficiency of hearing protectors in
practice, in Polish) Ochrona Pracy. Atest, 1. 13 – 15
129
June 2004
NoiseChem 130
Pawlas K : (1996) „ High Frequency audiometry above 8 kHz in prevention of hearing damage due to
impulse noise” ( in Polish0 ed IOMEH, Sosnowiec,
Pawlas KM, Bronder A, Powazka E (1997) Acoustic energy and other factors influencing hearing of workers
exposed to industrial noise. Book of abstracts AIHA &ACGIH Conference and Exhibition,: Dallas USA
17.05 - 23.05.,ed. AIHA &ACGIH,81
Pawlas K. (2000) Noise - induced hearing loss after exposures to different kind of Noise. NOPHER An
International Symposium on Noise -Induced Hearing Loss Basic mechanisms, prevention and control,
Cambridge. 7 - 10. 07.2000 Proceed. p38
Pekkarinen J.: (1987) ;Industrial impulse noise, crest factor and the effect of
earmuffs.,
Am.Ind.Hyg.Assoc.J.,48,861-866,
Pfaffi P, Saamanen (1993) The occupational scene of styrene. In: Butadiene and Styrene: Assessment of
Health Hazards IARCScientific Publication No 127 ed Sorsa M et al Lyon 15-26
Pilecki W, Mical-Strąk M Bogucki (2002) Wpływ wybranych parametrów stymulacji na wyniki badań
BAEP i ich kliniczne znaczenie (Effects of selected stimuls parameters on BAEP results and its clinical
meanings ( In Polish) Proceed IX Conference KOWBAN 2002, Wrocław 91-94
Polich J (1987) Task difficulty, probability, and inter stimulus interval as determinants of P300 from auditory
stimuli. Electorencephalogr Clin Neurophysiol 68, 311-320
Polich J ( 1989)Theoretical overview of P3a and P3b in detection of change: Event-related potential and
fMRI findings edJ Polich Kluwer Academic Press: Boston 83-98
Polich J, (1998) P300 clinical utility and control of variability J Clin Neurophysiol
15, 14-33
Polich J, Herbst K L (2000) P300 as a clinical assay: rationale, evaluation , and findings Intern J
Psychophysiol,38, 3-19
Powązka E., Zahorska - Markiewicz B., Pawlas K. (2002) A cross sectional study of ccupational noise
exposure and blood presure in Steelworkers Noise and Health 5.17, 15-22,
Pryor GT, Dickinson J, Feeney E, Rebert CS (1983). Transient cognitive deficits and high-frequency hearing
loss in weanling rats exposed to toluene. Neurobehav Toxicol Teratol, 5(1):53-57.
Rappaport B.Z., Fausti S.A., Schechter M.A., Frey R.H., Hartigan P.: (1985) Detection of
ototoxicity by high-frequency auditory evaluation., Seminars in Hearing, 6, 369 -377,
Rebert C.S., Day V.L., Matteucci M.J., Pryor G.T.: (1991) Sensory-evoked potentials in rats chronically
exposed to trichloroethylene: predominant auditory dysfunction, Neurotoxicology and Teratology, , Vol.13,
83-90
Reinhard F., Drexler H., Brickel A., Claus D., Ulm K., Angerer J., Lehnert G., Neundorfer B.: (1997a)
Elecktrophysiological investigation of central, peripheral and autonomic nerve function in workers with
long-term low-level exposure to carbon disulphide in the viscose industry, Int. Arch. Occup. Environ.
Health, 70, 249-256
Reinhardt F., Drexler H., Brickel A., Claus D., Angerer J., Ulm K., Lehnert G., (1997b) Neundorfer B.:
Neurotoxicity of long-term low-level exposure to carbon disulphide: results of questionnaire, clinical
neurological examination and neuropsychological testing, Int. Arch. Occup Environ. Health, 69, 332-338
Reshef (Haran) I, Attias J, Furst M (1993) Characteristic of click-evoked otoacoustic emissions in ears with
normal hearing and with noise-induced hearing loss. Brit.J of Audiology 27, 387-395
Rybak L.P. (1992) Hearing; the effects of chemicals. Otol. Head and Neck Surgery 106, , 677- 686
Sack D, Linz D, Shukla R, Rice C, Bhattacharya A, Suskind E (1993) Health status of pesticide applicators:
postural stability assessments. JOM 35, 1196-1202
Sass-Kortsak .AM, Corney PN, McD. Robertson J. . (1995) Ań investigation between exposure to styrene
and hearing loss. Ann Epidemiol; 5(1): 15-24.
Satoh Y, Kanzaki J., Uchi T., Yoshihara S .: (1998) , Age- related changes in transiently evoked otoacoustic
emissions and distortion product otoacoustic emissions in normal-hearing ears. Auris Nasus Larynx . 121130
Schwartz DM, Berry GA (1985) Normative Aspects of the ABR in The Auditory Brainstem Response
Jacobsen JT( eds) Taylor&Francis, London, 65-98
Seeber A., Sietmann B, Zupanic M. (1996) In search of dose-response relationships of solvent mixtures to
neurobehavioral effects in paint manufacturing and painters Food and Chemical Toxicology 34, 1113-1120
130
June 2004
NoiseChem 131
Shupak A, Bar-El E, Podoshin L, Spitzer O, Gordon R, Ben-David J ( 1994) Vestibular findings associated
with chronic noise induced hearing impairment Acta Otolaryngol (Stockh) 114, 579-585
Simonsen, L. and S.P. Lund (1995) Four weeks inhalation exposure to n-heptane causes loss of auditory
sensitivity in rats. Pharmacol.Toxicol. 76: 41-46
Śliwińska-Kowalska M, Bilski B, Zamysłowska-Szmytke E, Szymczak W, Kotyło P, Wesołowski W,
Pawlaczyk-Łuszczyńska M, Dudarewicz A, Fiszer M. (2000) Ocena uszkodzeń słuchu u pracowników
narażonych na mieszaniny rozpuszczalników organicznych w przemysle farb i lakierów Medycyna Pracy
50,1 , 1-10
Sliwińska- KowalskaM Zamyslowska- Szmytke E, Szymczak W, Kotyło P, Fiszer M, Wesolowski W,
Pawlaczyk_ Łuszczyńska M, , (2001) Ocena ryzyka wystąpienia zawodowego uszkodzenia słuchu w
przypadku łacznego narażenia na hałas i rozpuszczalniki (Risk assessment of hearing damage after
combined exposure to noise and solvents. In Polish) raport SPR 041040
Smith LB, Bhattacharya A, Lemasters G, Succop P, Puhala E,Medvedovic M, Joyce J ( 1997) Effect of
chronic low-level exposure to jet fuel on postural balance of US Air Force personel J. Occup Envieron Med.
39,7, 623-632
Stark J, Aalto H, Pyykko I, Ishizaki H (1994) The effect of low frequency noise on postural stability ACES
6(1-2) 83-88
Starck, J.,Pyykköö, I.,Toppila, E. (1996) , Individual risk factors in noise-induced hearing loss. Proceedings
of the 25th International Congress of Occupational Health, Stockholm September 15-20, Book II, p. 46.
Starck, J., Toppila, E Pyykköö, I.,(1999) Smoking as a risk factor in sensory neural hearing loss among
workers exposed to occupational noise Acta Otolaryngol (Stockh) 119, 302-305
Steinhauer SR, Morow LA, Condray R, Dougherty GG (1997) Event-related potentials in workers with
ongoing occupational exposure Biol Psychiatry 42, 854-858
Sturzebecher E, Werbs M (1988) Influence of age, sex, and hearing loss on audiotry brain stem response (
ABR) latencies, Scand Audiology 17, 248-250
Sułkowski WJ (1979) Badania nad przydatnością kliniczną audiometrii i elektronystagmografii w
diagnostyce przewlekłych zatruć dwusiarczkiem węgła. Med. Pracy; 30: 135-145.
Sułkowski W,Kowalska S, Matyja W. Guzek W. Wesolowski W, Szymczak w, Kostrzewski P: (2002)
Effects of occupational exposure to a mixture of solvents on the inner ear: a filed study. Intern J of Occup.
Med. And Environ Health 15,3, 247-256
Symanski E, Bergamaschi E, Mutti A, ( 2001) Inter- and intr-individual sources of variation in levels of
urinary styrene metabolites. Int Arch Occup Environ Health 74, 336-344
Toppila E, Pyykko I Starck J, Kaksonen R, Ischizaki H,(2000) Individuals risk factors in the development
of Noise-induced hearing loss Noise and Health 8, 59-70
Toppila E, Pyykko I Starck J (2001) Age and noise –induced hearing loss. Scand Audiol 30, 236-244
Trybalska G, Namyslowski G, Morawski K (1999) Ocena przydatności pomiarów emisji otakustycznej
wywowłanej trzaskiem w wykrywaniu wczesnych ubytków sluchu spowodowanych hałasem Assessment of
transient evoked otoacoustic emissions in early detection of noise-induced hearing loss ( in Polish)
Otolaryngologia polska 53,2. 207-211
Umberger Ch.J., Pierese F.F. (1963). „Colorimetric method for hippuric acid” Clin. Chem.,
Viaene MK, (1998) Review of the literature : Meurotoxicity due to occupational exposure to organic
solvents Brussel 1998
Vrca A., Karacic V., Bozicevic D., Bozikov V., Malinar M.: (1996) Brainstem auditory evoked potentials in
individuals exposed to long-term low concentrations of toluene, American Journal of Industrial Medicine, ,
30, 62-66
Vrca A., Karacic V., Bozicevic D., Fuchs R., Malinar M (1997) Cognitive evoked potential VEP p300 in
persons occupationally exposed to low concentrations of toluene Arh hig rada toksikol 48, 277-285
Wenker MAM, Kezic S, Monster AC, de Wolff FA, (2001), Metabolic capacity and interindividual variation
in toxicokinetics of styrene in volunteers Human and Experimental Tooxicology 20, 221-228
Wiencke JK, Mustacchi P, (1998) Styrene and butadiene. in: Environmental and occupational medicine ed.
Wlliams N Ropm Lippincoll-Raven Publishers, Philadelphia , 1129-1144
131
June 2004
NoiseChem 132
Yokoyama K, Araki S, Murata K, Nishikitani M, Nakaaki K, Yokota J, Ito A, Sakata E, (1997) Postural
Sway frequency analysis in workers exposed to n- hexane, xylene and toluene: Assessment of subclinical
cerebellar dysfunction. Environmental Research 74, 110-115
132
June 2004
NoiseChem 133
Lab 3: Mariola Sliwinska-Kowalska
Department of Physical Hazards,
Nofer Institute of Occupational Medicine, Poland
133
June 2004
NoiseChem 134
Styrene versus Noise Exposure effects on Hearing and Balance
Mariola Sliwinska-Kowalska, Ewa Zamyslowska-Szmytke, Piotr Kotylo, Marek Bak, Marta Fiszer,
Malgorzata Pawlaczyk-Luszczynska, Adam Dudarewicz, Anna Gajda-Szadkowska, Wieslaw
Szymczak
Department of Physical Hazards, Nofer Institute of Occupational Medicine, Lodz, Poland
Abstract
It has been shown that occupational exposure to styrene is associated with increased probability of
developing hearing loss, dizziness and vertigo. However, the sites of the lesion of the auditory and
vestibular organ remain unknown. The function of hearing of 114 workers from glass fiber reinforced
industry exposed to styrene (mean concentration around 55 mg/m3) was assessed bilaterally by means
of pure-tone audiometry (PTA), impedance audiometry, otoacoustic emission (transient-evoked
otoacoustic emission – TEOAE and distortion-product otoacoustic emission – DPOAE), auditory
brainstem responses (ABR) and cognitive event related potentials (wave P-300). The balance function
was assessed by posturography (modified Romberg test and Limit of Stability test - LOS). The results
were referred to the group of 154 workers exposed to noise (above 85 dB-A) and 110 white-collar
workers exposed neither to noise nor solvents. Age and gender were accounted for as confounding
variables in all statistical models used in the study.
PTA showed significantly poorer hearing thresholds in styrene-exposed and noise-exposed groups as
compared to controls, while TEOAE level was slightly lower in the right ear in noise-exposed
population only. Although there was no impairment in TEOAE, DPOAE revealed significantly lower
levels of emission at high frequencies (4, 5 and 6 kHz) in styrene-exposed as well as in noise-exposed
workers, with poorer values in the latter group. The ABR results showed shorter latency of wave III
and V, and inter-wave I-V interval in styrene-exposed group that would indicate cochlear effect of
exposure. The ABR results along with no difference between groups in the stapedial decay test
exclude retrocochlear damage after styrene exposure. P-300 test results did not differ significantly
between styrene-exposed and control group, denying auditory cortex deficit. Posturography indicated
to the central disturbances in styrene-exposed group with the most significant changes in modified
Romberg test on the foam with eyes closed, and reaction time along with movement velocity in the
LOS test.
The study suggests that occupational exposure to styrene at moderate concentration is associated with
the damage of the cochlea at high frequency region, and central lesion of the vestibular organ.
DPOAE, ABR and posturography should be considered as the most important tests that are useful in
differential diagnostics of styrene-induced hearing and balance impairment.
Introduction
Styrene is an aromatic solvent widely used as a precursor for polystyrene plastics. Pure styrene is a
colourless, easily evaporating liquid with characteristic sweetish odour, sparingly soluble in water. In
the industry, it occurs mainly in the production of polyester laminates and plastics, plastic products,
synthetic rubber and insulating materials (such as polyurethane foam), in glass-fibre-reinforced plastic
products industry (yachts, lavatory pans, wash-basins). The highest occupational exposures to styrene
occur particularly when laminating large items like boats. Styrene is absorbed through airways and
skin. Its metabolites are mainly mandelic acid (MA) and phenylgloxal acid (PGA).
Styrene affects functions of various human organs and systems. The most significant effects were
observed in the mucosa, liver and nervous system. As other organic solvents, it is a neurotoxin
damaging central nervous system. It has been suggested that styrene could also influence hearing and
balance.
In humans the effect of styrene on hearing was first assessed by Muijser et al. (1988) in 59 employees
of a company producing glass fiber-reinforced plastic products. As compared to the non-exposed
control group and after adjustment for presbyacusis, these employees did not reveal increased hearing
thresholds when assessed by pure-tone audiometry, both at standard and extended high frequencies. A
statistically significant difference in hearing thresholds was found, however, between the workers
directly exposed to styrene (at the mean concentration of 138 mg/m3) and those indirectly exposed to
styrene (at 61 mg/m3). Also, Möller et al. (1990), who examined the employees exposed to styrene at
134
June 2004
NoiseChem 135
concentrations remarkably below the Swedish Occupational Exposure Limit - OEL (119 mg/m3 in
1990), did not note any significant hearing loss in the solvent-exposed group as compared to the nonexposed controls.
Although the first studies failed to indicate a significant effect of styrene on hearing threshold, our
recent paper document that occupational exposure to that solvent significantly increases the odds ratio
of hearing loss. Exposure to styrene at mean concentration of 61.8 mg/m3 was associated with 5.2-fold
increased odds ratio of hearing loss, and this value rose up to 13.1-fold for the combined exposure to
styrene and toluene (Sliwinska-Kowalska et al., 2003). Mean hearing thresholds adjusted for age,
gender and noise exposures were significantly higher in styrene-exposed group than in non-exposed to
styrene control at all standard frequencies tested (1-8 kHz).
Two other studies showed poorer hearing thresholds in a population of workers exposed to styrene at
very low concentration (12-16 mg/m3), that did not exceed the OEL values (Morata et al., 2002;
Morioka et al., 1999; Morioka et al., 2000). This effect was associated with styrene concentrations in
air and mandelic acid concentrations in urine (Morata et al., 2002).
The results of the studies performed in populations exposed to both noise and styrene are also
equivocal. An early investigation by Sass-Kortsak et al. (1995) did not provide evidence of chronic
styrene-induced hearing acuity impairment in 299 employees exposed to styrene and noise at fibrereinforced plastics manufacturing enterprises. In that study, noise (Leq 87.2±6.5 dBA) and styrene
(concentration 73.5±88.6 mg/m3) exposures were highly correlated, and the effect of noise was shown
to be predominant. On the other hand, our latest study assessing odds ratio of hearing loss in styreneonly, noise-only and styrene and noise exposed populations suggests an additive effect of co-exposure
(Sliwinska-Kowalska et al., 2003). The chance of developing hearing loss increased 3.3-fold in noiseonly exposed workers, 5.2-fold in styrene-only exposed workers, 10.9-fold in noise and styrene coexposed subjects and over 21.5-fold in noise, styrene and toluene co-exposed individuals.
Although an increased probability of developing hearing loss (defined as a hearing impairment > 25
dB HL at any frequency tested) due to styrene exposure was clearly demonstrated in industry workers,
the site of lesion in the auditory organ remains unclear. Potentially, it could damage auditory cortex as
well as the cochlea. Up to date human studies point rather to the central than peripherial effect. Möller
et al. (1990) in 18 workers with long-term exposure to styrene at low levels showed no hearing loss
when assessed by pure-tone audiometry, whereas 7/18 subjects exhibited pathology with distorted
speech and/or cortical response audiometry (CRA). Similarly, other authors described central auditory
pathology in workers exposed to different organic solvents and their mixture (cognitive event related
potential, distorted speech, CRA) (Laukli and Hansen, 1995; Steinhauer et al., 1997; Niklasson et al.,
1998). On the other hand, animal studies demonstrated a clear ototoxic effect of styrene to the cochlea.
The rats exposed to styrene at the concentration 1000 ppm for several days and weeks exhibited the
lesion of supporting cells and outer hair cells of the third row (OHC3). It was followed by the
disruption of OHC2 and OHC1; the changes were ascending from the basal to the apical turn of the
cochlea (Campo et al., 2001).
Vestibular or balance disorders seem to be quite often complaints in subjects exposed to styrene. They
are usually reported as unsteadiness, dizziness and vertigo.
Ödkvist et al. performed vestibulo-oculomotor tests in healthy volunteers before, during and after
acute exposure to styrene at the concentration between 87 and 139 ppm (370-591 mg/m3). No
spontaneous nystagmus appeared, and the rotatory and optokinetic nystagmus was not influenced by
styrene. However, the speed of the saccade was significantly enhanced, and visual suppression was
disturbed indicating that, in healthy subjects, exposure to styrene blocked the cerebellar inhibition of
the vestibulo-oculomotor system (Ödkvist, 1982). Similar changes related to ocular smooth pursuit
and visual suppression of the vestibulo-occular reflex were described in workers with long-term
exposure to styrene at low concentrations (Möller, 1990). In the patients with chronic toxic
encephalopathy (CTE) induced by long-term exposure to organic solvents, impaired equilibrium was
demonstrated by both static and dynamic posturography (Ledin et al., 1989 i 1991). Exposed workers
showed larger sway areas with eyes open, as well as closed, along with pathology in visual
suppression test (Ledin, 1989). Similar abnormalities were shown in subjects exposed to styrene
135
NoiseChem 136
June 2004
without CTE signs, suggesting that posturography could be an important assay in early diagnosis of
this syndrome. In the dynamic posturography, a phase of sensory organization was impaired in
patients with styrene-induced CTE, whereas movement coordination phase did not change as
compared to control subjects (Ledin, 1991).
•
•
•
Aims of the study
to evaluate the effects of styrene exposure on the auditory organ with extensive hearing
test battery
to compare hearing test results in styrene-exposed and noise-exposed subjects
to evaluate the effects of styrene exposure on balance
Subjects and Methods
Study population
The study population included 114 workers of a plastic factory exposed to styrene, 154 workers of a
dockyard and metal factory exposed to noise, and 110 white collar workers exposed neither to noise
nor organic solvents. These groups were selected from larger populations, described elsewhere
(Sliwinska-Kowalska et al., 2003). Exclusion criteria were as follows: time of employment in
exposure <6 months; abnormal otoscopy; abnormal tympanogram; history of ear diseases; ear surgery;
severe head trauma in anamnesis. Characteristics of study groups are given in table 1.
Table 1. Characteristics of Study Groups
Variable
Number of subjects
% of men
Age [years]
Current styrene exposure [mg/m3]1
% overexposed
Lifetime styrene exposure [mg/m3]
Current exposure index 2
- % overexposed
Lifetime exposure index for mixture
Mean lifetime noise exposure [dBA]3
Total lifetime noise exposure [dBA]
Styreneexposed
114
95
Mean (SD) 35.2 (8.9)
Range
20-60
Mean (SD) 54.9 (32.9)
Range
1.4 – 307.6
40.3
Mean (SD) 749.3 (757.6)
Range
3.3 – 3807.0
Mean (SD)
4.4 (2.3)
Range
0.1- 7.6
94
Mean (SD) 70.4 (193.7)
Range
0.2 - 2047
Mean (SD) 78.9 (3.5)
Range
71.3 – 85.9
Mean (SD) 123.8 (4.4)
Range
109.6 – 134.9
Noiseexposed
Unexposed
154
77
39.5 (9.4)
21-61
-
110
65
37.5 (9.1)
21-57
-
-
-
-
-
-
-
91.3 (4.0)
85.1 – 100.1
137.4 (4.6)
125.6–149.0
75.7 (2.2)
75 - 85
120.4 (5.4)
106.6-133.1
1
Admissible Occupational Exposure Limit for styrene in Poland is 50 mg/m3 (=12ppm)
This value accounts for all solvents present in the workplace, i.e. styrene and acetone (or
dichloromethane); the normative value is 1
2
3
Noise exposure level normalized to a nominal 8h working day averaged over the total time of employment;
Polish Occupational Exposure Limit Value is LEX,8h = 85 dBA, with 3 dB exchange rate
136
June 2004
NoiseChem 137
Study subgroups
To avoid the influence of between-group differences of age, gender and hearing impairment on
audiometric and postural test results, a separate analysis was performed for men-only with normal
hearing (hearing thresholds no worse than 25 dB HL at the range of frequencies 3-8 kHz). The
subgroups included 64 control subjects, 68 noise-exposed, and 70 styrene exposed subjects.
Questionnaire
The questionnaire included detailed inquiries on present and previous employment exposure to
solvents and noise, medical history, physical features, life-style, military service, and exposure to
ototoxic factors beyond occupational environment. Medical history was focused on the evaluation of
signs and symptoms of auditory and vestibular disorders, past middle ear diseases and surgery,
hereditary disorders, chronic systemic diseases, cholesterol level and hypertension, head trauma,
current and past medications with ototoxic potential. The occupational history emphasized the changes
in the duties performed, place of work in given conditions, application of individual protection on a
regular basis (type, availability and actual application) along with detailed analysis of the exposure to
noxious factors in former places of work. The history non-related with the occupations was focused on
the ototoxic factors, including the noise exposure in the past, during army service or at the leisure time
in particular. The life style data were aimed to identify smokers and evaluate alcohol intake. The
physical data included questions on age, body weigh, and skin pigmentation.
Solvent exposure
The evaluation of current occupational exposure to organic solvents was based on assessing the
concentrations of all toxic compounds in the work environment using individual dosimetry method.
Air was sampled by passing a known volume of air through charcoal tubes manufactured by SKC
Company. The absorbed substances were desorbed with carbon disulfite. The eluate was analyzed by
gas chromatography (Hewlett-Packard, model HP-5890II). The duration of the sampling period was
always close to the nominal working time, not shorter than 80% of an eight-hour working shift, in
accordance with the adopted uniform criteria for work environment monitoring. In order to investigate
the exposures in the past and former places of work, the protocols of environmental inspections
performed for the last 15 years by local occupational safety agencies were explored.
The current styrene concentrations ranged from 8.1 to 289.6 mg/m3 (Polish OEL -50 mg/m3). In
addition of styrene used in the direct production line, two other organic solvents, namely acetone
(detected concentrations from 1.4 to 307.6 mg/m3, OEL in Poland 600 mg/m3) and dichloromethane
(detected concentrations from 1 to 145.4 mg/m3, OEL in Poland 20 mg/m3) were used.
In addition to each solvent concentration, the exposure index for the mixture was calculated as a sum
of the fractions (measured air concentration of a given chemical by its OEL value) for all mixture
compounds.
Since the employees were exposed to different solvent concentrations at different workplaces and
during various employment periods, the total worklife exposure was calculated according to the
formula:
[(a1 x b1) + ...(an x bn)] / (b1 + ...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 average exposure over each employee’s worklife, the total worklife exposure value was
divided by total years of exposure. Averaged solvent exposure above OEL, or exposure index for a
mixture exceeding 1, indicated overexposure to solvents.
In the whole group of solvent exposed workers, 51% of subjects were overexposed to styrene and 94%
of subjects were overexposed to the mixture of solvents.
Noise exposure
The individual work-life exposures to noise were evaluated on the basis of collected subjects’ work
histories and exposure data in different working conditions (jobs). In order to investigate the previous
exposures to noise, the records of mandatory periodical measurements made by the employers over the
last 20 years were explored. Missing data were substituted by the best available information on the
137
NoiseChem 138
June 2004
possible exposure parameters. The current exposure to noise was evaluated based on our own
measurements, which were performed in accordance with Polish standard PN-N-01307 and
international standards ISO 1999:1990 and ISO 9612:1997, using the using the Brüel & Kjær sound
pressure level meter, type 2231.
Generally, according to ISO1999: 1990 the assessment of occupational exposure to noise was based
on noise exposure level normalized to a nominal 8 h working day averaged over the total time of
employment (mean life-time noise exposure), calculated using the following formula:

 1
L = 10 lg
 ∑T

EX
N
N
∑ T ×10
0,1 L EX,8h i
i
i =1
i
i =1





where:
N − is the total number of various time intervals/ workplaces / jobs,
LEX,8h i – is the equivalent continuous A-weighted sound pressure level normalized to a nominal 8 hour
working day in the time interval/ workplace/ job i, in dB,
Ti – is the duration of time interval i, in years.
The critical noise exposure limit between two distinctive groups of the exposed and non-exposed
subjects was based on admissible A-weighted sound pressure level of 85 dB-A (≤85 dB-A unexposed, >85 dB-A - exposed to noise).
The total lifetime noise exposure calculation included the factor of the time of employment and was
calculated according to the formula:
N
L E A , t = 10 × log(E A , t / E o ) = 10 × log(∑ t i × 10
i =1
0 ,1×L EX , 8 h i
N
/ To ) = 10 ⋅ log(∑ Ti × 240 × 28800 × 10
0 ,1×L EX 8 h i
i =1
where:
EA, t – is the total A-weighted sound exposure, in Pa2s,
Eo - is the reference sound exposure (=4⋅10 -10 Pa2s),
N − is the total number of various time intervals/ workplaces / jobs,
LEX, 8h i – is the equivalent continuous A-weighted sound pressure level normalized to a nominal 8 hour
working day in the time interval/ workplace/ job i, in dB,
Ti – is the duration of time interval i, in years,
ti – is the effective time exposure at the workplace, in seconds (ti=Ti⋅x 48 weeks per years x 5 days per
week x 8 hours per day= Ti x 240 days per year x 28800 seconds per day),
To – is the reference time (To=1s).
Audiometric tests
Hearing examination was performed at least 16 hours after the last exposure to noise in the sound
proof booth meeting the requirements of ISO 6189:1983.
The following tests were included:
- standard and extended high frequency pure-tone audiometry (air conduction 1- 16 kHz, bone
conduction 1-4 kHz), with the clinical audiometer AC-40 model from Interacoustics Co.
- immitance audiometry (tympanometry, stapedial reflex, decay test) with Zodiac 901 model from
Madsen
- otoacoustic emissions with ILO96 model from Otodynamics:
- transient-evoked otoacoustic emission – TEOAE (click 80µs, presented at repetition rate of 50 Hz
with a peak reception level of 80 dB ± 3 dB SPL,
138
/ To )
NoiseChem 139
June 2004
- distortion-product otoacoustic emission – DPOAE (DP grams at 0.7, 1, 1.5, 2, 3, 4 and 5 kHz with
70 dB SPL equal levels stimuli f1 and f2; DP input/output function at 4 kHz),
- auditory brainstem evoked responses (ABR) – a 100 µs, 90 dB HL click stimulus with Nicolet
Spirit 2000 & Spirit 2000 Lite model from Nicolet Biomedical;
- a cognitive event related potential (wave P-300) (stimulus oddball paradigm with target frequency
of 500 Hz and non-target more frequent stimuli of 1 kHz (mode A); and with target frequency of 1500
Hz and non-target more frequent stimuli of 1 kHz (mode B); target/non-target ratio 30/70 in a random
sequence at a comfortable listening level of 70 dB HL.
Posturography
Posturography was performed using model VSR/Basic Balance Master from Neurocom Inc.
The following tests were included:
- Modified Romberg test (Clinical Test of Sensory Integration on Balance) (eyes open, firm surface;
eyes closed, firm surface; eyes open, foam; eyes closed, foam)
- Limit of Stability-LOS (reaction time, directional control, movement velocity, endpoint and
maximal excursions)
Data analysis
Statistical analysis was based on the following tests
Test Chi 2 - to compare the differences in the prevalence of variables between study groups,
Tuckey-B test - to compare the differences of mean hearing test results between study groups
The significance level p<0.05.
The data were analysed in the entire group of subjects with age and gender included as
confounders, and in the male subject subgroups without hearing loss (hearing thresholds ≤ 25
dB).
Results
Entire group
Questionnaire
Questionnaire-based prevalence of variables is given in table 2.
Table 2. Questionnaire-based Prevalence of Variables
Variable
Styrene-exposed
(n=114)
Gender (male)
94.7
Solvent exposure in the past (yes)
16.7
Noise exposure in the past (yes)
36.8
Vibration exposure in the past (yes)
13.2
Exposure to noise during army service 61.4
(yes)
Exposure to solvent during army service 1.8
(yes)
Noise exposure in leisure time (yes)
16.7
Solvent exposure in leisure time (yes)
1.8
Smoking (yes)
65.8
Meningitis (yes)
1.8
Head trauma (yes)
11.4
a
Cholesterol level (increased)
1.8
Treatment with non steroid anti- 2.6
inflammatory drugs (yes)
Treatment with aminoglycosides (yes)
2.6
Kidney diseases (yes)
0.9
Liver diseases (without hepatitis viralis)
6.1
Diabetes (yes)
0
Noise-exposed
(n=154)
76.6
1.9
17.5
7.1
35.7
Unexposed
(n=110)
64.5
2.7
6.4
1.8
26.4
p
<0.05
<0.05
<0.05
<0.05
<0.05
1.9
0.9
>0.05
16.9
8.4
51.3
0.6
7.8
1.3
1.9
30.9
15.5
38.2
0.9
8.2
3.6
0
<0.05
<0.05
<0.05
>0.05
>0.05
<0.05
>0.05
0.6
1.3
2.6
0.6
1.8
0
2.7
0
>0.05
>0.05
>0.05
>0.05
139
NoiseChem 140
June 2004
Hypertensionb (yes)
6.1
Middle ear inflammation in anamnesis 16.7
(yes)
Acoustic trauma (yes)
2.6
Balance disorders (yes)
0.9
Vertigo (yes/no)
3.5
Tinnitus (yes/no)
7.0
Family diseases (yes)
3.5
a
b
8.4
11.0
4.5
22.7
<0.05
<0.05
2.6
3.2
6.5
13.6
4.5
2.7
4.5
8.2
6.4
0.9
>0.05
>0.05
>0.05
>0.05
>0.05
no data available from 70 subjects (15, 54, 1 in respective groups)
no data available from 8 subjects (1, 7, 0 in respective groups)
The styrene-exposed group, noise-exposed group and control group differed in respect of exposure to
noise and solvents in the past, during army service and in leisure time. There were more smokers in
noise-group and styrene-group as compared to controls. Control group had elevated cholesterol level,
while the subjects exposed to noise exhibited more often blood hypertension. The frequency of selfreported tinnitus or balance disturbances did not differ between groups.
Auditory tests
Both styrene-exposed and noise-exposed groups exhibited significantly greater mean hearing loss as
compared to control group at all frequencies tested (fig. 1). The frequency of cases with no measurable
hearing threshold at extended high frequencies differed between study groups and was the highest in
noise-exposed subjects (tab. 3).
1000
2000
3000
4000
6000
8000
1000
2000
3000
4000
6000
0
control
noise
5
hearing threshold [dB]
8000
styrene
10
15
20
25
30
35
*
*
*
*
*
*
*
*
*
*
*
*
frequency [Hz]
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 1. Mean hearing thresholds in pure-tone audiometry in the study groups
The ipsilateral stapedial reflex appeared at similar intensities of stimulus in styrene-group and noisegroup, and these values were higher than in control-group (fig. 2a). Contralateral stapedial reflex in the
right ear appeared at significantly higher levels of stimulus in styrene-exposed group as compared to
both noise-exposed and control group (fig. 2b). There was no difference in the proportion of subjects
exhibiting stapedial reflex recruitment and pathological decay test (tab. 3). The last finding indicates
no retrocochlear hearing damage in exposed groups.
140
NoiseChem 141
June 2004
Table 3. Prevalence of pathological results in standard and extended high frequency pure tone
audiometry (PTA), stapedial reflex recruitment and pathological decay test
Variable
Styrene-exposed
(n=114)
Noise-exposed Unexposed
(n=154)
(n=110)
p
53.5
51.9
23.6
<0.05
57.9
56.3
31.8
<0.05
7.9
9.7
2.7
>0.05
28.1
40.9
16.4
<0.05
7.9
15.6
3.6
<0.05
24.6
40.9
26.4
<0.05
2.9
5.7
4.3
1.4
2.9
1.4
11.4
0
0
2.9
0
4.8
7.9
7.1
1.6
1.6
1.6
1.7
3.3
3.3
4.7
0.24
0.05
0.63
0.44
0.83
0.16
0.33
Pathological decay test at 1 kHz, right ear
10.0
10.0
4.7
0.40
Pathological decay test at 500 Hz, left ear
10.0
7.9
5.0
0.55
Pathological decay test at 1 kHz, left ear
11.4
4.8
6.7
0.33
Standard and high frequency PTA
Hearing damage at any standard frequency
(HT>25 dBHL), right ear
Hearing damage at any standard frequency
(HT>25 dBHL), left ear
No measurable hearing threshold at 12 kHz,
right ear
No measurable hearing threshold at 16 kHz,
right ear
No measurable hearing threshold at 12 kHz,
left ear
No measurable hearing threshold at 16 kHz,
left ear
Stapedial reflexes
Recruitment* at 1 kHz, right ear
Recruitment at 2 kHz, right ear
Recruitment at 4 kHz, right ear
Recruitment at 1 kHz, left ear
Recruitment at 2 kHz, left ear
Recruitment at 4 kHz, left ear
Pathological decay test at 500 Hz, right ear
* Stapedial reflex <60 dB above hearing threshold
141
NoiseChem 142
June 2004
a) ipsilateral stapedial reflex
Frequency [Hz]
500
100 2000 4000
500
100 2000 4000
reflex threshold [dB HL]
75
80
85
control
90
95
noise
*
*
*
*
*
styrene
*
100
*
*
105
adjusted for age and gender
* significant difference between groups (p<0.05),
b) contralateral stapedial reflex
reflex threshold [dB HL]
94
96
98
100
102
104
106
108
110
*
*
50
0
10
0
20
00
40
00
w
hi
te
50
0
10
0
20
00
40
00
w
hi
te
no
is
no
is
e
e
frequency [Hz]
*
control
noise
styrene
*
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 2. Ipsi and contralateral stapedial reflex thresholds in the study groups
TEOAE whole response was weaker in noise-exposed subjects in right ear only (fig. 3a). The analysis
of TEOAE levels in frequency bands revealed weaker signals only at 4 kHz in both exposed groups
and both ears than in controls (fig. 3b). Larger spectrum of frequencies was involved when TEOAE
response was assessed as the signal to noise ratio (fig. 3c).
a) whole response
142
NoiseChem 143
June 2004
TEOAE level [dB SPL]
12
*
10
8
control
6
noise
styrene
4
2
0
right ear
left ear
10
00
20
00
30
00
40
00
50
00
10
00
20
00
30
00
40
00
50
00
b) frequency spectrum - signal
0
TEOAE level [dB SPL]
-5
-10
control
noise
-15
styrene
-20
*
*
-25
-30
c) frequency spectrum – S/N ratio
8
6
*
4
2
*
0
-2
control
noise
*
styrene
*
*
10
00
20
00
30
00
40
00
50
00
-4
*
*
10
00
20
00
30
00
40
00
50
00
TEOAE S/N [dB SPL]
10
frequency [Hz]
Figure 3. Transient-evoked otoacoustic emission (TEOAE) in the study groups (values adjusted for
age and gender)
a) DPgram; signal
143
NoiseChem 144
June 2004
control
noise
styrene
10
6
*
*
*
*
*
3000
2000
1500
6000
5000
4000
3000
2000
1500
*
500
-4
1000
-2
*
*
4000
*
1000
0
6000
2
5000
4
500
DP level [dB]
8
frequency [Hz]
adjusted for age and gender
* significant difference between groups (p<0.05),
20
18
16
14
12
10
8
6
4
2
0
*
*
*
control
*
*
noise
styrene
60
00
40
00
*
20
00
50
00
30
00
*
10
00
*
*
15
00
50
0
DP S/N [dB SPL]
b) DPgram; S/N ratio
frequency [Hz]
adjusted for age and gender
* significant difference between groups (p<0.05),
c) DPOAE - grow-rate function; signal
144
NoiseChem 145
June 2004
DP level [dB SPL]
10
f2 = 4000 Hz
5
0
*
-5
control
*
*
-10
noise
*
styrene
*
*
-15
*
*
-20
70 65 60 55 50 45
70 65 60 55 50 45
stimulus level [dB SPL]
adjusted for age and gender
* significant difference between groups (p<0.05),
DP S/N [dB SPL]
d) DPOAE - grow-rate function; S/N ratio
16
14
12
10
8
6
4
2
0
-2
-4
-6
f2 = 4000 Hz
*
*
*
control
noise
styrene
*
*
*
*
*
*
70 65 60 55 50 45
*
70 65 60 55 50 45
stimulus level [dB SPL]
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 4. Distortion product otoacoustic emission (DPOAE) in the study groups
Although the TEOAE changes in styrene-exposed and noise-exposed groups were rather subtle,
DPOAE showed a substantial decrease in otoacoustic emission signals in both exposed groups as
compared to controls. The changes were noted in both ears at high frequency region (fig. 4 a, b, c, d).
This finding suggests a cochlear site of the lesion in noise-exposed group, but also in styrene-exposed
group. A shortening of the latency of wave III, and inter-wave I-III interval in both ears and inter-wave
I-V interval in the right ear was observed exclusively in the group exposed to styrene. No similar
changes were seen in noise-exposed group (fig. 5). The differences in the ABR results between noiseexposed and styrene-exposed group might be the result of a different pathomechanism of the lesion in
the cochlea. While for noise - mechanical injury to hair cells and their stereocilia dominates, in styrene
ototoxicity cell membrane rupture could be most crucial.
145
NoiseChem 146
June 2004
6
control
noise
5
styrene
*
*
4
*
3
*
*
2
1
0
I
III
V
I-III III-V I-V
I
III
V
I-III III-V I-V
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 5. Auditory brainstem response (ABR) in the study groups
345
*
*
340
335
330
control
325
noise
320
styrene
315
310
305
300
A
B
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 6. Cognitive event -related response P-300 in the study groups (A, B – see description in the
text)
The P-300 latency was not prolonged in styrene exposed workers compared to control group, denying
a central hearing effect of the exposure (fig. 6). On the other hand, unexpectedly, it was prolonged in
noise-exposed group. This finding requires more research in the future.
Posturography
There were no differences between groups in modified Romberg test performed on a firm surface with
eyes opened, and slight difference in favour of control group with eyes closed. When the foam pad
was used and eyes were opened - sway velocities differed mainly between styrene-exposed and noiseexposed groups. When eyes were closed, the sway velocity in styrene-exposed group was the highest
and differed significantly from the velocities of noise-exposed and control groups (fig. 7).
146
NoiseChem 147
June 2004
In the Limits of Stability (LOS) test, majority of parameters were poorest in styrene-exposed group
and differed significantly from control group (fig. 8a and 8b). The most remarkable impairment was
seen for reaction time, movement velocity and endpoint excursion. All these parameters were poorer
compared to noise-group.
sway velocity [deg/s]
2.5
*
2
control
1.5
noise
1
styrene
*
0.5
*
0
FIRM_EO
FIRM_EC
FOAM_EO
FOAM_EC
adjusted for age and gender
* significant difference between groups (p<0.05),
Firm EO: firm base, eyes opened
Firm EC: firm base, eyes closed
Foam EO: unstable base (foam), eyes opened
Foam EO: unstable base (foam), eyes closed
Figure 7. Sway velocity in the Modified Clinical Test of Sensory Interaction on Balance (mCTSIB)
a) directional control
6
*
5
4
control
3
noise
styrene
2
*
1
0
reaction time [s]
movement velocity [deg/s]
adjusted for age and gender
* significant difference between groups (p<0.05),
147
NoiseChem 148
June 2004
b) reaction time and movement velocity
100
*
95
[%]
90
*
85
80
control
noise
*
styrene
75
70
endpoint
excursion
maximum
excursion
directional
control
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 8. Limits of Stability (LOS) test parameters in the study groups
Male subject subgroups with normal hearing
The results of PTA, TEOAE, and DPOAE were similar in male subject subgroups to that of the entire
groups. The percent of ears with present DP signals was similar in styrene-exposed and control subjects, and
decreased in noise-exposed group for single frequencies or stimulus levels, only (tab. 4 and 5). Alike in the
entire group, at high frequencies (4, 5 and 6 kHz) the signals were weaker both in noise-exposed and styreneexposed group as compared to controls. At some frequencies, the differences reached the level of statistical
significance (fig. 9 a, b, c, d).
Table 4. The percent of ears with distortion-product otoacoustic emission signals in DP-grams in male
subject subgroups with normal hearing
Frequency Control Noise
Styrene p
Right ear
500 Hz
84.4 (54) 89.4 (59)
82.9 (58) 0.517
1000 Hz
96.9 (62) 90.0 (63)
90.0 (63) 0.187
1500 Hz
98.4 (63) 97.1 (68)
95.7 (67) 0.636
2000 Hz
100.0
97.1 (68)
92.9 (65) 0.033
(64)
3000 Hz
98.4 (63) 95.7 (67)
97.1 (68) 0.636
4000 Hz
98.4 (63) 91.4 (64)
95.7 (67) 0.150
5000 Hz
98.4 (63) 94.3 (66)
94.3 (66) 0.342
6000 Hz
96.9 (62) 88.6 (62)
94.7 (66) 0.148
Left ear
500 Hz
83.3 (50) 80.6 (50)
77.1 (54) 0.673
1000 Hz
93.3 (56) 93.7 (59)
88.6 (62) 0.502
1500 Hz
96.7 (58) 96.8 (61)
95.7 (67) 0.934
2000 Hz
100.0
96.8 (61)
97.1 (68) 0.220
(60)
3000 Hz
100.0
98.4 (62)
98.6 (69) 0.471
(60)
4000 Hz
98.3 (59) 96.8 (61)
97.1 (68) 0.847
5000 Hz
100.0
95.2 (60)
97.1 (68) 0.127
(60)
6000 Hz
100.0
90.5 (57)
94.3 (66) 0.015
(60)
148
NoiseChem 149
June 2004
Table 5. The percent of ears with distortion-product otoacoustic emission signals assessed as DP-growth
rate function at 4 kHz in male subject subgroups with normal hearing
Stimulus level [dB SPL] Control Noise Styrene p
n=64
n=68 n=70
Right ear
70
96.9
88.2 94.9
0.134
65
96.9
82.4 95.7
0.004
60
90.6
83.8 91.4
0.326
55
82.8
60.3 78.6
0.007
50
57.8
54.4 60.0
0.800
45
39.1
39.7 31.4
0.529
Left ear
70
93.3
93.3 95.7
0.799
65
95.0
90.0 91.3
0.552
60
91.7
88.3 82.6
0.292
55
70.0
76.7 69.6
0.609
50
61.7
60.0 42.0
0.043
45
36.7
40.0 29.0
0.397
a) DPgram, signal
DP level [dB SPL]
control
12
noise
10
styrene
8
6
4
*
*
*
2
60
00
40
00
20
00
10
00
50
00
30
00
15
00
50
0
0
frequency [Hz]
adjusted for age and gender
* significant difference between groups (p<0.05),
b) DPgram, S/N ratio
25
DP S/N [dB]
20
15
*
10
control
noise
styrene
*
5
60
00
40
00
20
00
10
00
50
00
30
00
15
00
50
0
0
frequency [Hz]
adjusted for age and gender
* significant difference between groups (p<0.05),
149
NoiseChem 150
June 2004
c) DPOAE - grow-rate function; signal
f 2 = 4000 Hz
DP level [dB SPL]
10
5
*
0
control
noise
-5
*
styrene
-10
-15
70 65 60 55 50 45
70 65 60 55 50 45
Stim ulus level [dB SPL]
adjusted for age and gender
* significant difference between groups (p<0.05),
d) DPOAE - grow-rate function; S/N ratio
f 2 = 4000 Hz
DP S/N [dB SPL]
20
control
noise
15
10*
styrene
*
*
*
*
*
5
0
70
65
60
55
50
45
70
65
60
55
50
45
stim ulus level [dB SPL]
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 9. Distortion product otoacoustic emission (DPOAE) male subjects with normal hearing
In the ABR, the latency shortening effect was even more clear than for the entire group. Apart from the
latency of the wave III, the latency of the wave V was also decreased, as well as inter-wave latencies I-III
and I-V in both ears (fig. 10). The latency of the wave P-300 was significantly longer in noise-exposed
subjects as compared to both other groups (fig. 11), and this finding needs to be further examined.
150
NoiseChem 151
June 2004
control
noise
styrene
7
*
6
*
[ms]
5
4
*
*
*
*
3
*
2
*
1
I
III
V
I-III III- I-V
V
I
III
V
I-III III- I-V
V
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 10. Auditory brainstem response (ABR) in male subjects with normal hearing
340
335
*
*
[ms]
330
control
325
noise
320
styrene
315
310
305
A
B
adjusted for age and gender
* significant difference between groups (p<0.05),
Figure 11. Cognitive event-related response P-300 in male subjects with normal hearing
Summary and Conclusions
The results of the study indicate that:
1. Occupational exposure to styrene at moderate concentrations (mean value about 55 mg/m3 = app. 13
ppm) may cause a damage at the level of cochlea as shown by distortion-product otoacoustic emission
(DPOAE). Like for noise exposure, it is associated with high frequency region. This finding is consistent
with the results obtained in animals (rats).
2. Shortening of inter-wave I-V latency in the auditory brainstem responses (ABR) in normal hearing
subjects could indicate a subclinical disrupting effect of styrene at the level of the cochlea.
3. No retrocochlear or central auditory pathology was observed in styrene-exposed workers, while in noiseexposed workers the latency of the wave P-300 was prolonged.
4. Occupational exposure to styrene at moderate concentrations (mean value about 55 mg/m3 = app. 13
ppm) may cause subclinical balance disturbances, that could be revealed by static posturography. The
findings of this study point to the central disturbances in the vestibular system. The most crucial
tests/parameters for screening styrene toxicity are: modified Romberg test on the foam with eyes closed, and
reaction time along with movement velocity in the Limit of Stability Test.
151
June 2004
NoiseChem 152
5. Otoacoustic emission, particularly DPOAE, auditory brainstem responses and posturography should be
considered as the most important tests that are useful in differential diagnostics of styrene-induced hearing
and balance impairment.
152
June 2004
NoiseChem 153
The effects of isolated and combined exposure to organic solvent mixtures and noise on hearing
and balance
Mariola Sliwinska-Kowalska, Ewa Zamyslowska-Szmytke, Piotr Kotylo, Marek Bak, Marta Fiszer,
Malgorzata Pawlaczyk-Luszczynska, Adam Dudarewicze, Anna Gajda-Szadkowska, Wieslaw Szymczak
Department of Physical Hazards
Nofer Institute of Occupational Medicine, Lodz, Poland
Summary and Conclusions
The results of the study indicate that:
1. Occupational exposures to organic solvent mixtures (with xylene as main compound) are related to
hearing impairment at the degree comparable to occupational noise-induced hearing loss.
2. The site of the lesion is cochlea (at least at part). That is supported by the findings of increased
prevalence of recruitment and decrease of otoacoustic emissions (particularly distortion product
otoacoustic emissions – DPOAE) at high frequencies as well as prolongation of wave I latency in the
auditory brainstem response (ABR) audiometry.
3. The pathomechanism of cochlear damage of noise and organic solvent mixtures is different, what is
supported by the findings of increased otoacoustic emission signals at low ( around threshold)
intensity of stimulus in case of organic solvent exposures (suggesting increased activity of outer hair
cells).
4. Combined occupational exposure to noise and organic solvent mixtures cause increased incidence of
cochlear hearing loss as compared to isolated exposure to noise or organic solvents; however, the
effect of noise on the profoundness of hearing impairment is predominant.
5. Neither noise exposure nor organic solvent exposure do not increase the frequency of cases with
retrocochlear or central hearing impairment, as assessed by contralateral stapedial reflex, stapedial
decay test, ABR and P-300 wave. However, the latency of wave P-300 was prolonged in case of
impulse noise-exposed subjects.
6. There was only slight negative influence of organic solvent exposure in balance organ assessed with
posturography. It was comparable to the influence of noise exposure. No cumulative effect of
combined exposure to noise and solvents was observed.
References
1. Campo P, Lataye R, Loquet G, Bonnet P. Styrene-induced hearing loss: a membrane insult. Hear Res
2001; 154(1-2): 170-180.
2. Laukli E, Hansen PW. An audiometric test battery for the evaluation of occupational exposure to
industrial solvents. Acta Oto-Laryngol 1995; 115(2): 162-164.
3. Ledin T, Ödkvist LM, Möller C. Posturography findings in workers exposed to industrial solvents. Acta
Otolaryngol (Stockh) 1989; 107; 357-361
4. Ledin T, Jansson E, Möller C, Ödkvist LM. Chronic toxic encephalopathy investigated using dynamic
posturography. Am J Otolaryngol 1991; 12; 96-100
5. Morata TC, Johnson AC, Nylèn P et al. Audiometric findings in workers exposed to low levels of
styrene and noise. J Occup Environ Med 2002; 44(9): 806-814.
6. Morioka I, Miyai N, Yamamoto H, Miyashita K. Evaluation of combined effect of organic solvents and
noise by the upper limit of hearing. Industrial Health 1999; 54(5): 341-46.
7. Morioka I, Kuroda M, Miyashita K, Takeda S. Evaluation of organic solvent ototoxicity by the upper
limit of hearing. Arch Environ Health 2000; 38(2): 252-257.
8. Möller C, Ödkvist L, Larsby B et al. Otoneurological findings in workers exposed to styrene. Scand J
Work Environ Health 1990; 16: 189-194.
9. Mujiser H, Hoogendijk E, Hooisma J. The effects of occupational exposure to styrene on high-frequency
hearing thresholds. Toxicol 1988; 49: 331-340.
10. Niklasson M, Arlinger S, Ledin T et al. Audiological disturbances caused by long-term exposure to
industrial solvents. Relation to the diagnosis of toxic encephalopathy. Scand Audiol 1998; 27(3): 131-136.
11. Ödkvist LM, Larsby R, Tham R et al. Vestibulo-oculomotor disturbances in humans exposed to styrene.
Acta Otolaryngol 1982; 94; 487-93.
153
June 2004
NoiseChem 154
12. Sass-Kortsak AM, Corney PN, McD Robertson J. An investigation of the association between exposure
to styrene and hearing loss. Ann Epidemiol 1995; 5(1): 15-24.
13. Sliwinska-Kowalska M, Zamyslowska-Szmytke E, Szymczak W et al. Ototoxic effects of occupational
exposure to styrene and co-exposure to styrene and noise. J Occup Environ Med 2003; 45(1): 15-24.
14. Steinhauer SR, Morrow LA, Condray R, Dougherty GG Jr. Event-related potentials in workers with
ongoing occupational exposure. Biol Psychiatry 1997; 42: 854-858.
154
June 2004
NoiseChem 155
Lab 4: Jukka Starck
Finnish Institute of Occupational Health
Finland
155
NoiseChem 156
June 2004
The subjects
The measurement were taken in 15 small boat shipyards. Totally 311 workers were measured of which 256 were men.
The number of workers varied between 3 and 45. One of the companies (Nr 15) was not using styrene because they
were using ABS-plastic in their boats. Companies (3 and 4) had the rule that the workers are using all the time
respiratory protective devices (RPD). The RPDs in use was a power assisted filtering device.
50
40
Number of subjects
30
20
10
0
.5
2.5
1.5
4.5
3.5
6.5
5.5
8.5
7.5
10.5
9.5
12.5
11.5
14.5
13.5
15.5
Company
Figure 1. Number of workers in companies participating to study (Note the figure shows the mean value of each class)
The methods are described in Appendix II.
The mean age of the laminators was 37 years, non-laminators 40 years and office workers was 43 years.
44
42
Mean Age (Years)
40
38
36
Non-laminating
laminating
office
Figure 2. The age distribution of workers in different occupations
The laminators and non-laminators have relatively high blood pressures (means systolic over 140 mmHg and mean
diastolic over 84 mmHg) compared to the office group. The cholesterol levels are about normal. The body mass index
(BMI) indicates typically overweight (normal values 20-25).
156
NoiseChem 157
June 2004
Table 1. Basic health data on blood pressure, cholesterol and body mass index (BMI)
Parameter
Laminators
Non-laminators
Mean
Stdev
Mean
Stdev
Systolic blood
140
17
143
15
pressure (mmHg)
diastolic blood
84
13
87
11
pressure (mmHg)
total Cholesterol
5.0
1.1
5.4
1.6
(mmol/l)
BMI
26
3
26
3
Office
Mean
130
Stdev
12
80
11
5.1
0.9
26
2
Results
Styrene measurements
Biological monitoring of styrene
The magpa measurements showed a clear difference between the laminators and non-laminators. For laminators the
mean value was 0.9 (mmol/l) and non-laminators the mean was 0.3(mmol/l). A large variation exists (Fig 3)
32
116
4
184
27
225
162
U_MAGPA
2
193
222
59
179
195
122
64
76
20
33
152
210
0
-2
N=
140
104
12
Non-laminating
laminating
office
Figure 3. The distribution of magpa among laminators and non-laminators.
157
NoiseChem 158
June 2004
12.0
10.0
8.0
162
122
64
91
202
228
215
204
43
U_MAGPA(mmol/l)
6.0
12
10
U_MAGPA (mmol/l)
8
6
4.0
2.0
0.0
4
184
-2.0
2
158
N=
0
1
6
Missing
-2
N=
1
8
Missing
1
15
2
1
12
7
4
3
9
3
7
5
6
36
10
9
2
3
13
11
5
2
1
1
4
26
16
4
3
20
1
6
5
5
8
8
7
14
9
10
9
4
1
12
11
8
12
14
13
15
15
14
Company number
Company number
A) laminators
B) non-laminators
Figure 4. The magpa levels in different companies.
The results of the styrene in respiratory zone are given in table 2. In single samples large variations exist. Mainly this
is due to fact that most of the subjects are working in small companies, where the duties vary from day to day. Thus
heavy exposure lamination did no occur every day.
Hygienic monitoring of styrene
Table 2. Styrene concentration at the breathing zone
Laminators
Non-laminators
Parameter
Mean
Stdev
Mean
Stdev
Styrene (mg/m3)
104
107
11
12
Office
Mean
50
Stdev
1
Noise measurements
Noise measurement were taken from totally 104 workers. As the workplaces had not many noisy tools, no high noise
exposures were found and the difference between groups was only 5 dB.
Table 3. Noise levels in different occupations
Laminators (N=61)
Parameter
Mean
Stdev
Noise level
81
6
(dB(A))
Non-laminators (37)
Mean
Stdev
83
9
Office (N=2)
Mean
78
Stdev
5
Audiometry
From audiometric data all data where the form of the audiogram did not correspond to age-related or noise induced
hearing loss were removed. As criteria the following was used: The maximum hearing loss must occur at frequencies
3-6 kHz. At speech frequencies the audiogram must be flat to max (5 dB, 0.15times mean hearing loss at 0.5-2 kHz).
The hearing loss of laminators corresponded well with prediction of ISO 1999, but the non-laminators and the office
workers had exceptionally many with a large hearing loss.
158
NoiseChem 159
June 2004
-5
R0.5
R1
R2
R3
R4
-5
R6
0
50%
60%
70%
80%
90%
5
10
15
L1
L2
L3
L4
L6
10
15
20
25
25
A) Right ear of laminators
R0.5
R1
R2
50%
60%
70%
80%
90%
5
20
-5
L0.5
0
B) Left ear of laminators
R3
R4
-5
R6
0
L0.5
L1
L2
L3
L4
L6
0
50%1
60%
70%
80%
90%
5
10
15
50%
60%
70%
80%
90%
5
10
15
20
20
25
25
C) Right ear of non-laminators
D) Left ear of non-laminators
Figure 5 Comparison of hearing loss of laminators and non-laminators to ISO1999-1990.
Otoacoustic Emission
The transient otoacoustic emission was measure two times; the second time with contralateral inhibition of 75 dB white
noise. From the data all results with low repeatability (<70= were removed). After that there were no difference
between the groups.
100.0
90.0
100.0
90.0
80.0
70.0
80.0
70.0
60.0
50.0
40.0
Non-laminating
laminators
60.0
50.0
40.0
30.0
20.0
30.0
20.0
10.0
0.0
10.0
0.0
N1K
N2K
N3K
N4K
N5K
Non-laminating
laminators
C1K
C2K
C3K
C4K
C5K
A) TOAE
B) TOAE with contralateral inhibition
Figure 6. Transient Otoacustic emission results
Tinnitus
Tinnitus occurred more often among-non-laminators. They had more often tinnitus and the symptoms were
experienced more severe (figure 7). There is no significant deviation from normal population.
159
NoiseChem 160
June 2004
20
16
14
12
10
10
8
6
Non-laminating
Count
4
laminating
Count2
office
0
sometimes
Often (daily)
Non-laminating
laminating
0
Always
office
Not at all
Frequency of tinnitus
Little
Much
Very much
Disturbance of tinnitus
A) Frequency of tinnitus
Figure 7. Tinnitus data
B) Disturbance of tinnitus
Quality of life
There were no clear difference between the groups in different topics. The self-evaluated hearing was lower among
non-laminators, which is in accordance with audiogram results.
The self-reported quality life (QoL) showed a clear difference between laminators and non-laminators. For ages under
30 there were no difference. In both groups the QoL decreased exceptionally fast, however the decrease was still faster
among laminators.
Mean self-evaluated quailty of life
100
90
80
Non-laminating
laminating
70
office
Age<30
Age30-40
Age40-50
Age>5=50
Figure 8. The self-evaluated general quality of life (thermometer of EQoL-5D) in different age groups
Balance measurements
Balance measurements consisted of static and virtual tests. The static tests did not show differences between the groups
(table 4).
160
NoiseChem 161
June 2004
Table 4. Static balance parameters (See appendix III for names of parameters)
PP_1 PP_2 PP_3 PP_4 PP_5 PP_6 PP_7 PP_8 PP_9 PP_10 PP_11
Non-laminating Mean
13.0
2.0
11.4
1.3
1.3
1.9
6.7
-5.8
-8.8
9.7
12.6
N
140
140
140
140
140
140
140
140
140
140
140
Std. Dev
6.0
1.5
6.5
0.7
0.8
1.8
2.6
2.0
2.8
2.2
3.2
laminating
Mean
11.7
1.9
11.6
1.2
1.2
1.7
6.7
-6.1
-9.0
9.8
12.9
N
103
103
103
103
103
103
103
103
103
103
103
Std. Dev
5.3
1.4
6.2
0.6
0.7
1.9
2.8
1.6
2.3
2.1
2.6
office
Mean
10.8
1.6
10.0
1.2
1.3
1.5
6.7
-6.4
-9.8
9.8
13.1
N
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
Std. Dev
5.9
1.3
4.0
0.4
1.0
0.9
1.8
1.2
1.4
1.8
1.8
Total
Mean
12.4
1.9
11.4
1.3
1.3
1.8
6.7
-6.0
-8.9
9.7
12.8
N
255
255
255
255
255
255
255
255
255
255
255
Std. Dev
5.8
1.5
6.3
0.6
0.8
1.8
2.6
1.8
2.5
2.1
2.9
PP_12 PP_13 PP_14 PP_15 PP_16 PP_17 PP_18 PP_19 PP_20 PP_21 PP_22
Non-laminating Mean
18.7 192.3
5.0
14.2
24.3
25.2
12.9
5.7
15.6
21.2
18.9
N
140
140
140
140
135
140
139
140
140
138
140
Std. Dev
3.8
68.7
3.6
12.8
15.6
24.6
11.2
5.7
15.3
18.5
24.5
laminating
Mean
18.9 195.2
5.5
14.0
22.3
26.2
11.5
6.0
15.0
17.0
12.9
N
103
103
104
104
102
104
104
104
104
104
104
Std. Dev
3.9
67.3
4.4
11.8
14.1
25.7
9.4
6.0
16.3
14.6
22.1
office
Mean
19.6 204.2
5.3
13.7
23.9
29.4
12.7
5.0
11.4
11.3
8.8
N
12.0
12
12
12
12
12
12
12
12
12
12
Std. Dev
2.9
50.5
2.8
7.8
9.9
20.7
6.0
6.2
14.7
12.3
21.7
Total
Mean
18.8 194.0
5.2
14.1
23.5
25.8
12.3
5.8
15.1
19.0
16.0
N
255
255
256
256
249
256
255
256
256
254
256
Std. Dev
3.8
67.2
3.9
12.2
14.8
24.8
10.3
5.8
15.6
16.9
23.6
PP_23 PP_24 PP_25 PP_26 PP_27 PP_28 PP_29 PP_30 PP_31 PP_32
Non-laminating Mean
15.5
5.3
15.9
23.3
12.9
12.0
5.3
17.0
23.5
17.3
N
139
140
140
140
140
138
140
140
138
140
Std. Dev
21.8
4.7
13.2
18.2
19.3
11.6
5.0
15.4
19.1
23.2
laminating
Mean
12.7
4.4
13.6
21.3
11.2
11.4
4.4
13.5
20.1
17.2
N
103
104
104
103
104
103
104
104
104
104
Std. Dev
16.0
4.8
14.2
20.0
16.9
12.1
5.0
14.1
19.6
23.6
office
Mean
9.8
4.8
14.1
18.5
12.8
9.7
4.1
9.8
13.6
11.2
N
12
12
12
12
12
12
12
12
12
12
Std. Dev
11.5
5.5
13.9
17.5
17.6
11.7
6.2
13.6
16.1
17.6
Total
Mean
14.1
4.9
14.9
22.3
12.2
11.7
4.9
15.3
21.6
17.0
N
254
256
256
255
256
253
256
256
254
256
Std. Dev
19.3
4.8
13.7
18.9
18.3
11.8
5.1
14.9
19.2
23.1
7
168
6
6
2.2
5
2.0
232
192
4
137
3
114
88
174
126
220
131
2
17
142
1.8
240
127
1.6
173
136
36
28
1.4
laminating=1, non-la
PP_5
1
Non-laminating
laminating
0
office
-1
N=
34 30
2
Age<30
32
33
1
Age30-40
37
29
6
Age40-50
37
11
3
Age>5=50
1.2
1.0
95% CI PP_5
.8
.6
.4
N=
140
103
Non-laminating
laminating
12
office
AGEGRP
161
NoiseChem 162
June 2004
A) Sway velocity by age group and work task
B) Group means of sway velocity
2.4
20
2.2
6
168
2.0
1.8
10
232
126
114
173
131
253
233
213 17
108
220 94
137
125 136
1.6
240
164
127
1.4
laminating=1, non-la
0
PP_6
Non-laminating
-10
N=
34 30
2
Age<30
32 33
1
Age30-40
37 29
6
Age40-50
37 11
1.2
95% CI PP_6
1.0
laminating
.8
office
.6
3
Age>5=50
N=
140
103
12
Non-laminating
laminating
office
AGEGRP
C) Maximal sway velocity by age group and task
D) Group means of maximal sway velocity
Figure 9. Romberg quotient calculated for sway velocity (A and B) and for maximal velocity (C and D).
Differences between groups were found in virtual tests.
In tilting platform tests the laminators and the non-laminators differ from each others. (figure B) The difference is
smallest in the age group below <30 and increases with age (Figure xA).
162
NoiseChem 163
June 2004
500
300
400
280
300
260
200
240
220
100
36
95% CI PP_51
Non-laminating
12
200
116
PP_51 0
laminating
180
office
-100
N=
34 30
2
Age<30
32 33
1
37 29
6
37
11
160
3
N=
Age30-40 Age40-50 Age>5=50
140
103
12
Non-laminating
laminating
office
AGEGRP
A) Tilting platform number of large lateral
corrections (as to free sway, eyes open)
Figure 10. Example of tilting platform test results
B)Tilting platform group means
In cone tests no age dependency was found. Also the difference between groups was smaller.
1.0
187
.5
.1
0.0
101
-.5
219
228
141
150
233
-1.0
176
188
0.0
27
82
-.1
-1.5
Non-laminating
67
PP_47
-2.0
laminating
95% CI PP_47
-.2
office
-2.5
N=
33 30 2
30 32 1
Age<30
Age30-40
37 29
6
37 11
Age40-50
3
-.3
Age>5=50
N=
137
102
12
Non-laminating
laminating
office
AGEGRP
A) Cone test deviation during first rotation
Figure 11.Examples of the results of the cone tests
B) Group means of deviation in first rotatiom
1000
40
500
177
82
800
10
600
126
400
400
200
Non-laminating
PP_60
95%300
CI PP_60
0
laminating
office
-200
N=
34 30
2
Age<30
32 33
1
Age30-40
37 29
6
Age40-50
37 11
3
Age>5=50
200
N=
140
103
12
Non-laminating
laminating
office
AGEGRP
A) Tunnel test power consumption in various groups
Figure 12. Examples of the results of the tunnel tests
B) Group means of power consumption
163
NoiseChem 164
June 2004
In independent comparison group was formed from construction and maintenance workers of Tampere university
hospital (N=29, age range 25-37 years). Combined with the ABS-plastic worker and office workers a reference group
was formed. In virtual test there was no correlation between groups.
0.800
0.700
Significance
0.600
0.500
Tampere/nonexp
0.400
Tampere/office
0.300
non-exposed/office
0.200
0.100
0.000
46.000
-0.100
51.000
56.000
61.000
Test number
Figure 13. Comparison various virtual test for parameters (46-63, see appendix III )between different groups forming
the comparison groups.
For virtual test the laminators were compared to workers with a hearing loss exceeding 15 dB at speech region (0.5 -2
kHz). In all virtual parameters no significant differences were found.
1000
40
800
600
Power consumption
400
200
0
-200
N=
94
33
1.00
2.00
1=laminators 2= workers with hard of hearing
Figure 14. Example of the virtual parameter comparison between laminators and workers with a mean hearing loss
exceeding 15 dB (hard of hearing) at speech region
Vision
Color vision (C-index) by both tests did not differ in the three occupation groups (office n=7, non-laminator n=65,
laminator n=46) with different level of acute styrene exposure (Figure 15). The color vision, tested in high luminance,
was not affected by age.
164
NoiseChem 165
June 2004
Contrast vision (spatial frequency value in the D-column) did not differ in the three occupations (office n=11, nonlaminator n=78, laminator n=57) with different level of acute styrene exposure (Figure 16). The contrast vision, tested
in high luminance, correlated to age.
F15Rigth
3.000
F15Left
D15Rigth
2.000
D15Left
1.800
2.500
1.600
1.400
2.000
1.200
1.000
1.500
0.800
1.000
0.600
0.400
0.500
0.200
0.000
0.000
laminating
non-laminating
A) F15
Figure 15 Color vision
office
laminating
non-laminating
office
B) L15
contrast RigthD
8.000
Contrast LeftD
7.000
6.000
5.000
4.000
3.000
2.000
1.000
0.000
laminating
non-laminating
office
Figure 16. Contrast vision
165
NoiseChem 166
June 2004
Appendix I: The working procedure:
The moulded inner surface of the outer shell is covered with glass plastic fibre (Fig 1) by spraying layers of plastic
fibers and styrene.
A) General view
B) Detail view
Figure 1. Spraying glass fibre to the boat shell
After spraying the work finalized manually by adding fibres manually to those parts that are not perfectly filled.(Fig 2).
Figure 2. The Finishing of the work.
After the glass fibre has dried hard the inner shell of is mounted on and the finishing and furnishing can start (Fig. 3).
166
June 2004
NoiseChem 167
Figure 3. Examples of the finishing procedure.
167
June 2004
NoiseChem 168
Appendix II. The data collection procedure
The measurements were performed on site using the field clinical bus providing the laboratories needed.. Each
company was contacted and find out the number and ids of subjects. The questionnaire and the written description of
the project was sent to each subject at least one week before the field measurements. At the same time the subjects got
an appointment for examination. The measurements were performed in five phases (Fig 2):
- Checking the questionnaire, measurement of the blood pressure
- Vision tests and blood samples for the cholesterol
- Otoacoustic emission and audiometry
- Posturography
- Urine-sample reception
In addition selected subjects were equipped with passive styrene concentration batch and/or noise dosimeters.
Figure 1. The field clinic bus
168
June 2004
NoiseChem 169
A) Checking of questionnaire and
measurement of blood pressure
B) Blood sample for cholesterol
C) Vision contrast measurements
D) Color contrast measurements
E) Audiometric boot in the bus
F) Otoacoustic emission measurement
169
NoiseChem 170
June 2004
G) The station for noise and styrene dosimeter
distribution
Figure 2. The measurement stands.
F) Balance measurements
Measurements:
Questionnaire
The questionnaire contained 115 questions grouped as follows:
- Diseases affecting to the hearing (central nervous system diseases, ear diseases..) and accidents effecting Central
Nervous Functions.
- Exposure to noise and organic solvents (including free time and military service noise exposure)
- European Quality of life 5 Dimension (EQoL-5D)
- Tinnitus
- Individual susceptibility (use of pain killers, smoking habits, cholesterol, blood pressure)
The questionnaire was sent to the subjects in advance and it was screened by a trained nurse when it was delivered.
Styrene and noise exposures
Passive styrene diffusion monitor (3M.3500) and noise dosimeters (Larson-Davis 705/706) were given to the first five
subjects of every morning. The microphone of the noise dosimeter was mounted at the left shoulder of the subject. The
styrene dosimeter was mounted on the collar of the subject. At the end of the work shift the subjects returned the
dosimeters. The styrene dosimeters were sealed and put to refrigerator. The contents of the noise dosimeters were read
using a software provided by the manufacturer and the daily dose was calculated.
Blood pressure
Blood pressure was measured using an automatic clinical blood pressure meter (Omron M5) after 30 min rest from the
work.
Cholesterol measurement
For the cholesterol measurement a blood sample of 5 ml was taken. It was centrifuged on site in the clinical bus and
stored in the refrigerator before sending to the laboratory analysis.
Biological monitoring of styrene
The biological monitoring of styrene was done from the urine mandelic acid (magpa). The sample was taken from the
morning urine before eating lunch day after the other measurements. The samples were stored in refrigerator and sent
to laboratory for analysis.
Audiogram: The measurement was performed using a Madsen Midimate 602 (Denmark) audiometer, which was
calibrated at the Finnish Institute of Occupational Health. The measurement order deviated from normal. The
measurement was started with left ear at 4 kHz and then 8 kHz and after 2, 1 and 0.5 kHz respectively. After that the
right ear was measured using the same sequence.
Distortion product Otoacoustic measurement: The measurement was performed using Otodynamics ILO 88 (USA)
equipment from left ear. To measure the contralateral inhibition white noise was generated using a RIO MP3-player
(USA) and Sony RTF-headphone. The sound level was 80 dB(A) and it was calibrated using a artificial ear
170
NoiseChem 171
June 2004
(Bruel&Kjaer 4120). Under the influence of the white noise the transient otoacoustic emission was measured again
from the left ear.
Balance measurements were done using postural plate (figure 1). The platform is a circle with a radius of 250 mm. It
has three pressure censors at the circle forming a equilateral circle. The pressure censors are powered by strain gages.
The signals from the strain gages were connected to a datalogger (Hewlett-Packard 34790A, USA) capable of sampling
data at the speed of 250 samples per second. The platform is supported from the middle point by ball pivot. Two
motors can incline the plate independently ±10 degrees in two directions under the control a programmable logic. The
movement patterns were pre-programmed to the memory of the programmable logic and selected using binary control.
This arrangement enabled conventional and virtual measurements. In conventional measurements the platform did not
move. In virtual measurements the platform moved or the test person was looking (figure 2) at a moving scenery using
virtual glasses (V8, Virtual Research Systems Inc USA). These glasses were stereo glasses with a resolution of
640*480 and a viewing angle of 60 degrees. The virtual sceneries were developed using World Toolkit (Sense8, USA
). The system was equipped with a head direction tracker Intertrax 30 (Intersense, USA) and the angle of the head
affected to the viewpoint of the virtual scenery.
Figure 1. The posture platform
Figure 2. The measurement setup
The measurement protocol was as follows:
To prevent injuries due to fall, the test persons were asked to use protection against fall. The protector was sewed in a
overall (figure 2). The subject was connected by a rope to a hook mounted in the roof of the clinical bus. The guiding
position of the feet was marked on the surface of the platform. The test persons were using their own shoes during the
test. During conventional measurements the test persons had their arms crossed in front of them. They were instructed
to move their center point of force by bending the ankles and staying otherwise straight.
The conventional measurement were as follows:
Reach test: The test subject leaned forward, backward, to left and to right directions.
The following parameters were calculated
-AP-range (reach forward- reach backward)
- AP range corrected with shoe size
-ML-range (reach forward- reach backward)
Target test: A square with the size of 2.5 cm appeared in the screen at five positions ((0,(forwardbackward)/2),(0,forward-2.5), (0,backward+2.5), (left-2.5, (forward-backward)/2), (right-2.5, (forward-backward)/2)).
Each target appeared for 10 seconds. On the screen was also a pointer showing the position of the centre point of force.
The test subject was instructed to move the pointer the square as fast as possible and to keep it there.
The following parameters were calculated
anterio reach time (s)
,anterio reach path (cm)
171
June 2004
NoiseChem 172
,anterio reach velocity (mm/s)
test,anterio hold percent (in target) (%)
test,anterio velocity in target (mm/s)
,posteri reach time (s)
,posteri reach path (cm)
,posteri reach velocity (mm/s)
,posteri hold percent (in target) (%)
,posteri velocity in target (mm/s)
,left reach time (s)
,left reach path (cm)
,left reach velocity (mm/s)
,left hold percent (in target) (%)
,left velocity in target (mm/s)
,right reach time (s)
,right reach path (cm)
right reach velocity (mm/s)
,right hold percent (in target) (%)
,right velocity in target (mm/s)
Step test: The test person was asked to step on the platform a constant speed (Figure 3).
Figure 3. A typical recording of the centre point of force from in a the step test. The larger deviation is lateral
position and the smaller is the frontal position.
The following parameters were calculated
time spent on right leg per step, mean (s)
time spent on right leg per step, std (s)
time spent on left leg per step, mean (s)
time spent on left leg per step, std (s)
most common frequency in lateral trace, mean (Hz)
most common frequency in lateral trace, std(Hz)
most common frequency in anterio-posterior trace, mean (Hz)
most common frequency in anterio-posterior trace, mean (Hz)
excess path spent in air, right leg mean (cm)
excess path spent in air, right leg std (cm)
excess path spent in air, left leg mean (cm)
172
NoiseChem 173
June 2004
Romberg test: In Rombergs test the subject was asked to stay in relaxed stance on the platform and not to
move. First 20 s was recorded with eyes open and then with eyes closed.
The following parameters were calculated
Free sway,eyes open
Free sway,eyes open
Free sway,eye close
Free sway,eye close
Free sway,Romberg
Free sway,Romberg
1
2
3
4
5
6
sway velocity
maximal sway amplitude
sway velocity
maximal sway amplitude
ratio of sway velocity with eyes open/and closed
ratio of maximal sway amplitude with eyes open and closed
In addition two virtual tests were performed.
Tilting platform: The virtual helmet has no image and platform started to tilt slowly for 15 s. The following
parameters were calculated
- sway velocity (mm/s)
- number of large lateral corrections (as to free sway, eyes open)
- total energy in sway per second, lateral sway (J/s)
- energy (per second) spent on large ML corrections (J/s)
- number of large AP corrections (as to free sway, eyes open)
- total energy in sway per second, AP sway (J/s)
- energy (per second) spent on large AP corrections (J/s)
Cone test: A cylinder appears in the virtual helmet. After 5 s the tunnel starts to rotate with accelerating speed
to left. After 10 seconds the speed starts to decrease. The same sequence is repeated to the right. The platform is
not moving during this test.
Figure 4. Visual stimulus: Cylinder which rotates during the test.
The following parameters were calculated
- sway velocity, first rotation direction (mm/s)
- Dev, first rotation direction (cm)
- sway velocity, second rotation direction (mm/s)
- Dev, second rotation direction (cm)
Tunnel Test: A tunnel (figure 5) appears in the virtual helmet.
173
June 2004
NoiseChem 174
Figure 5. Virtual stimulus: The tunnel where the test subject moves. The movement of the platform is
syncronized with the view in the virtual helmet for half of test lengh (20 sec) after that the phase of visual input
changes 180 degrees.
The following parameters were calculated
-sway velocity (mm/s)
-number of large lateral corrections (as to free sway, eyes open)
- total energy in sway per second, lateral sway (J/s)
-total energy (per second) spent on large ML corrections (J/s)
- number of large AP corrections (as to free sway, eyes open)
- total energy in sway per second, AP sway (J/s)
- energy (per second) spent on large AP corrections (J/s)
Vision
In this test a questionnaire consisting of use of alcohol, eye symptoms and quality of life was given. In addition
the gray scale and color test ..... was performed on clearly illuminated surface.
Visual tests
Methods:
Color vision: Color vision was tested with two color arrangement tests (15caps), easier Fahrnswoth 15 test and
then by a test with very hue colors,the Lanthony D15. Fahrnsworth was done once and Lanthony D15 three
times ifneeded for correct performance. The best result of three trials with Lanthony D15 was chosen.
Contrast vision: Contrast sensitivity was tested with Vistech 40cm-charts with either own eye glasses or when
subjectively needed eye correction, eye glasses ranging from -3 to +3 were available.
Tests vere done separately for both eyes under 2000 Lux day light luminance.
174
June 2004
NoiseChem 175
Appendix III. Postural parameters.
____________________ ___
__________
Test situation
Nr. Results
____________________ ___
__________
Free sway,eyes open
1
sway velocity
Free sway,eyes open
2
maximal sway amplitude
Free sway,eye close
3
sway velocity
Free sway,eye close
4
maximal sway amplitude
Free sway,Romberg
5
sway velocity
Free sway,Romberg
6
maximal sway amplitude
Leaning test
7
forward reach
Leaning test
8
backward reach
Leaning test
9
left reach
Leaning test
10
right reach
Leaning test
11
anterior-posterior range
Leaning test
12
lateral range
Leaning test
13
body support area (elliptic)
Target test,anterio
14
reach time (s)
Target test,anterio
15
reach path (cm)
Target test,anterio
16
reach velocity (mm/s)
Target test,anterio
17
hold percent (in target) (%)
Target test,anterio
18
velocity in target (mm/s)
Target test,posteri
19
reach time (s)
Target test,posteri
20
reach path (cm)
Target test,posteri
21
reach velocity (mm/s)
Target test,posteri
22
hold percent (in target) (%)
Target test,posteri
23
velocity in target (mm/s)
Target test,left
24
reach time (s)
Target test,left
25
reach path (cm)
Target test,left
26
reach velocity (mm/s)
Target test,left
27
hold percent (in target) (%)
Target test,left
28
velocity in target (mm/s)
Target test,right 29
reach time (s)
Target test,right 30
reach path (cm)
Target test,right 31
reach velocity (mm/s)
Target test,right 32
hold percent (in target) (%)
Target test,right 33
velocity in target (mm/s)
Stepping test
34
time spent on right leg per step, mean (s)
Stepping test
35
time spent on right leg per step, std (s)
Stepping test
36
time spent on left leg per step, mean (s)
Stepping test
37
time spent on left leg per step, std (s)
Stepping test
38
most common frequency in lateral trace, mean (Hz)
Stepping test
39
most common frequency in lateral trace, std(Hz)
Stepping test
40
most common frequency in anterio-posterior trace,
mean (Hz)
Stepping test
41
most common frequency in anterio-posterior trace,
mean (Hz)
Stepping test
42
excess path spent in air, right leg mean (cm)
Stepping test
43
excess path spent in air, right leg std (cm)
Stepping test
44
excess path spent in air, left leg mean (cm)
Stepping test
45
excess path spent in air, left leg std (cm)
Rotating cylinder
46
sway velocity, first rotation direction (mm/s)
Rotating cylinder
47
deviation, first rotation direction (cm)
Rotating cylinder
48
sway velocity, second rotation direction (mm/s)
Rotating cylinder
49
deviation, second rotation direction (cm)
Tilting platform
50
sway velocity (mm/s)
Tilting platform
51
number of large lateral corrections
(as to free sway, eyes open)
Tilting platform
52
total energy in sway per second, lateral sway (J/s)
Tilting platform
53
energy (per second) spent on large ML
175
NoiseChem 176
June 2004
Tilting platform
Tilting platform
Tilting platform
Tunnel test
Tunnel test
57
58
Tunnel test
Tunnel test
59
60
Tunnel test
61
Tunnel test
Tunnel test
62
63
corrections (J/s)
54
number of large AP corrections
(as to free sway, eyes open)
55
total energy in sway per second, AP sway (J/s)
56
energy (per second) spent on large AP
corrections (J/s)
sway velocity (mm/s)
number of large lateral corrections
(as to free sway, eyes open)
total energy in sway per second, lateral sway (J/s)
energy (per second) spent on large ML
corrections (J/s)
number of large AP corrections
(as to free sway, eyes open)
total energy in sway per second, AP sway (J/s)
energy (per second) spent on large AP
corrections (J/s)
176
NoiseChem 177
June 2004
Lab 5: Wieslaw Sulkowski
Nofer Institute, Poland
177
June 2004
NoiseChem 178
Effects of Occupational Exposure to a Mixture of Solvents on the Inner Ear: A field study
Wiesław J. Sułkowski1, Sylwia Kowalska1, Wojciech Matyja1, Wojciech Guzek1,
Wiktor Wesołowski2, Wiesław Szymczak3 and Przemysław Kostrzewski2
1 ENT and Audiology Division
2 Department of Chemical and Dust Hazards
3 Department of Environmental Epidemiology
Nofer Institute of Occupational Medicine Łódź, Poland
Introduction
The assessment of health effects of occupational exposure to organic solvents has recently been an
important problem for occupational medicine specialists and numerous toxicological research centers
dealing mainly with neurology and otoneurology. The reason is that organic solvents, chemicals
widely spread in industry, are characterized by high volatility and lipid solubility, which enhance their
absorption in tissues and their binding to lipids. As the nervous tissue is mostly composed of lipids, it
is particularly sensitive to solvent toxicity.
Chronic exposure to vapors of solvents, frequently a mixture of different compounds, may lead to
long-term or even permanent functional pathology in the central and peripheral nervous systems,
especially in workers exposed to high concentrations of these chemicals [1,2]. Some studies [3,4]
report that solvents at low doses, below the recommended threshold limit values, may produce mild
but clinically detectable sensory impairments. However, the quantitative data on exposure have not
been available to allow the dose-response relationship to be characterized.
178
June 2004
NoiseChem 179
Nowadays, a lot of data provide evidence that the subtle sensorineural elements of the inner ear may
also be damaged by solvents.
A recent literature review on the ototoxic effect of exposure to solvents in both laboratory animals and
humans is given in Table 1 [5–13] and Table 2 [14–20]; it suggests two distinct patterns of cochlear
dysfunction. One pattern produced by toluene involves the impairment of outer hair cells that normally
encode middle frequency tones located in the middle turns [21]. The ototoxicity appears to stem from
a preferential perturbation in motility of these cells and, thereby of sensitivity to sound. Preferential
dysmorphia in these cells and impaired regulation of free intracellular calcium level occur rapidly and
at low concentrations of toluene predicted to occur in the brain of humans exposed to its permissible
levels. Because the outer hair cell alone shows rapid electromotility, a process sensitive to [Cai2+], it
may particularly be vulnerable to ototoxic agents that disrupt intracellular calcium regulation. The
second pattern produced by trichloroethylene, unlike toluene, impairs preferentially inner hair cellspiral ganglion cell function. It is yet to be determined whether this reflects excitotoxic injury at this
synapse [5]. There is still a lack of well documented investigations trying to explain the mechanism of
ototoxicity of the solvent, mixture and evaluate the function of both the parts of the inner ear (cochlear
and vestibular) at large in workers employed in real industrial conditions. The major objective of this
study was to examine the physiological effects of occupational exposure to a mixture of organic
solvents on the auditory and vestibular systems in the factory workers involved in the production of
paints and varnishes.
Materials and Methods
Of the 95 workers exposed to a mixture of organic solvents, a final sample of 61 subjects (only men,
aged
22–58 years; mean, 39.8 ± 11.2) was selected following a questionnaire survey and
otolaryngological examinations. Those with middle ear pathology, past ear surgery, head injuries,
ototoxic drug treatment, diabetes, hypertension, neurologic diseases, alcohol abuse and history of
noise exposure were excluded from the study. The duration of employment and exposure to the
mixture of solvents in the study group ranged from 2 to 34 years (mean, 15.8 ± 9.1). People working in
direct contact with solvent vapors, such as resin synthesis analyzers, dry component mixers, mill
operators, dispenser operators, colorists and packers of final products were only eligible for the study.
The control group consisted of 40 non-exposed healthy workers, aged 25–56 years (mean 39.2 ± 10.5),
employed in the administration and transport service of the same factory. In order to identify the work
environment pollutants, individual dosimeters were used for air sampling at all workposts; the workers
were provided with personal pumps which they carried during all daily routine operations, and the
duration of sampling was always close to the nominal working time, usually not shorter than 80% of
an eight-hour working shift. The ventilation systems were in use at the time of the study. Biological
monitoring of exposure to solvents involved the analysis of their blood levels and urinary excretion of
the relevant metabolites, performed by gas chromatography with a flame-ionizing detector; the
capillary blood samples were collected within 15–20 min after working shift;
urine fractions were taken from the last 4 h of the shift. Clinical examinations were carried out mostly
in
audiobus, a mobile audiological vehicle equipped with a sound-proof cabin and diagnostic apparatus
for a comprehensive evaluation of hearing [22]; only electronystagmographic (ENG) investigations
with use of the electronically steered rotatory armchair, optokinetic projector and Enthermo system for
caloric stimulation were performed in the out-patient clinic laboratory.
The following audiological tests preceded by anamnesis and otolaryngological checking were applied
in all subjects: air and bone pure tone audiometry (PTA), impedance audiometry with tympanometry
(T), acoustic reflex threshold (ART) measurement, and otoacoustic emissions (OAE), both transiently
evoked emissions (TEOAE) and distortion product otoacoustic emissions (DPOAE) if present,
spontaneous emissions were also recorded. The ENG investigations included a battery of tests,
including: saccadic and eye-tracking test, spontaneous and positional nystagmus, optokinetic test,
rotatory (2°/sec2–90°/sec) and bithermal (30 and 44°C) caloric test.
Both the clinical data and exposure measurements were assessed and interpreted according to the
routine obligatory rules and procedures published earlier [23, 24]. In the statistical analysis of the data,
the t-Student test, calculation of means and linear regression analysis were used.
179
June 2004
NoiseChem 180
Results
Different solvents were identified in the breathing zone of workers (Table 3). Only concentrations of
xylene (about 20% of solvent mixture) exceeded slightly the MAC values binding in Poland (100
mg/m3); the concentrations of remaining mixture constituents: toluene, ethyltoluene, styrene, npropylbenzene, ethylbenzene and trimethylbenzene isomers (the latter about 42% of solvent mixture)
were lower or within the MAC values. However, the calculated rate of combined total exposure (the
sum of quotients of individual compounds concentrations by respective exposure limit values)
amounted to 0.94–3.73 mg/m3 (the geometric means) and fluctuated within max-min values of 12.38–
0.30 (hygiene norm = 1), depending on individual workposts.
180
June 2004
NoiseChem 181
A more detailed characteristics of exposure and subjects
are shown in Table 4; the so called cumulative dose of
exposure, i.e. the product of exposure duration in years
and of the total exposure rate was coined and the exposed
subjects were categorized into three groups (I, II, III)
accordingly to its value.
Interestingly, the measured levels of noise associated with
the production process in workrooms were low (within
Leq60 –75 dBA), considerably below the permissible
hygiene standard (85 dBA).
The blood and urine concentrations of solvents and urinary
excretion of their metabolites are given in Table 5;
although they did not exceed the limit values, they confirmed
a direct contact of the workers under study with industrial solvents.
One may presume, bearing in mind the rigid criteria for
selecting study subjects, that hearing loss recognized in
181
June 2004
NoiseChem 182
42% of the exposed subjects (v 5% of controls) and other audiological findings (Table 6) are the
effects of exposure to industrial solvents.
The solvent mixture-induced hearing loss was a sensorineural high frequency (above 1 kHz) loss of
various degrees with significantly reduced amplitudes of otoacoustic emissions as exemplified in Fig.
1. The crude audiometric data were finally age-adjusted according to ISO-7029 [25] and then
subjected to an
analysis. Both the hearing thresholds defined in pure tone audiometry and amplitudes of otoacoustic
emissions closely corresponded with cumulative dose of exposure; the more increased dose, the more
lowered amplitudes and the highest thresholds were observed, as illustrated by the DPOAE data in
Table 7. Similar relations also applied to TEOAE amplitudes. The most significant relationships
between DPOAE amplitudes and exposure were found in the subjects’ breathing zone, in which
trimethylbenzene isomers (pseudocumene, mesitylene and hemimellitene) predominated as the main
constituents of the solvent mixture (Figs. 2, 3 and 4). The ENG tests yielded interesting findings; they
demonstrated the presence of vestibular disorders of mild or advanced degree inasmuch as 47.5% of
182
NoiseChem 183
June 2004
workers
employed
at
the
production
of
paints
and
varnishes
(v 5% of controls) as shown in Table 8. The comparative analysis of mean parameters of vestibularoculomotor induced reactions revealed the significantly decreased duration, amplitude and slow phase
angular velocity of nystagmus versus normal values in the control group of non-exposed sub- jects. An
example of abnormalities in the ENG tracings is presented in Fig. 5.
183
June 2004
NoiseChem 184
184
June 2004
NoiseChem 185
Discussion and Conclusion
An increasing interest in biological effects imposed by a mixture of organic solvents in general, and
their influence on hearing and balance in particular, results from a growing use of these chemicals in
185
June 2004
NoiseChem 186
various branches of industry. This entails the need for recognizing thoroughly early preclinical
symptoms of possible intoxication in order to establish the relevant MAC values, especially for
individual constituents of the mixture in the combined exposure to solvents.
Although organic solvents have been used in industrial production for over 150 years, serious concern
for their ototoxic effect on the exposed workers began to grow only about 15 years ago. A lot of
research data have been gathered to date and all of them provide evidence that solvents can induce
permanent hearing loss in both experimental animals (Table 1) and humans (Table 2). The problem
was neglected due to the fact that substantial noise is often present in most occupational settings where
solvent exposures occur, and thereby hearing impairments observed in these situations have been
attributed exclusively to noise exposure. On the other hand, some recent studies suggest that the
combined exposure to noise and ototoxic chemical compounds induces hearing loss more severe than
that evoked by one of these agents acting alone [26,27]. The explanation of this possible synergistic
effect is the objective of the “NoiseChem” research project just developed by the European
Commission [5]. The US National Institute for Occupational Safety and Health (NIOSH) is also
involved in the project. The Institute has initiated much earlier studies of the effects of noise and
solvents occurring alone, and in a combined exposure (Fig. 6) [28]. The study reported here was
focused on the assessment of ototoxicity of the mixture of organic solvents identified in the factory of
paints and varnishes, therefore only those subjects who were employed at workposts with the low
noise levels, not hazardous to hearing (60–75 dBA), were eligible for the study. An audiological
testing was extended to include ENG investigations to check the function of vestibular organ of the
inner ear, seldom examined in the solvent exposed workers because of the difficulties faced in the use
of the sophisticated ENG set in the field conditions. The results of our study showed a significant
prevalence
of
ENG
abnormalities
(47.5%
in
the
exposed
group
vs
5%
of controls), although complaints of vertigo were reported by only 26.1 % of study subjects. This may
be explained by a very well known high sensitivity of the method that enables us to detect slight
preclinical signs of vestibular dysfunction. A predominant number of complaints of vertigo in
employees of the paint and lacquer industry (40.5%) were found in the neurological screening carried
out by Indulski et al. [29].
The ENG findings similar to ours (Table 8) were reported by Pollastrini et al. [19] who examined a
group of 53 industrial painters exposed to benzene derivatives. They detected asymmetric post-
186
June 2004
NoiseChem 187
rotatory and/or caloric reactions, disordered saccades and spontaneous nystagmus in the study
subjects. The frequency and advancement of these and other pathologies, revealed in our material,
markedly increased with growing cumulative dose of the exposure defined as a product of exposure
duration in years and of the calculated total exposure rate for the mixture of solvents. The qualitative
and quantitative evaluation of the ENG findings allows to assume that they are possibly due to toxic
lesions in the peripheral (reflected inter alia by the canal paresis) and central vestibular (evidenced for
example by abnormalities in saccadic, eye-tracking and optokinetic tests) systems; they probably
depend on the predominating components of the mixture. Also, the auditory findings of this study
(Table 6) correspond closely with a cumulative dose of exposure to the mixture of solvents. They
showed mainly a high frequency (above 1 kHz) sensorineural hearing loss, identified in 42% of those
exposed versus 3% of non-exposed subjects.
Adjusted accordingly to the presbyacusis data, proposed in ISO-7029 [9], the hearing loss in the
exposed group seemed to be of the mild or moderate degree and the exposed-non-exposed hearing
thresholds differed significantly. This was accompanied by the lowered amplitudes of otoacoustic
emissions, both TEOAE and DPOAE, or their absence if hearing loss exceeded 40–50 dB. Similar
findings were disclosed in animals intoxicated with toluene by Johnson [10]. The peripheral or central
auditory disorders in people working in contact with various solvent vapors were reported among
others by Odkvist et al. (18), Morata et al. [17], Laukli and Hansen [16] and Śliwińska-Kowalska et al.
[30]. Thus, it appears again that the site of hearing lesion due to solvent exposure, like in the vestibular
deficit, may be associated with individual constituents of solvent mixtures. The results of the present
study provide convincing evidence that occupational exposure to a mixture of organic solvents is
ototoxic.
All confounding factors (past and current noise exposure, past and current middle and inner ear
pathology, presbyacusis and presbyastasis, head injuries, neurological and metabolic diseases,
medication, abuse of alcohol, tobacco and drugs, chemicals used in hobbies) were excluded following
a thorough analysis of medical and occupational history and ENT examinations), therefore the
observed symptoms of the inner ear damage in the subjects under study can be attributed to the longterm exposure to mixtures of solvents.
Since the mean concentrations of individual solvents fell rather within the MAC limits or slightly
exceeded them, the total rate of the combined exposure to the mixture seems to be responsible for the
findings. It could be explained by the additivity rule of health effects that is often used to cope with the
problem of combined exposure to solvents [2]. As proved by environmental and biological
monitoring, as
well as by the measurements of otoacoustic emissions, the exposure to trimethylbenzene isomers:
pseudocumene, mesitylene and hemimetillene, the main components of the mixture solvents,
contributed most significantly to the development of clinically detectable inner ear disorders in the
187
June 2004
NoiseChem 188
workers employed in the factory of paints and varnishes. The precise mechanism by which some
toxicants exert their detrimental effect on the internal ear has not as yet been fully elucidated; perhaps
it is due to their direct assault upon the sensory cells of the cochlear and vestibular neuroepithelia,
followed by the effect of solvents on metabolic processes and enzymatic systems, which probably
inhibit the protein synthesis.
References
1. Griffin JW. Hexacarbon neurotoxicity. Neurobehav Toxicol Teratol 1981; 3 (4): 437–44.
2. Ikeda M. Interactions and health aspects of exposure to mixtures of organic solvents. In: Health
Effects of Combined exposures to chemicalsin Work and Community Environments. Interim Document
11. Copenhagen: WHO Regional Office for Europe; 1983.
3. Husman K, Karli P. Clinical neurological findings among car painters exposed to a mixture of
organic solvents. Scand J Work Environ Health 1980; 6 (2): 33–9.
4. Maizlish NA, Fine LJ, Albers JW, Whitehead L, Langolf GD. A neurological evaluation of workers
exposed to mixtures of organic solvents.Br J Ind Med 1987; 44 (2): 14–25.
5. Bushnell PJ, Kelly KL, Crofton KM. Effects of toluene inhalation on detection of auditory signals
in rats. Neurotoxicol Teratol 1994; 16: 149–60.
6. Campo P, Latanye R, Cossec B, Placidi V. Toluene-induced hearing loss – a mid-frequency
location of the cochlear lesions. Neurotoxicol Teratol 1997; 19: 129–40.
7. Crofton KM, Lassiter TL, Rebert CS. Solvent-induced ototoxicity inrats: an atypical selective midfrequency hearing deficit. Hear Res 1994; 80: 25–30.
8. Fechter LD, Lin Y, Herr DW, Crofton KM. Trichloroethylene ototoxicity– evidence for a cochlear
origin. Toxicol Sci 1998; 42: 28–35.
9. Gagnaire F, Becker MN, de Ceaurriz J. Alterations of brainstem auditoryevoked potentials in
diethylbenzene and diacetylbenzene-treated rats. J Appl Toxicol 1992; 12: 343–50.
10. Johnson AC. The ototoxic effect of toluene and the influence of noise,acetyl salicylic acid on
genotype. A study in rats and mice. Scand Audiol Suppl. 1993; 39: 1–40.
11. Niklasson M, Tham R, Larsby B, Eriksson B. Effects of toluene,styrene, trichloroethylene and
trichloroethane on the vestibulo- and optooculomotor system in rats. Neurotoxicol Teratol 1993; 15:
327–34.
12. Rebert CS, Boyes WK, Pryor GT, Svensgaard DJ, Kassay KM, Gordon GR, et al. Combined
effects of solvents on the rat’s auditorysystem styrene and trichloroethylene. Int J Psychophysiol 1993;
14: 29–59.
13. Rebert CS, Schwartz RW, Svensgaard DJ, Pryor GT, Boyes WK. Combined effects of paired
solvents on the rat’s auditory system. Toxicology 1995; 105: 345–54.
14. Abbate C, Giorganni C, Munao F, Brecciarolli R. Neurotoxicity induced by exposure to toluene.
An electrophysiologic study. Int Arch Occup Environ Health 1993; 64: 392–8.
15. Hirata M, Ogawa Y, Okayama A, Goto S. A cross-sectional study on the brainstem auditory
evoked potentials among workers exposed tocarbon disulfide. Int Arch Occup Environ Health 1992;
64: 321–4.
16. Laukli E, Hansen PW. An audiometic test battery for the evaluation of occupational exposure to
industrial solvents. Acta Otolaryngol (Stockh) 1995; 115: 162–4.
17. Morata T, Fiorini AC, Fischer FM, Colacioppo S. Toluene-induced hearing loss among
rotogravure printing workers. Scand J Work Environ Health 1997; 23: 289–98.
18. Odkvist LM, Moller C, Thuomas KA. Otoneurologic disturbances caused by solvents pollution.
Otolaryngol Head Neck Surg 1992; 106: 687–92.
19. Pollastrini L, Abramo A, Cristalli G. Early signs of occupational ototoxicity caused by inhalation
of benzene derivative industrial solvents. Acta Otorhinolaryngol Ital 1994; 14: 503–12.
20. Sułkowski WJ. Occupational exposure to carbon disulfide (CS2) and dysfunction of vestibular
system: a clinical study. Cah Not Document 1990; 139: 472–5 [in French].
21. Prasher D, Morata T, Campo P, Fechter L, Johnson AC, Lund S P, et al. NoiseChem: An European
Commission research project on the effects of exposure to noise and industrial chemicals on hearing
and
balance. Int J Occup Med Environ Health 2002; 15 (1): 5–11.
188
June 2004
NoiseChem 189
22. Sułkowski WJ, Sward-Matyja W, Matyja W. AUDIOBUS – the first Polish audiological mobile
unit. Int J Occup Med Environ Health 2001; 14 (1): 79–80.
23. Sułkowski WJ. Principles of Prevention of Noise-Induced Hearing Loss. Łódź: Nofer Institute of
Occupational Medicine; 2001 [in Polish].
24. Wesołowski W, Gromiec JP. Occupational exposure in Polish paint and lacquer industry. Int J
Occup Med Environ Health 1997; 10 (1): 79–88.
25. ISO-7029: Acoustic. Threshold of hearing by air conduction as a function of age and sex for
otologically normal persons. Geneva: International Organization for Standardization; 1984.
26. Morata T, Dunn DE, Kretschner LW, Lemasters GK, Keith RW. Effects of occupational exposure
to organic solvents and noise on hearing. Scand J Work Environ Health 1993; 19 (4): 245–54.
27. Morioka I, Miyai N, Yomamoto H, Miyashito K. Evaluation of combined effect of organic solvents
and noise by the upper limit of hearing. Ind Health 2000; 38 (2): 252–7.
28. NIOSH. The National Occupational Research Agenda (NORA).Cincinnati Ohio, USA: National
Institute for Occupational Safety and Health; 1996.
29. Indulski J, Sińczuk-Walczak H, Szymczak W, Wesołowski W. Neurological and
neuropsychological cxaminations of workers occupationally exposed to organic solvent mixtures used
in the paint and varnish production. Int J Occup Med Environ Health 1996; 9 (3): 235–44.
30. Śliwińska-Kowalska M, Zamysłowska-Szmytke E, Kotyło P, Wesołowski W, Dudarewicz A,
Fiszer M, et. al. Assessment of hearing impairment in workers exposed to mixtures of organic solvents
in the paint and lacquer industry. Med Pr 2000; 51 (1): 1–10 [in Polish].
189
NoiseChem 190
June 2004
Preliminary Data Analysis on Noise, Solvent Mix and Carbon disulphide (CS2)
Table 1. Exposed study groups according to age and gender
Study group
Men
Age
Women
n
%
n
%
Solvents only
39.3 ± 7.2
71
47.0
80
53.0
Solvents + noise
38.5 ± 8.2
48
100.0
0
0.0
Noise only
37.4 ± 9.7
45
78.9
12
21.1
Probability
0.272
Comments
< 0.0005
frequencies of men and women are significantly different in
no two
study groups
groups are
significantly
different at
the 0.05 level
Table 2. Pure tone hearing thresholds adjusted to age and gender (right ear)
Study group
1000 Hz
2000 Hz
4000 Hz
6000 Hz
8000 Hz
Solvents only
13.7 ± 7.6
14.4 ± 7.1
22.6 ± 12.4
28.5 ± 15.4
23.4 ± 15.3
Solvents + noise
14.1 ± 8.2
15.5 ± 11.5
19.3 ± 13.1
25.7 ± 16.3
23.0 ± 18.9
Noise only
13.0 ± 7.0
12.5 ± 8.6
24.6 ± 18.8
29.5 ± 18.3
23.5 ± 19.3
0.722
0.131
0.111
0.437
0.988
Probability in ANOVA
Comments
no two
no two
no two
no two
no two
groups are
groups are
groups are
groups are
groups are
significantly significantly significantly significantly significantly
different at
different at
different at
different at
different at
the 0.05 level the 0.05 level the 0.05 level the 0.05 level the 0.05 level
190
NoiseChem 191
June 2004
Table 3. Pure tone hearing thresholds adjusted to age and gender (left ear)
Study group
1000 Hz
2000 Hz
4000 Hz
6000 Hz
8000 Hz
Solvents only
13.4 ± 8.7
15.6 ± 9.6
23.0 ± 13.4
30.6 ± 15.6
25.6 ± 17.1
Solvents + noise
13.3 ± 10.3
15.3 ± 13.4
23.2 ± 14.5
27.0 ± 14.7
23.8 ± 17.2
Noise only
11.9 ± 8.6
13.5 ± 9.9
25.0 ± 18.8
27.7 ± 15.5
21.9 ± 18.0
0.546
0.443
0.642
0.276
0.365
Probability in ANOVA
Comments
no two
no two
no two
no two
no two
groups are
groups are
groups are
groups are
groups are
significantly significantly significantly significantly significantly
different at
different at
different at
different at
different at
the 0.05 level the 0.05 level the 0.05 level the 0.05 level the 0.05 level
Table 4. Prevalence of hearing losses (pure tone hearing threshold greater than 15 dB at least in one of
examined frequencies) in exposed study groups
Right ear
Study group
Left ear
Right and/or left ear
n
%
n
%
n
%
Solvents only
126
84.6
126
84.6
138
92.6
Solvents + noise
42
87.5
42
87.5
45
93.8
Noise only
43
75.4
47
82.5
53
93.0
Comments
hearing losses in study
hearing losses in study
hearing losses in study
groups are not
groups are not
groups are not
significantly different (p significantly different (p significantly different (p
= 0.964)
= 0.771)
= 0.213)
Table 5. Odds Ratios (OR) adjusted to age and gender in subjects with pure tone hearing threshold
greater than 15 dB at least in one of examined frequencies
Right ear
Study group
OR
probability
Left ear
OR
probability
OR
probability
Solvents only
1,00
Solvents + noise
1,53
0,429
1,19
0,756
1,24
0,773
Noise only
0,73
0,453
1,12
0,815
1,34
0,655
Comments
1,00
Right and/or left ear
1,00
no ORs are significantly no ORs are significantly no ORs are significantly
different from 1.00
different from 1.00
different from 1.00
Table 6. DP amplitudes adjusted to age and gender (right ear) in exposed study groups
Study group
2000 Hz
3000 Hz
4000 Hz
191
NoiseChem 192
June 2004
Solvents only
3.8 ± 6.8
- 1.5 ± 7.6
- 4.6 ± 7.8
Solvents + noise
3.3 ± 6.7
- 0.9 ± 7.2
- 4.1 ± 6.7
Noise only
4.3 ± 5.6
- 0.8 ± 7.0
- 5.0 ± 7.9
0.723
0.838
0.879
Probability in ANOVA
Comments
no two groups are
no two groups are
no two groups are
significantly different significantly different significantly different
at the 0.05 level
at the 0.05 level
at the 0.05 level
Table 7. SNR amplitudes adjusted to age and gender (right ear) in exposed study group
Study group
2000 Hz
3000 Hz
4000 Hz
Solvents only
15.3 ± 7.2
14.5 ± 7.6
12.8 ± 8.1
Solvents + noise
14.2 ± 7.0
14.5 ± 7.1
12.8 ± 6.7
Noise only
13.5± 5.8
14.4 ± 7.5
11.7 ± 8.5
0.293
0.990
0.714
Probability in ANOVA
Comments
no two groups are
no two groups are
no two groups are
significantly different significantly different significantly different
at the 0.05 level
at the 0.05 level
at the 0.05 level
192
NoiseChem 193
June 2004
Table 8. DP amplitudes adjusted to age and gender (left ear) in exposed study groups
Study group
2000 Hz
3000 Hz
4000 Hz
Solvents only
3.0 ± 7.1
- 2.1 ± 7.2
- 4.5 ± 6.6
Solvents + noise
3.1 ± 6.6
- 3.4± 6.7
- 7.6 ± 6.1
Noise only
3.3 ± 6.8
- 1.7 ± 6.6
- 4.8 ± 7.1
0.968
0.512
0.064
Probability in ANOVA
Comments
no two groups are
no two groups are
groups: solvents only
significantly different significantly different and solvents + noise
at the 0.05 level
at the 0.05 level
are different (p =
0.023)
Table 9. SNR amplitudes adjusted to age and gender (left ear) in exposed study group
Study group
2000 Hz
3000 Hz
4000 Hz
Solvents only
14.6 ± 7.4
14.3 ± 7.4
13.4 ± 8.5
Solvents + noise
15.4 ± 6.1
13.6 ± 11.1
14.5 ± 19.9
Noise only
13.7 ± 7.2
13.8 ± 7.1
12.5 ± 7.6
0.505
0.883
0.750
Probability in ANOVA
Comments
no two groups are
no two groups are
no two groups are
significantly different significantly different significantly different
at the 0.05 level
at the 0.05 level
at the 0.05 level
Table 10. DPOAE. Odds Ratios (OR) of REFER adjusted to age and gender (right ear)
2000 Hz
Study group
OR
probability
3000 Hz
OR
probability
OR
probability
Solvents only
1.00
Solvents + noise
1.09
0.884
0.86
0.742
0.73
0.411
Noise only
1.97
0.132
2.10
0.048
2.83
0.003
Comments
1.00
4000 Hz
1.00
risk of REFER in the
risk of REFER in the
noise only group is
noise only group is
significantly greater than significantly greater than
in the solvents groups
in the solvents groups
Table 11. DPOAE. Odds Ratios (OR) of REFER adjusted to age and gender (left ear)
no ORs are significantly
different from 1.00
2000 Hz
Study group
OR
probability
3000 Hz
OR
probability
1.00
4000 Hz
OR
probability
Solvents only
1.00
1.00
Solvents + noise
1.74
0,288
2.16
0.095
1.05
0.909
Noise only
3.17
0,006
6.79
< 0.0005
3.52
0.0004
193
NoiseChem 194
June 2004
Comments
risk of REFER in the
risk of REFER in the
risk of REFER in the
noise only group is
noise only group is
noise only group is
significantly greater than significantly greater than significantly greater than
in the solvents only
in the solvents only
in the solvents groups
group
group
Dynamic Posturography
Table 12. Postać bokiem, open eyes. Adjusted to age and gender
Study group
A max
(platform)
A ave
(platform)
S ave
(platform)
S ave
(head)
Solvents only
1.39 ± 1.32
0.23 ± 0.27
0.51 ± 0.30
2.31 ± 4.27
Solvents + noise
1.28 ± 1.03
0.21 ± 0.21
0.51 ± 0.25
2.08 ± 3.07
Noise only
1.97 ± 1.16
0.32 ± 0.21
0.65 ± 0.28
1.53 ± 1.94
0.005
0.022
0.006
0.421
value in the noise
only group is
significantly
greater than in the
others (p = 0.005)
value in the noise
only group is
significantly
greater than in the
others (p = 0.016)
Probability in
ANOVA
Comments
value in the noise no two groups are
significantly
only group is
different at the
significantly
0.05 level
greater than in the
others (p = 0.005)
194
NoiseChem 195
June 2004
Table 13. Postać bokiem, close eyes. Adjusted to age and gender
Study group
A max
(platform)
A ave
(platform)
S ave
(platform)
S ave
(head)
Solvents only
5.82 ± 3.65
1.02 ± 1.29
2.05 ± 1.25
3.51 ± 4.48
Solvents + noise
6.49 ± 3.17
1.15 ± 0.65
2.25 ± 1.15
3.43 ± 3.05
Noise only
7.61 ± 3.37
1.36 ± 0.67
2.76 ± 1.25
2.66 ± 2.12
0.008
0.143
0.002
0.361
Probability in
ANOVA
Comments
value in the noise no two groups are
significantly
only group is
different at the
significantly
0.05 level
greater than in the
solvents only
group (p < 0.05)
value in the noise no two groups are
only group is
significantly
significantly
different at the
greater than in the
0.05 level
others (p = 0.039)
Table 14. Postać przodem, open eyes. Adjusted to age and gender
Study group
A max
(platform)
A ave
(platform)
S ave
(platform)
S ave
(head)
Solvents only
1.92 ± 1.49
0.33 ± 0.32
0.67 ± 0.40
2.32 ± 4.31
Solvents + noise
1.90 ± 2.07
0.26 ± 0.26
0.61 ± 0.41
2.22 ± 3.07
Noise only
2.53 ± 1.58
0.40 ± 0.25
0.78 ± 0.44
1.42 ± 1.76
0.044
0.044
0.077
0.303
value in the noise
only group is
significantly
greater than in the
others (p = 0.050)
value in the noise
only group is
significantly
greater than in the
solvents + noise
group (p = 0.013)
Probability in
ANOVA
Comments
value in the noise no two groups are
significantly
only group is
different at the
significantly
0.05 level
greater than in the
solvents + noise
group (p = 0.033)
195
NoiseChem 196
June 2004
Table 15. Postać przodem, close eyes. Adjusted to age and gender
Study group
A max
(platform)
A ave
(platform)
S ave
(platform)
S ave
(head)
Solvents only
7.29 ± 4.30
1.27 ± 0.89
2.29 ± 1.26
3.11 ± 4.31
Solvents + noise
7.87 ± 5.05
1.42 ± 1.04
2.43 ± 1.10
2.97 ± 2.96
Noise only
7.84 ± 3.59
1.40 ± 0.69
2.57 ± 1.21
2.05 ± 1.72
0.641
0.538
0.358
0.188
Probability in
ANOVA
Comments
no two groups are no two groups are no two groups are no two groups are
significantly
significantly
significantly
significantly
different at the
different at the
different at the
different at the
0.05 level
0.05 level
0.05 level
0.05 level
Table 16. Age and frequency of men and women in study groups
Study group
Men
Age
Women
n
%
n
%
CS2 only
54.5 ± 7.3
38
88,4
5
11,6
CS2 + noise
58.3 ± 6.6
23
85,2
4
14,8
Noise only
37.4 ± 9.7
45
78.9
12
21.1
Probability
< 0.0005
0.434
Comments
noise only
group is
significantly
younger than
the others
groups
frequencies of men and women are similar in study groups
196
NoiseChem 197
June 2004
Audiometry
Table 17. Audiogram – right ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
31.0 ± 16.2
35.8 ± 16.7
48.6 ± 20.0
56.4 ± 22.9
53.2 ± 25.2
CS2 + noise
36.8 ± 17.1
41.8 ± 20.5
50.6 ± 21.3
59,6 ± 20.1
58.0 ± 21.3
Noise only
19.8 ± 7.0
22.0 ± 8.6
39.7 ± 18.8
45.0 ± 18.3
39.2 ± 19.3
Probability in
ANOVA
< 0.0001
< 0,0001
0.093
0.024
0.008
value in the
noise only
group is
significantly
less than in the
others (p <
0.0001)
value in the
noise only
group is
significantly
less than in the
others (p <
0.0001)
value in the
noise only
group is
significantly
less than in the
others (p =
0.044)
value in the
noise only
group is
significantly
less than in the
others (p =
0.010)
value in the
noise only
group is
significantly
less than in the
others (p =
0.003)
Comments
Table18. Audiogram – left ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
27.4 ± 12.0
31.3 ± 15.3
45.6 ± 20.1
51.8 ± 20.7
49.3 ± 21.9
CS2 + noise
36.4 ± 14.9
42.1 ± 19.1
52.8 ± 20.8
58.9 ± 22.9
58.1 ± 22.6
Noise only
17.1 ± 8.6
20.7 ± 9.9
38.6 ± 18.8
39.8 ± 15.5
35.3 ± 18.0
Probability in
ANOVA
< 0.0001
< 0.0001
0.041
0.004
0.001
value in the
noise only
group is
significantly
less than in the
others (p <
0.0001)
value in the
noise only
group is
significantly
less than in the
others (p <
0.0001) and
CS2 + noise is
significantly
greater than
CS2 only (p =
0.001)
value in the
noise only
group is
significantly
less than in the
CS2 + noise
group (p =
0.012)
value in the
noise only
group is
significantly
less than in the
others (p <
0.0001)
value in the
noise only
group is
significantly
less than in the
others (p =
0.0002)
Comments
197
NoiseChem 198
June 2004
Table19. Frequency of impairments (threshold 25 dB).
Right ear
Study group
Left ear
Right and/or left ear
n
%
n
%
n
%
CS2 only
42
97.7
41
95.3
43
100.0
CS2 + noise
27
100.0
27
100.0
27
100.0
Noise only
25
43.9
26
45.6
33
57.9
Comments
frequency of
impairments at 25 dB
threshold in group noise
only is significantly less
(p < 0.0005) than in the
others groups
frequency of
impairments at 25 dB
threshold in group noise
only is significantly less
(p < 0.0005) than in the
others groups
frequency of
impairments at 25 dB
threshold in group noise
only is significantly less
(p < 0.0005) than in the
others groups
Table 20. Hearing impairment (threshold 25 dB). Odds Ratios (OR) adjusted to age and gender
Right ear
Study group
OR
Left ear
probability
OR
Right and/or left ear
probability
probability
CS2 only
1,00
CS2 + noise
458.8
0.838
808.3
0.828
0.70
0.996
Noise only
0.09
0.035
0.17
0.046
0.0001
0.814
Comments
1,00
OR
1,00
risk of hearing
risk of hearing
no ORs are significantly
impairment in the noise impairment in the noise
different from 1.00
only group is
only group is
significantly less than in significantly less than in
the CS2 groups
the CS2 groups
Table 21. Frequency of impairments (threshold 15 dB).
Study group
Right ear
Left ear
Right and/or left ear
n
%
n
%
n
%
CS2 only
43
100.0
43
100.0
43
100.0
CS2 + noise
27
100.0
27
100.0
27
100.0
Noise only
43
75.4
47
82.5
53
93.0
Comments
frequency of
impairments at 15 dB
threshold in group noise
only is significantly less
(p = 0.0001) than in the
others groups
frequency of
impairments at 15 dB
threshold in group noise
only is significantly less
(p = 0.0021) than in the
others groups
frequencies of
impairments at 15 dB
threshold in study
groups are not
significantly different (p
= 0.167)
Table 22. Hearing impairment (threshold 15 dB). Odds Ratios (OR) adjusted to age and gender
Study group
Right ear
Left ear
Right and/or left ear
198
NoiseChem 199
June 2004
OR
probability
OR
probability
probability
CS2 only
1,00
CS2 + noise
0,97
0,999
0,72
0,996
0,97
0,999
Noise only
0,0001
0,821
0,0002
0,834
0,0001
0,890
Comments
1,00
OR
1,00
no ORs are significantly no ORs are significantly no ORs are significantly
different from 1.00
different from 1.00
different from 1.00
DPOAE
Table 23. DPOAE, DP gram – right ear. Adjusted to age and gender
Study group
2000 Hz
3000 Hz
4000 Hz
CS2 only
7.7 ± 10.6
7.8 ± 12.9
6.2 ± 11.2
CS2 + noise
9.7 ± 11.2
10.8 ± 11.2
10.3 ± 13.9
Noise only
2.8 ± 5.6
- 5.2 ± 7.0
- 11.6 ± 7.9
0.108
< 0.0005
< 0.0005
value in the noise
only group is
significantly less than
in the CS2 + noise
group (p = 0.044)
value in the noise
only group is
significantly less than
in the others (p =
0.0001)
value in the noise
only group is
significantly less than
in the others (p <
0.0001)
Probability in ANOVA
Comments
Table 24. DPOAE, Signal to Noise Ratio – right ear. Adjusted to age and gender
Study group
2000 Hz
3000 Hz
4000 Hz
17.9 ± 9.1
19.5 ± 13.9
18.6 ± 11.3
CS2 + noise
20.2 ± 10.7
21.9 ± 11.8
20.8 ± 12.1
Noise only
10.7 ± 5.8
8.9 ± 7.5
5.1 ± 8.5
0.006
0.003
0.001
value in the noise
only group is
significantly less than
in the others (p =
0.003)
value in the noise
only group is
significantly less than
in the others (p =
0.002)
value in the noise
only group is
significantly less than
in the others (p =
0.001)
CS2 only
Probability in ANOVA
Comments
199
NoiseChem 200
June 2004
Table 25. DPOAE, DP gram – left ear. Adjusted to age and gender
Study group
2000 Hz
3000 Hz
4000 Hz
CS2 only
5.5 ± 12.2
5.3 ± 11.2
6.4 ± 14.0
CS2 + noise
7.2 ± 15.8
10.7 ± 16.0
12.4 ± 14.1
Noise only
1.9 ± 6.8
- 3.5 ± 6.6
- 9.6 ± 7.1
0.452
0.005
< 0.0005
Probability in ANOVA
Comments
value in the noise
only group is
significantly less than
in the others (p <
0.0001)
no two groups are
value in the noise
significantly different only group is
at the 0.05 level
significantly less than
in the others (p =
0.001)
Table 26. DPOAE, Signal to Noise Ratio – left ear. Adjusted to age and gender
Study group
2000 Hz
3000 Hz
4000 Hz
CS2 only
16.2 ± 11.2
16.7 ± 11.6
18.8 ± 11.8
CS2 + noise
18.1 ± 14.2
21.9 ± 15.9
24.8 ± 14.4
Noise only
11.2 ± 7.2
11.5 ± 7.1
7.4 ± 7.6
0.196
0.067
0.001
Probability in ANOVA
Comments
no two groups are
value in the noise
significantly different only group is
at the 0.05 level
significantly less than
in the CS2 + noise
group (p = 0.021)
value in the noise
only group is
significantly less than
in the others (p =
0.0003)
Table 27. Age and frequency of men and women in study groups
Study group
Men
Age
Women
n
%
n
%
CS2 only
54.5 ± 7.3
38
88,4
5
11,6
CS2 + noise
58.3 ± 6.6
23
85,2
4
14,8
Probability
0.032
Comments
0.726
the CS2 only frequencies of men and women are similar (in the statistical
sense) in compared groups
group is
significantly
younger than
the CS2 +
noise group
200
NoiseChem 201
June 2004
Audiometry
Table 28. Audiogram – right ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
3000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
35.5 ± 16.2
41.9 ± 16.7
50.5 ± 18.9
57.2 ± 20.0
65.5 ± 22.9
62.8 ± 25.2
CS2 + noise
39.6 ± 17.1
45.5 ± 20.5
52.0 ± 21.2
57.6 ± 21.3
66.5 ± 20.1
64.7 ± 21.3
Probability in
ANOVA
0.294
0.367
0.728
0.916
0.822
0.713
Comments
compared
compared
compared
compared
compared
compared
groups are
groups are
groups are
groups are
groups are
groups are
not
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
different
different
different
different
different
different
Table 29. Audiogram – left ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
3000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
30.8 ± 12.0
36.0 ± 15.3
45.4 ± 19.0
53.2 ± 20.1
59.1 ± 20.7
57.3 ± 21.9
CS2 + noise
38.5 ± 14.9
45.0 ± 19.1
53.4 ± 20.3
59.5 ± 20.8
64.2 ± 22.9
64.0 ± 22.6
Probability in
ANOVA
0.016
0.024
0.079
0.169
0.297
0.173
value in the
CS2 only
group is
significantly
less than in
the CS2 +
noise group
value in the
CS2 only
group is
significantly
less than in
the CS2 +
noise group
Comments
compared
compared
compared
compared
groups are
groups are
groups are
groups are
not
not
not
not
significantly significantly significantly significantly
different
different
different
different
201
NoiseChem 202
June 2004
Table 30. Test SISI – right, 1 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
27
87.1
3
9.7
1
3.2
31
100.0
CS2 + noise
17
89.5
2
10.5
0
0.0
19
100.0
Probability
0.701
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 31. Test SISI – right, 2 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
24
77.4
6
19.4
1
3.2
31
100.0
CS2 + noise
15
88.2
2
11.8
0
0.0
17
100.0
Probability
0.608
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 32. Test SISI – right, 4 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
10
38.5
14
53.8
2
7.7
26
100.0
CS2 + noise
7
43.8
7
43.8
2
12.5
16
100.0
Probability
0.704
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
202
NoiseChem 203
June 2004
Table 33. Test SISI – left, 1 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
28
93.3
2
6.7
0
0.0
30
100.0
CS2 + noise
18
90.0
1
5.0
1
5.0
20
100.0
Probability
0.450
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 34. Test SISI – left, 2 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
23
76.7
6
20.0
1
3.3
30
100.0
CS2 + noise
17
89.5
1
5.3
1
5.3
19
100.0
Probability
0.332
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 35. Test SISI – left, 4 kHz
Study group
Equal to 0 %
Greater than 0 %
and less than or
equal to 60 %
Greater than 60 %
Total
n
%
n
%
n
%
n
%
CS2 only
10
37.0
13
48.1
4
14.8
27
100.0
CS2 + noise
11
68.8
4
25.0
1
6.3
16
100.0
Probability
0.150
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
203
NoiseChem 204
June 2004
Table 36. Impedance audiometry – tympanometry. Right
Study group
A
B
C
Total
n
%
n
%
n
%
n
%
CS2 only
36
87.8
0
0.0
5
12.2
41
100.0
CS2 + noise
20
80.0
1
4.0
4
16.0
25
100.0
Probability
0.386
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 37. Impedance audiometry – tympanometry. Left
Study group
A
B
C
Total
n
%
n
%
n
%
n
%
CS2 only
35
83.3
2
4.8
5
11.9
42
100.0
CS2 + noise
18
69.2
3
11.5
5
19.2
26
100.0
Probability
0.345
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 38. Impedance audiometry – tympanometry. PK Press. Right
Study group
- 50 ÷ + 50 (correct)
Less than – 50
Total
n
%
n
%
n
%
CS2 only
36
87.8
5
12.2
41
100.0
CS2 + noise
20
80.0
5
20.0
25
100.0
Probability
0.485
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
204
NoiseChem 205
June 2004
Table 39. Impedance audiometry – tympanometry. PK Press. Left
Study group
- 50 ÷ + 50 (correct)
Less than – 50
Total
n
%
n
%
n
%
CS2 only
36
90.0
4
10.0
40
100.0
CS2 + noise
20
80.0
5
20.0
25
100.0
Probability
0.288
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
Table 40. Acoustic Reflex (ART) – right ear. Adjusted to age and gender
Study group
1000 Hz
2000 Hz
4000 Hz
CS2 only
91.3 ± 8.3
93.8 ± 8.8
90.8 ± 6.8
CS2 + noise
96.2 ± 8.9
98.8 ± 8.5
89.9 ± 4.7
0.065
0.091
0.736
compared groups are
not significantly
different
compared groups are
not significantly
different
compared groups are
not significantly
different
Probability in ANOVA
Comments
Table 41. Acoustic Reflex (ART) – left ear. Adjusted to age and gender
Study group
1000 Hz
2000 Hz
4000 Hz
CS2 only
94.0 ± 7.6
97.8 ± 12.8
89.5 ± 7.4
CS2 + noise
95.6 ± 10.2
94.3 ± 10.2
87.0 ± 8.2
0.562
0.399
0.454
compared groups are
not significantly
different
compared groups are
not significantly
different
compared groups are
not significantly
different
Probability in ANOVA
Comments
205
NoiseChem 206
June 2004
Table 42. Gradient. Adjusted to age and gender
Study group
Right ear
Left ear
CS2 only
77.9 ± 32.9
78.5 ± 37.3
CS2 + noise
65.8 ± 30.3
72.6 ± 34.8
0.130
0.536
compared groups are not
significantly different
compared groups are not
significantly different
Probability in ANOVA
Comments
ABR latencies
Table 43. ABR latencies (ms) – right ear. Adjusted to age and gender
Study
group
I
III
V
CS2 only
1.73 ± 0.26
4.03 ± 0.26
CS2 +
noise
1.71 ± 0.28
Probability
in ANOVA
Comments
Interpeak differences
I - III
III - V
I-V
6.03 ± 0.42
2.28 ± 0.34
2.07 ± 0.66
4.28 ± 0.49
4.07 ± 0.29
5.94 ± 0.33
2.36 ± 0.41
1.94 ± 0.24
4.22 ± 0.39
0.870
0.592
0.390
0.454
0.424
0.637
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
Table 44. ABR latencies (ms) – left ear. Adjusted to age and gender
Study
group
I
III
V
CS2 only
1.74 ± 0.24
4.12 ± 0.42
CS2 +
noise
1.82 ± 0.44
Probability
in ANOVA
Comments
Interpeak differences
I - III
III - V
I–V
6.10 ± 0.50
2.38 ± 0.47
1.98 ± 0.38
4.36 ± 0.59
4.03 ± 0.71
6.16 ± 1.25
2.17 ± 0.60
2.18 ± 0.80
4.33 ± 1.10
0.384
0.573
0.800
0.179
0.287
0.916
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
compared
groups are
not
significantly
different
206
NoiseChem 207
June 2004
DPOAE
Table 45. DPOAE, DP gram – right ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
10.5 ± 9.0
8.4 ± 9.7
6.3 ± 10.6
4.8 ± 12.9
2.3 ± 11.2
- 1.4 ± 11.2
CS2 + noise
11.9 ± 9.6
7.4 ± 10.6
9.5 ± 11.2
9.1 ± 11.2
7.1 ± 13.9
7.5 ± 13.3
Probability in
ANOVA
0.670
0.756
0.378
0.327
0.328
0.113
Comments
compared
compared
compared
compared
compared
compared
groups are
groups are
groups are
groups are
groups are
groups are
not
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
different
different
different
different
different
different
Table 46. DPOAE, Signal to Noise Ratio – right ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
13.8 ± 8.9
15.5 ± 9.9
16.2 ± 9.1
16.1 ± 13.9
14.7 ± 11.3
12.4 ± 11.7
CS2 + noise
17.3 ± 10.2
14.1 ± 9.9
18.9 ± 10.7
19.6 ± 11.8
17.6 ± 12.1
19.2 ± 12.0
Probability in
ANOVA
0.300
0.664
0.403
0.464
0.542
0.232
Comments
compared
compared
compared
compared
compared
compared
groups are
groups are
groups are
groups are
groups are
groups are
not
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
different
different
different
different
different
different
Table 47. DPOAE, DP gram – left ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
9.4 ± 9.2
6.9 ± 10.8
4.3 ± 12.2
3.8 ± 11.2
3.5 ± 14.0
- 0.5 ± 13.9
CS2 + noise
13.8 ± 12.3
10.5 ± 12.1
6.8 ± 15.8
9.0 ± 15.9
10.5 ± 14.1
3.1 ± 10.7
Probability in
ANOVA
0.230
0.341
0.601
0.305
0.199
0.488
Comments
compared
compared
compared
compared
compared
compared
groups are
groups are
groups are
groups are
groups are
groups are
not
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
different
different
different
different
different
different
207
NoiseChem 208
June 2004
Table 48. DPOAE, Signal to Noise Ratio – left ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
15.0 ± 8.1
14.1 ± 10.5
14.5 ± 11.2
15.0 ± 11.6
15.9 ± 11.8
11.4 ± 12.2
CS2 + noise
18.7 ± 13.3
17.6 ± 11.7
17.1 ± 14.2
19.6 ± 16.2
22.6 ± 14.4
15.4 ± 11.2
Probability in
ANOVA
0.306
0.327
0.522
0.373
0.189
0.402
Comments
compared
compared
compared
compared
compared
compared
groups are
groups are
groups are
groups are
groups are
groups are
not
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
different
different
different
different
different
different
Electroneurography
Table 49. Motor conduction studies – right median nerve. Adjusted to age and gender
Study group
CV (m/s)
Amp (mV)
dLAT (ms/cm)
CS2 only
53.4 ± 6.2
14.0 ± 4.3
0.74 ± 0.09
CS2 + noise
55.2 ± 5.1
13.2 ± 5.1
0.76 ± 0.12
Probability in
ANOVA
0.360
0.649
0.595
Comments
compared groups are not compared groups are not compared groups are not
significantly different
significantly different
significantly different
208
NoiseChem 209
June 2004
Table 50. Motor conduction studies – right pereoneal nerve. Adjusted to age and gender
Right
Study group
Left
CV (m/s)
Amp (mV)
dLAT
(ms/cm)
CV (m/s)
Amp (mV)
dLAT
(ms/cm)
CS2 only
53.4 ± 6.7
11.4 ± 5.0
0.66 ± 0.07
50.7 ± 5.1
8.9 ± 4.9
0.70 ± 0.09
CS2 + noise
51.4 ± 5.7
11.8 ± 4.7
0.71 ± 0.12
50.7 ± 6.7
9.4 ± 5.4
0.79 ± 0.12
Probability in
ANOVA
0.330
0.806
0.126
0.981
0.814
0.030
Comments
compared value in the
compared
compared
compared
compared
groups are CS2 only
groups are
groups are
groups are
groups are
group is
not
not
not
not
not
significantly significantly significantly significantly significantly significantly
less than in
different
different
different
different
different
the CS2 +
noise group
Table 51. Sensory conduction studies – right sural nerve. Adjusted to age and gender
Study group
CV (m/s)
Amp (µV) - min
Amp (µV) – max
CS2 only
51.9 ± 8.9
6.7 ± 3.9
11.6 ± 7.1
CS2 + noise
49.3 ± 6.8
8.9 ± 3.4
12.7 ± 5.9
Probability in
ANOVA
0.400
0.118
0.664
Comments
compared groups are not compared groups are not compared groups are not
significantly different
significantly different
significantly different
ENG - VNG
Table 52. ENG – spontaneous nystagmus
SNG right
SNG left
n
%
n
%
n
%
n
%
n
%
CS2 only
6
14.0
2
4.7
1
2.3
1
2.3
10
100.0
CS2 + noise
4
14.8
1
3.7
1
3.7
0
0.0
6
100.0
Study group
Square waves
SNG
directionally
changeable
Probability
0.749
Comment
distributions of frequencies are similar (in the statistical sense) in
both groups
Total
209
NoiseChem 210
June 2004
Table 53. ENG – eye-tracking test
Normal
Study group
Abnormal
Total
n
%
n
%
n
%
CS2 only
19
57.6
14
42.4
33
100.0
CS2 + noise
16
72.7
6
27.3
22
100.0
HE left, DP
left
Total
Probability
0.248
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
Table 54. ENG – rotatory test
Study
group
Symetry
HE left, DP
right
HE right, DP
left
HE bilateral
n
%
n
%
n
%
n
%
n
%
n
%
CS2 only
2
6.7
22
73.3
4
13.3
1
3.3
1
3.3
30
100.0
CS2 +
noise
4
21.1
7
36.8
6
31.6
2
10.5
0
0.0
19
100.0
Probabilit
y
0.090
Comment
distributions of frequencies are similar (in the statistical sense) in both
groups
Table 55. ENG – optokinetic test
Study group
Normal
Abnormal
Total
n
%
n
%
n
%
CS2 only
21
63.6
12
36.4
33
100.0
CS2 + noise
16
72.7
6
27.3
22
100.0
Probability
0.479
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
210
NoiseChem 211
June 2004
VNG
Table 56. VNG – caloric test. Excitability
Study group
Right
Left
Total
n
%
n
%
n
%
CS2 only
6
85.7
1
14.3
7
100.0
CS2 + noise
2
25.0
6
75.0
8
100.0
Probability
0.041
Comment
Frequency in each group are significantly different
Table 57. VNG – caloric test. DP
Study group
Right
Left
Total
n
%
n
%
n
%
CS2 only
3
42.9
4
57.1
7
100.0
CS2 + noise
6
75.0
2
25.0
8
100.0
Probability
0.315
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
Table 58. VNG – caloric test. CP
Study group
Right
Left
Total
n
%
n
%
n
%
CS2 only
1
14.3
6
85.7
7
100.0
CS2 + noise
5
71.4
2
28.6
7
100.0
Probability
0.103
Comment
distributions of frequencies are similar (in the statistical
sense) in both groups
211
NoiseChem 212
June 2004
Table 59. VNG – caloric test. Rotatory test
Right
Study group
Normal
Left
Total
n
%
n
%
n
%
n
%
CS2 only
7
77.8
2
22.2
0
0.0
9
100.0
CS2 + noise
1
50.0
0
0.0
1
50.0
2
100.0
Probability
0.164
Comment
distributions of frequencies are similar (in the statistical sense)
in both groups
Table 60. Age and frequency of men and women in study groups
Study group
Men
Age
Women
n
%
n
%
CS2 only
54.5 ± 7.3
38
88,4
5
11,6
CS2 + noise
58.3 ± 6.6
23
85,2
4
14,8
N-hexan
31.3 ± 9.7
4
57.1
3
42.9
Probability
< 0.0005
Comments
0.137
frequencies of men and women are similar (in the statistical
N-hexan
sense) in study groups
group is
significantly
younger than
the others
groups
212
NoiseChem 213
June 2004
Audiometry
Table 61. Audiogram – right ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
3000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
27.1 ± 16.2
30.2 ± 16.7
38.9 ± 18.9
44.0 ± 20.0
50.2 ± 22.9
46.8 ± 25.2
CS2 + noise
31.0 ± 17.1
33.5 ± 20.5
40.3 ± 21.2
44.4 ± 21.3
51.4 ± 20.1
49.1 ± 21.3
N-hexan
45.5 ± 13.5
57.4 ± 16.0
57.6 ± 14.0
59.3 ± 15.2
65.4 ± 14.7
63.7 ± 12.5
0.061
0.007
0.154
0.292
0.311
0.277
value in the
N-hexan
group is
significantly
greater than
in the CS2
only group
(p = 0.032)
value in the
N-hexan
group is
significantly
greater than
in the others
(p = 0.014)
Probability in
ANOVA
Comments
no two
no two
no two
no two
groups are
groups are
groups are
groups are
significantly significantly significantly significantly
different at different at different at different at
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
Table 62. Audiogram – left ear. Hearing thresholds adjusted to age and gender
Study group
1000 Hz
2000 Hz
3000 Hz
4000 Hz
6000 Hz
8000 Hz
CS2 only
24.5 ± 12.0
27.3 ± 15.3
34.7 ± 19.0
41.3 ± 20.1
43.3 ± 20.7
43.3 ± 21.9
CS2 + noise
31.9 ± 14.9
36.0 ± 19.1
42.4 ± 20.3
47.5 ± 20.8
51.3 ± 22.9
49.8 ± 22.6
N-hexan
40.0 ± 13.8
47.0 ± 17.9
49.5 ± 13.5
49.6 ± 11.7
55.0 ± 12.7
51.8 ± 13.8
0.006
0.008
0.070
0.274
0.398
0.272
value in the
N-hexan
group is
significantly
greater than
in the others
(p = 0.025)
and CS2 +
noise is
significantly
greater than
CS2 only (p
= 0.018)
value in the
CS2 only
group is
significantly
less than in
the others (p
= 0.027)
Probability in
ANOVA
Comments
no two
no two
no two
no two
groups are
groups are
groups are
groups are
significantly significantly significantly significantly
different at different at different at different at
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
213
NoiseChem 214
June 2004
DPOAE
Table 63. DPOAE, DP gram – right ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
12.3 ± 9.0
13.5 ± 9.7
9.6 ± 10.6
10.2 ± 12.9
8.2 ± 11.2
2.5 ± 11.2
CS2 + noise
13.4 ± 9.6
12.6 ± 10.6
12.8 ± 11.2
14.4 ± 11.2
12.5 ± 13.9
11.2 ± 13.3
N-hexan
4.4 ± 6.1
- 5.2 ± 8.5
- 0.05 ± 2.6
- 3.7 ± 0.7
- 0.1 ± 6.6
0.4 ± 7.7
0.481
0.027
0.289
0.181
0.455
0.232
no two
groups are
significantly
different at
the 0.05
level
value in the
N-hexan
group is
significantly
less than in
the others (p
= 0.022)
Probability in
ANOVA
Comments
no two
no two
no two
no two
groups are
groups are
groups are
groups are
significantly significantly significantly significantly
different at different at different at different at
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
Table 64. DPOAE, Signal to Noise Ratio – right ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
16.0 ± 8.9
20.3 ± 9.9
19.1 ± 9.1
21.2 ± 13.9
19.6 ± 11.3
15.5 ± 11.7
CS2 + noise
19.0 ± 10.2
19.0 ± 9.9
22.0 ± 10.7
24.4 ± 11.8
21.8 ± 12.1
21.9 ± 12.0
5.0 ± 8.0
0.2 ± 7.4
8.1 ± 7.8
8.2 ± 2.5
11.9 ± 6.7
7.3 ± 4.1
0.213
0.015
0.202
0.316
0.641
0.351
no two
groups are
significantly
different at
the 0.05
level
value in the
N-hexan
group is
significantly
less than in
the others (p
= 0.016)
N-hexan
Probability in
ANOVA
Comments
no two
no two
no two
no two
groups are
groups are
groups are
groups are
significantly significantly significantly significantly
different at different at different at different at
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
214
NoiseChem 215
June 2004
Table 65. DPOAE, DP gram – left ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
12.8 ± 9.2
10.4 ± 10.8
6.1 ± 12.2
5.2 ± 11.2
8.1 ± 14.0
3.5 ± 13.9
CS2 + noise
16.5 ± 12.3
13.4 ± 12.1
8.0 ± 15.8
9.4 ± 15.9
14.9 ± 14.1
7.1 ± 10.7
N-hexan
2.5 ± 11.4
3.5 ± 8.1
0.8 ± 9.9
3.3 ± 11.0
1.1 ± 8.2
- 4.1 ± 10.7
0.285
0.499
0.824
0.684
0.311
0.523
Probability in
ANOVA
Comments
no two
no two
no two
no two
no two
no two
groups are
groups are
groups are
groups are
groups are
groups are
significantly significantly significantly significantly significantly significantly
different at different at different at different at
different at different at
the 0.05
the 0.05
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
level
level
Table 66. DPOAE, Signal to Noise Ratio – left ear. Adjusted to age and gender
Study group
1000 Hz
1500 Hz
2000 Hz
3000 Hz
4000 Hz
5000 Hz
CS2 only
18.6 ± 8.1
18.0 ± 10.5
16.8 ± 11.2
16.8 ± 11.6
20.2 ± 11.8
15.3 ± 12.2
CS2 + noise
21.7 ± 13.3
21.1 ± 11.7
18.8 ± 14.2
20.3 ± 16.2
26.6 ± 14.4
19.3 ± 11.2
8.4 ± 9.3
10.1 ± 7.8
9.0 ± 9.4
14.8 ± 13.7
12.3 ± 7.0
6.1 ± 7.7
0.300
0.404
0.659
0.782
0.278
0.357
N-hexan
Probability in
ANOVA
Comments
no two
no two
no two
no two
no two
no two
groups are
groups are
groups are
groups are
groups are
groups are
significantly significantly significantly significantly significantly significantly
different at different at different at different at
different at different at
the 0.05
the 0.05
the 0.05
the 0.05
the 0.05
the 0.05
level
level
level
level
level
level
215
NoiseChem 216
June 2004
right ear
left ear
frequency (Hz)
1000
2000
4000
6000
8000
0
5
5
10
10
15
15
2000
4000
6000
8000
dB
dB
0
frequency (Hz)
1000
20
20
25
25
30
30
35
35
solvents
solvents+noise
noise only
solvents
solvents+noise
noise only
Fig. 1 Pure tone hearing thresholds adjusted to age and gender in exposed study groups
(means and standard error)
Mean PTA thresholds for the three groups show no significant differences
right ear
18
17
17
16
16
15
15
dB
dB
18
14
left ear
14
13
13
12
12
11
11
10
10
2000
2000
solvents
frequency (Hz)
solvents+noise
3000
4000
frequency (Hz)
3000
solvents
solvents+noise
noise only
noise only
Fig. 2 SNR amplitudes adjusted to age and gender in exposed study groups
(means and standard error)
Transient emissions at 2, 3 and 4kHz for the groups show some divergence across frequencies and
ears.
216
NoiseChem 217
June 2004
left ear
10
8
8
6
6
4
4
2
2
0
dB
dB
right ear
10
-2
0
-2
-4
-4
-6
-6
-8
-8
-10
-10
2000
solvents
frequency (Hz)
3000
solvents+noise
frequency (Hz)
noise only
solvents
solvents+noise
noise only
Fig. 3 DP amplitudes adjusted to age and gender in exposed study groups
(means and standard error)
Distortion Product amplitudes across frequencies show best data for solvents only group and worst for
noise. Addition of noise to solvents has significant effect on DP amplitude.
right ear
0
left ear
frequency (Hz)
1000
2000
frequency (Hz)
4000
1000
4000
6000
8000
0
10
10
20
20
30
30
dB
dB
2000
40
40
50
50
60
60
70
70
CS2
CS2+noise
Noise only
CS2
CS2+noise
Noise only
PTA thresholds are significantly worse when CS2 exposure occurs with noise.
217
NoiseChem 218
June 2004
right ear
30
25
25
20
20
15
dB
dB
30
15
10
10
5
5
0
left ear
0
2000
CS2
2000
3000
frequency (Hz)
CS2+noise
3000
4000
frequency (Hz)
Noise only
CS2
CS2+noise
Noise only
Transient otoacoustic emission amplitudes are affected considerably by noise compared to CS2.
Combined exposure reduces amplitude further from the CS2 alone.
218
June 2004
NoiseChem 219
Lab 6: Deepak Prasher
Institute of Laryngology and Otology,
University College London, UK
219
June 2004
NoiseChem 220
Effect of exposure to a mixture of solvents and noise on hearing and balance in aircraft
maintenance workers
Deepak Prasher, Haifa Al-Hijaj, Susan Aylott and Alexander Aksentijevic
Institute of Laryngology and Otology, University College London
Introduction
Toxic nature of solvents is well recognised and in particular their acute and chronic effects on
the central nervous system. Dizziness is a commonly reported feature of the effects but has
not been extensively studied. Furthermore the effect of solvent exposure on hearing has, for
some time been masked by the concomitant presence of noise in the workplace where solvent
exposure occurred. It is beginning to emerge that not only are solvents ototoxic but in the
presence of noise can compound the effect on hearing. Recent studies examining the effect of
solvent exposure alone and in combination with noise have begun to show synergistic effects
on hearing.
Makitie et al (2003) have shown in rats that exposure to styrene at 600 ppm (for 12 h/ day 5
days/ week for 4 weeks) caused a 3dB hearing loss at 8kHz and exposure to noise at 100105dB industrial noise caused a loss between 2-9dB but an exposure to a combination of
styrene and noise caused a flat loss between 23-27 dB. Furthermore the lower concentrations
at 300 and 100 ppm only induced a hearing loss when combined with noise. Clear syngersitic
effect was observed above a critical level.
Sliwinska-Kowalska et al (2001) showed that the relative risk for hearing loss was increased
to 4.4 in workers exposed to a mixture of solvents within the exposure limits. SliwinskaKowalska et al (2003) examined styrene exposed workers and showed that there was almost a
4-fold increase in odds of developing a hearing loss related to styrene exposure and that the
odds ratios were 2-3 times higher when styrene and noise co-exposure existed compared to
each acting alone.
Aircraft maintenance workers are exposed to a mixture of solvents, which include stripping
agents to remove polymer based coatings and residues which contain dimethylacetamide
whilst some other cleaning agents include trichloroethane which has been phased out of use
by greenhouse emission protocols but has been used as a degreaser and cleaner of metals and
plastics. They are also exposed to adhesives, sealants, adhesion promoters, isocyanate, zinc
chromate and glycol ester paints in addition to exposure to hydrocarbon fuels such as jet fuel,
and diesel which are a complex mixture including benzene, n-hexane, toluene, xylenes,
naphthalene. It has been pointed out (Ritchie etal 2001) that while hydrocarbon fuel
exposures occur typically below permissible exposure limits for their constituent chemicals, it
is unknown whether additive or synergistic interactions may result in unpredicted
neurotoxicity.
During occupational use of solvents, absorption into the body is through breathing in solvent
vapour in to the lungs and via contact with skin. It is estimated that breathing occurs around
12 times per minute with five to eight litres of air exchanged every minute. During exercise or
heavy manual work with deep inspiration, the number of litres exchanged may increase to
over a 100 per minute, thereby increasing the solvent absorption through this route. Once
absorbed the body metabolism transforms the solvent into water soluble components that are
excreted in urine. The clinical effects depend on the amount absorbed and the duration of
exposure. For many chemicals an exposure limit is given for an average eight-hour industrial
exposure to which workers may be repeatedly exposed without adverse effects. In addition to
this there are odour detection levels and short term exposure limits. Therefore the key
elements necessary for exposure effect relationship evaluation are the absorption,
biotransformation and toxicity. Furthermore if a combination of exposures occurs then the
interaction between the toxic elements have to be considered.
A summary of the observed effects of some specific solvents are given in Table 1.
220
NoiseChem 221
June 2004
It has been shown (Smith et al 1997) that low-level chronic exposure to jet fuel vapour in
aircraft maintenance personnel can lead to increased postural sway. A significant association
between solvents (benzene, toluene, xylene) and increased postural sway was observed.
Hearing loss and brainstem response latency prolongation were reported by Chen et al (1992)
in aircraft maintenance workers and firemen working at airport facilities. The fact that both
peripheral and central auditory pathway damage was reported by the authors implies that the
effect was not only due to aircraft noise but additional exposures to solvents which may be
responsible for the more central effects observed. Exposure to noise alone rarely affects the
central conduction time in the auditory pathway as indicated by the prolongation of the
brainstem response latencies.
Materials and Methods
Four groups of people were examined, aircraft maintenance workers exposed to solvents and
noise, and mill workers exposed to noise alone, printed circuit board operatives exposed to
solvents only and those exposed to neither acted as controls. The number of subjects tested in
each group with their mean age are shown in Table 1.
Table 1: Study Groups
Exposure
N
group
control
39
Noise
153
solvent mix
13
solvent mix
174
+ noise
Mean (SD)
Age
47.6 (14.8)
53.3 (7.8)
49.6 (14.2)
47.4 (7.5)
Procedure:
The subjects were interviewed to complete a questionnaire (Appendix1) relating to their
personal health and work history and exposure. Each subject had height, weight (to determine
body mass index) and blood pressure measurements taken. Prior to audiometric testing, the
otoscopic examination of the ears was undertaken to rule out any excessive presence of wax
or any perforation or other abnormalities. The following tests were conducted on each subject.
Tympanometry and Acoustic Reflex Measurements:
Tympanometry which measures the compliance of the tympanic membrane with change in
pressure was conducted to exclude any middle ear dysfunction, prior to the pure tone
audiometry. Subjects with conductive hearing loss were excluded from the analysis. An
automatic GSI 38 tympanometer was used for the purpose. Ipsilateral and contralateral reflex
thresholds were also recorded at 500Hz, 1kHz and 2kHz. 4kHz was not considered as it is
very frequently absent.
Pure Tone Audiometry:
Air conduction audiometry was conducted in a sound proof booth using the HughsonWestlake automatic assessment with a Castle Excalibur GA1001 PC based audiometer. The
patient was clearly instructed to respond to the lowest audible sound with a button press and
the threshold recorded automatically by the computerised system. Any significant departures
from the pattern expected over the frequency range were re-examined manually to correct for
any discrepancies. The results were stored on computer for further analysis.
Transient Otoacoustic Emissions:
Transient emissions were evoked by 80dBSPL click stimuli using the Biologic Scout Sport
OAE system with a test protocol covering 1.0kHz to 6kHz range. The subject had the probe
placed in the ear canal with a disposable ear tip of appropriate size to secure a good seal. The
221
NoiseChem 222
June 2004
programme performs an in-the-ear calibration and adjust the stimulus intensity level to the
protocol target value. If the calibration procedure is successful, the programme automatically
moves to the averaging phase. The TOAE reproducibility and response amplitude with
respect to the noise floor value (TE-NF) were analysed for each frequency band namely 1,
1.5,2,3,4kHz and across the whole range1.2-3.4kHz.
Both ears were tested and the results analysed separately and summed across the two ears.
Distortion Product Otoacoustic Emissions:
The distortion product emissions were recorded using the standard ototoxic protocol covering
the 1.5kHz to 10kHz range. The Lower frequency primary tone intensity was was set at 65dB
and higher frequency intensity level was set at 55dB. The ratio of F1 to F2 was set at 1.22.
The reproducibility and the response level with respect to the noise floor (DP-NF) were
analysed by frequency band across the range. There were four points per octave. The test was
repeated for each ear to check for reliability.
Auditory Evoked Potentials:
The auditory brainstem response was recorded using the Biologic Navigator system.
Disposable electrodes were attached to the mastoids, chin and the forehead at the hairline
after cleansing the skin and slight abrasion to reduce contact impedance of the electrodes.
Ipsilateral and contra-lateral responses to stimulation of each ear separately were recorded on
two occasions with an alternating click stimulus presented at 80dB. The responses were
recorded over a time window of 10ms and analysed for Waves I, III and V latencies.
Responses were considered outside the normal range if the inter-wave interval between I-V
exceeded 4.4ms or if either wave I or III were present in the absence of Wave V.
Nystagmography:
Video-nystagmography was performed using the Chartr VNG system. The subject was seated
4 feet away from the light bar, the range sensor provided an indication of the distance
between the light bar and the subject. The video goggles were positioned on the subject’s eyes
whilst the subject looked straight at the centre of the light bar. The video image was viewed
and the subject’s pupils were aligned prior to the test session by adjusting the camera and the
mirrors. The brightness and contrast adjustments were made to acquire a satisfactory image.
The saccades were recorded to horizontal random position of light on the bar. Light target
moved randomly every 1.25 seconds over a 34 degree arc. Horizontal eye position and target
position were recorded. The gaze test consisted of the measurement of horizontal eye position
with light target centred, 30 degrees to right and left in the light and without vision (cover on
goggles). The tracking or pursuit test examined horizontal eye position with the light target
moving sinusoidally
at frequencies of 0.2,0.3,0.4,0.5,0.6 and 0.7 Hz over a 34 degree arc. The optokinetic test
consisted of a multiple light target moving right/leftward at 20 to 60 degrees per second.
Posturography:
The posturographic measurements were made using the NeuroCom system. The subject stood
on a platform initially with eyes open then closed, first on a firm surface then on a foam. Each
test had three trials and the mean sway velocity as well as the centre of gravity alignment
were recorded and compared with control data for an age range within the decade of the
individual under test.
Noise Measurements
Measurements of the noise level over short durations were taken on 9 occasions over three
days over a period of a year. Both Leq and Lpeak levels including levels over the spectral
range were recorded.
Measurement Time Leq
(seconds)
Lpeak
222
NoiseChem 223
June 2004
25
22
34
19
20
16
32
16
10
95.2
95.9
97.9
92.5
59.6
93
97.5
75.1
66.3
112.7
114.2
115.9
108.1
84.9
114
115.2
111.1
108.9
As can be seen from the table above, the range of levels varies on different occasions
depending on whether the aircraft engines are running at the time. The relatively quieter
periods occur when the engine is turned off.
The major difference in the noise exposure between the two groups is the fact that the aircraft
maintenance workers are exposed to a wide range of levels with relative quiet periods
whereas the noise alone group are in constant noise during their work time.
Solvent Exposure
Aircraft maintenance workers were exposed to a complex mixture of solvents which occur in
stripping agents, cleaning fluids, paints, and jet fuel. These are a mixture of benzene, nhexane, toluene, xylenes, naphthalene, trichloroethane, dimethylacetamide, etc. The exposure
was estimated from years of work and the type of work undertaken in that time. It was not
possible to take urine or blood measurements of the workers.
Results
Pure tone audiometry
Predictably, the main effect of frequency was highly significant [*F (3.2, 1039.8) = 377.5, p <
0.001; MSE = 148.1]. Figure 1 below shows that a highly significant main effect of exposure
group [F (1, 320) = 77.4, p < .001; MSE = 1516.5] was due to the fact that the noise group’s
thresholds were higher at all frequencies. The divergence of the two groups’ thresholds at the
upper end of the frequency range was reflected in a significant interaction [F (6, 1920) = 21.1,
p < 0.001; MSE = 80.2].
223
NoiseChem 224
June 2004
PTA
60
50
Mean threshold (dB HL)
40
30
exposure
20
noise
solvent mix + noise
10
0.5
1
2
3
4
6
8
Frequency (kHz)
The pure tone audiometric thresholds for the groups tested are shown in Table 2.
Table 2: Pure Tone Audiometric (PTA) Thresholds (dB)
Exposure group
N
Mean (SD) PTA (dB)
Control
39
20.1 (12.1)
Noise
153
35.3 (17.7)
solvent mix
13
26.9 (14.8)
solvent mix + noise
174
20.8 (11.3)
The number of subjects in the controls and solvents only group were relatively small
compared to the other groups which meant that meaningful threshold comparisons across all
groups were not possible. In addition the mean threshold (across frequencies and ears) for the
noise group was significantly higher at 35.3 dB compared with the solvent and noise
combined group with a mean threshold of 20.8dB. This is largely due to the difference in the
level and duration of noise exposure across these two groups.
It is worth noting that if the mean PTA across the frequency range is considered and a cut-off
of 20dB is applied, 5.6% of the controls fall outside compared with 33.3% of the solvent +
noise group although both groups have very similar mean PTA thresholds.
Unfortunately it was not possible to match the noise exposure levels in the two groups.
PTA:Ear Differences
There were significant differences (Table 3) between ears at 500Hz and 1kHz for the controls
but not at any other frequency, for noise 500Hz, 2kHz, 3kHz, 4kHz, and 8kHz showed
significant differences but none were observed at 1kHz or 6kHz. For solvents and noise
group, significant differences were observed at 1kHz, 4kHz and 6kHz but not at 500Hz,
2kHz, 3kHz and at 8kHz.
Table3: Significant differences in threshold between ears
224
NoiseChem 225
June 2004
Audiometry
Controls
Noise
Solvents+
Noise
Significant LT vs RT
Difference at Frequencies
500Hz
1k
500Hz
2k 3k
1k
-
4k
4k
6k
8k
-
Distortion product emissions
The distortion amplitude declined with frequency, as confirmed by a highly significant main
effect of frequency [*F (6.5, 1244.1) = 22.8, p < .001; MSE = 50]. In addition, the noise
group exhibited lower DP amplitude compared to sm + n group [F (1, 191) = 28.7, p < .001;
MSE = 301]. However, the downward trend was reversed at the high end of the frequency
range. Although, both groups showed recovery from 6 kHz upwards, the reversal was
particularly prominent for the noise group at 7.7 kHz. This was reflected in a highly
significant interaction [F (11, 2101) = 12.6, p < .001; MSE = 29.6].
Mean distortion product emissions
14
12
10
8
Response (dB)
6
4
Exposure group
2
noise
0
solvent mix + noise
1.4
2.1
3.0
4.2
5.9
8.4
Frequency f2 (kHz)
225
NoiseChem 226
June 2004
Transient Otoacoustic Emissions (TOAE) S/N ratio
A mixed-design ANOVA was used to examine the between-group difference on the TOAE
amplitude. As in the previous analyses, the noise group was more affected [F (1, 191) = 24, p
< .001; MSE = 44.3]. The highly significant main effect of frequency [*F (2.9, 555) = 18.4, p
< .001; MSE = 157.3] reflected the peak at 1.5 kHz. The interaction failed to reach
significance.
TOAE S/N ratio
7
6
TOAE amplitude (dB SPL)
5
4
3
exposure
2
noise
solvent mix + noise
1
1
1.5
2
3
4
Frequency (kHz)
TOAE reproducibility
The main effect of frequency was highly significant [*F (2.8, 526.9) = 27.7, p < .001; MSE =
460.7], as was the main effect of exposure group [F (1, 190) = 21.4, p < .001; MSE = 2256.7].
The reproducibility for the noise group was lower by approximately 20 %.
226
NoiseChem 227
June 2004
TOAE reproducibility
80
70
Reproducibility (%)
60
50
exposure
40
noise
solvent mix + noise
30
1
1.5
2
3
4
Frequency (kHz)
Auditory Brainstem Responses
A significant number of subjects (32.4%) had abnormally prolonged inter-wave interval
(WaveI-V) in the solvent and noise group. However, a comparison of the mean latencies for
all waves across the groups failed to reveal any differences.
Acoustic Reflex Thresholds
Mean Ipsilateral and contralateral reflex thresholds for the right and left ears for Noise and
Solvents and Noise Groups are shown in Table 4 below. The only significant differences in
the mean reflex thresholds were observed in the contralateral recordings from the right ear at
500Hz, 1kHz and 2kHz.
Table 4: Mean Acoustic Reflex Thresholds
Reflex
Group
Right
Ear
Mean
Threshold+SD
Ipsi500
Noise
91.9+9.2
S+N
92.0+10.8
Ipsi1k
Noise
91.5+9.0
S+N
90.8+10.7
Ipsi2k
Noise
93.0+8.6
S+N
93.0+10.9
Contra500
Noise
95.7+8.21*
S+N
90.4+27.5
Contra1k
Noise
93.4+7.6*
S+N
89.1+25.2
Contra2k
Noise
96.3+6.7*
S+N
91.1+24.5
* significance p<0.05
Left Ear
Mean
Threshold+SD
90.9+8.5
92.3+7.7
91.8+8.7
91.6+7.5
94.1+8.4
93.8+7.6
94.9+8.2
92.7+22.5
92.8+8.4
91.4+20.6
95.9+7.8
92.2+21.1
No significant difference was observed between Right Ipsilateral and Left
Ipsilateral or between Right Contralateral and Left Contralateral reflex thresholds in any
group. But as shown in Table 5 below the correlation between ipsilateral and contralateral
227
NoiseChem 228
June 2004
reflex thresholds for the right ear and left ear were significant for almost all frequencies for
the control and noise groups but not significant for the left ear for the solvent and noise group.
Furthermore the threshold difference between ipsilateral and contralateral reflex thresholds
(shown in Table 6) is only found to be significant for the control and noise groups and not for
the solvent+noise group. These findings indicate a pattern of differences in reflex
measurements which differentiate between noise and solvent+noise group. The contralateral
pathway appears to be differentially affected by solvent exposure. The number of subjects in
the solvent +noise group that had an absent reflex ipsilaterally was 25.1% compared with
41.2% contralaterally.
Table 5: Correlations between Ipsi and contralateral reflex thresholds across groups
Ipsi
vs Controls
Noise
Solvents +Noise
Contra
Correlations
Ear/Freq
Correlation Significance Correlation Significance Correlation Significance
R500
0.19
0.342
0.61
0.000
0.18
0.036
R1k
0.48
0.008
0.43
0.002
0.16
0.045
R2k
0.39
0.028
0.26
0.092
0.28
0.001
L500
0.44
0.022
0.36
0.013
0.04
0.654
L1k
0.82
0.000
0.52
0.000
0.06
0.491
L2k
0.65
0.000
0.77
0.000
0.03
0.702
Table 6: Mean Reflex Threshold
across Groups
Controls
Ear/Frequency Ipsi-contra
reflex
Threshold
R500
-11.1
R1k
-5.7
R2k
-1.78
L500
-11.5
L1k
-6.2
L2k
-3.9
differences between Ipsilateral and contralateral reflexes
Sig
0.000
0.000
0.170
0.000
0.000
0.003
Noise
Ipsi-contra
reflex
Threshold
-5.5
-3.13
-4.17
-5.32
-1.6
-2.02
Sig
0.000
0.012
0.005
0.000
0.159
0.015
Solvents+Noise
Ipsi-contra
reflex
Threshold
1.56
1.56
1.52
-1.44
-0.03
1.69
Sig
0.51
0.46
0.43
0.45
0.98
0.38
Posturography
Subjects were examined under four conditions namely on firm surface with eyes open and
closed and on foam with eyes open and closed. Mean centre of gravity sway velocity outside
the age-adjusted normal values constituted an abnormality. 32.3% of subjects in the solvents
and noise group had an abnormal posturographic finding.
228
June 2004
NoiseChem 229
Video-nystagmography
Table 7: VNG Test results in Solvents +Noise Group (n=140)
Test
% Normal % Abnormal % Indeterminate
Saccades 19
74
6
Gaze
69
6
26
Pursuit
39
56
6
OKN
18
45
37
Saccades
Three aspects of saccadic activity were analysed in the solvent and noise group. These were
the accuracy of reaching target position, latency to reach target and the velocity of movement
to target. Abnormality in any one or combination of aspects were considered abnormal if
outside the age adjusted normal control values. It can be seen from Table 7 that 74% of
subjects had abnormal saccadic activity.
Gaze
Any nystagmic activity in the centre, left or right gaze were recorded and analysed. Only 6%
of subjects showed any significant nystagmus in any position of gaze.
Pursuit
Eye movements to following of a target stimulus were recorded and analysed with respect to
velocity, gain and any asymmetry in the response. Abnormality in any aspect constituted a
response abnormality. 56% of subjects showed an abnormality.
OKN
Optokinetic nystagmus was recorded at three speeds of target rotation. The velocity and any
asymmetry beyond age-adjusted normal values were considered abnormal under any speed of
target. 45% of the subjects showed some abnormal OKN function.
Discussion
There have been a number of limitations to this study, the four groups originally envisaged of
the same size proved to be an impossible task due to the great reluctance of the industrial
concerns to take part in this study. Much time was wasted in many meetings trying to
persuade the printing, manufacturing and other industries to take part in the study. Both
management and unions as well as the European associations of major industrial groups were
approached and several presentations were made by the NoiseChem group to encourage these
groups to participate but none were willing for fear of litigation from their members.
Thus it proved that we were left with unmatched groups at the end of the study as the time for
the study was limited and no extensions were permitted. Within the constraints of the groups
and the absence of a match for noise exposure level between the noise alone and
solvent+noise group only limited conclusions may be drawn from this study. However many
interesting and valid observations have been possible and with grouping of data across labs
within the NoiseChem group further conclusions may be drawn from the studies conducted by
the group.
It is clear from the data that despite the control and solvents group having a similar mean age
and mean PTA thresholds, 33.3% of the solvent group had a hearing loss (mean PTA
threshold across the frequency range greater than 20dB) compared to 5.6% of the controls.
The noise alone group had a greater exposure to noise than the solvent and noise, and this was
confirmed by the greater mean loss of 35.3dB compared to solvents+noise group of 20.8 dB.
The distribution of the loss across the frequency range was very similar for the two groups
showing a significant shift in threshold from 2kHz with a maximum at 6kHz in both groups.
As it was not possible to match the noise exposure in these two groups, it is not possible to
comment further on the combined exposure although the solvents group without noise
229
June 2004
NoiseChem 230
exposure did show a greater mean loss than the controls and solvents +Noise groups.
However the significant difference in the number of subjects across groups requires caution in
further interpretation.
Further analysis of the comparison across the ears revealed interesting differences between
groups. There were no significant differences between the right and left ears for the noise
alone group at 1kHz and 6kHz but the differences were significant for the Solvents+Noise
group at these frequencies. The significance of the asymmetry due to the combined exposure
is not clear but may be associated with more central effects on the auditory pathway.
The comparison of the noise alone and solvents +noise groups for distortion product
emissions reveals that both groups show a decline in DP amplitude with frequency as would
be expected from the hearing threshold deterioration in the high frequencies in both groups.
The mean level of DP amplitude was significantly worse for the noise alone group
particularly at around 4kHz. It is interesting to note that although both groups show a
significant recovery of emission amplitude from 6kHz, this was much more pronounced for
the noise than with solvents+noise group.
The transient emissions amplitude and reproducibility showed a similar pattern of decline
with frequency and being worse for the noise alone group.
It is clear that the emissions reflect the relative noise exposure in the two groups. The solvent
+noise group with periods of relative quiet fared better than those with continuous high noise
exposure levels.
Ipsilateral and contralateral acoustic reflex thresholds for both the right and left ears were
compared across the two groups. A significant difference in the mean acoustic reflex
threshold between noise alone and solvents+noise group was only observed in the
contralateral reflex thresholds in the right ear. This asymmetric threshold difference also adds
to the asymmetry observed in the pure tone audiometric thresholds at certain frequencies. The
ipsilateral/contralateral difference in reflex threshold was significant at most frequencies for
both ears for the control and noise alone group but not at any frequency for either ear for the
solvents+noise group. This significant observation shows an absence of any increase in reflex
thresholds contralaterally for the solvents+noise group. Normally there is 5-7 dB increase in
contralateral reflex threshold which is absent in the solvents+noise group. As there is no
difference in the ipsilateral reflex thresholds across groups, the lower contralateral reflex
thresholds in the solvents+noise groups is the key difference between groups.
Another interesting finding is that the correlation between ipsi and contralateral reflexes in the
control and noise groups is highly significant especially for the left ear but is not at all
significant for the solvents+noise group especially for the left ear.
These findings of altered contralateral reflexes imply an effect of solvents on the central
crossed auditory pathway as no deficiencies were observed in the noise alone group or the
controls.
The effect of solvent exposure on the central auditory nervous system is further supported by
abnormalities of the auditory brainstem response in the solvent+noise group. In this group
prolongation of central conduction time for the auditory brainstem response wave I-V
interval, presence of Waves I and III in the absence of Wave V, and unrepeatable responses
were observed although no significant differences in the mean latencies of the Waves was
noted across groups.
Furthermore, effects on the balance system were observed with posturographic recordings
showing an abnormality of postural sway in 32.3% of subjects in the solvents+noise group.
These investigations have shown that there are clear observable effects on the audiovestibular system of solvent+noise exposure. The hearing impairment shows a similar
pattern to that observed with noise alone and is dependent on the level and duration of noise
exposure. It is clear that the relative quiet periods in the exposure for the solvent+noise group
compared with the continuous high level of noise in the noise alone group provided some
protection in that the mean loss in the noise alone group was significantly worse than that for
the solvent+noise group. However, in the solvent group changes to the central auditory
pathway were observed with contralateral reflex thresholds showing group differences.
230
June 2004
NoiseChem 231
Auditory brainstem response abnormalities in the solvent+noise group also indicated that the
central auditory pathway was affected in this group. This is in agreement with a study of
airport employees by Chen et al (1992) who reported prolongation in central conduction time
in intervals I-V and III-V.
Postural sway abnormalities were detected in about a third of the aircraft maintenance
workers. This is in agreement with Smith et al (1997) who showed a significant association
between solvent exposure and increased postural sway response. Clearly, the solvents have a
subtle influence on the vestibular proprioceptive interaction.
Occupational exposures to solvents are common particularly mixtures of solvents. Odour
detection for some solvents can occur at very low concentrations. The threshold limit value or
TLV is an average eight-hour exposure to which workers may be repeatedly exposed without
harmful effects. These values are defined for specific solvents but not for any combinations as
found in work environments. The determination of limit values considers effects on animals
and where possible humans and are based on adverse effects in terms of carcinogenicity or
toxic effects on the central nervous system. The effect on hearing or balance systems is rarely
considered in the setting of limit values. Solvent exposure has been implicated in specific
sensory impairment as for example colour perception or hearing damage but again there is
very little research which has examined damage of the senses in the same individual worker.
Styrene, toluene, n-hexane and carbon disulfide have been shown to affect both colour vision
and hearing.
The effect of a mixture of solvents on the auditory system appears to occur both at the end
organ level as well as in the nervous pathway.
References
Chen TJ, Chiang HC., Chen SS. Effects of Aircraft noise on hear8ing and auditory pathway
function of airport employees. J. Occup. Med, 1992 Jun; 34(6): 613-9
Mäkitie AA., Pirlova U., Pyykkö I., Sakakibara H., Riihimäki V., Ylikoski J. The ototoxic
interaction of styrene and noise. Hearing research, 2003, 179:9-20
Ritchie GD., Still KR., Alexander WK., Nordholm AF., Wilson CL., Rossi J. 3rd , Mattie DR.
A review of the neurotoxicity risk of selected hydrocarbon fuels. J. Toxicol. Environ Health B
Crit. Rev., 2001 Jul-Sep;4(3):223-312.
Sliwinska-Kowalska M., Zamyslowska-Szmytke E., Szymczak W., Kotylo P., Fiszer M.,
Wesolowski W., Pawlaczyk-Luszczynska M. Ototoxic effects of occupational exposure to
styrene and co-exposure to styrene and noise. J. Ocupp. Environ. Med., 2003 Jan, 45(1):1524.
Sliwinska-Kowalska M., Zamyslowska-Szmytke E., Szymczak W., Kotylo P., Fiszer M.,
Dudarewicz A., Wesolowski W., Pawlaczyk-Luszczynska M., Stolarek R. Hearing Loss
among workers exposed to moderate concentrations of solvents. Scand. J. Work. Environ
Health., 2001 Oct; 27(5):335-42.
Smith LB., Bhattacharya A., Lemasters G., Succop P., Puhala E. 2nd, Medvedovic M., Joyce
J. Effect of Chronic low-level exposure to jet fuel on postural balance of US Air Force
personnel. J. Occup Environ Med., 1997 Jul; 39(7):623-32.
231
June 2004
NoiseChem 232
Effects of Combined Environmental Exposure to Noise and Solvents on Factory
Workers across Europe: Preliminary Data analysis of combined data
Deepak Prasher and Alexander Aksentijevic
232
NoiseChem 233
June 2004
TABLE 1: Combined databases
Lab
Exposure group
N
Age
1
Noise
8-h exposure
control
noise (n)
styrene
styr. + n
Total
59
78
65
88
290
26.5 (11.2)
32 (9.1)
31.3 (10.4)
31 (6.3)
2
< 70 dB
88.9 (4.2)
73.6 (9.5)
88.2 (3.2)
8-h exposure
control
noise
97
17
26.5 (11.1)
43.3 (11.1)
< 70 dB
87.4 (5.8)
styrene
41
31.5 (9.9)
73.6 (9.5)
styr. + n
62
32 (8.8)
88.9 (4.2)
solv. mix
solv. mix + n
Total
20
51
288
27.2 (2.7)
30.6 (6.4)
76.5 (3.3)
88.2 (3.3)
3
Mean life exposure
control
noise
110
154
37.5 (9.1)
39.5 (9.3)
75.7 (2.2)
91.3 (4)
styrene
114
35.2 (8.9)
79.7 (3.8)
solv. mix + n
Total
137
515
41.1 (8.3)
91.7 (5.5)
noise
styrene
Total
Carbon
disulfide
solvent mix
solv. mix + n
Total
180
126
306
77
control
noise
solv. mix
solv. mix + n
Total
TOTAL
39
153
13
174
379
2117
Styrene
Mean PTAa
MA
in
urine
27.9 (18.3)
26.5 (23.3)
16.3 (2.7) 36.3 (19.2)
11.5 (1.2) 32.1 (20.1)
MA
urine
in
5.2 (6.8)
18.1 (11.4)
42.1
(49.2)
25.7
(34.4)
11 (8.3)
12.2 (9.6)
9.1 (4.2)
8 (8.3)
MA
urine
in
14.5 (7.7)
23.1 (14.6)
11.2
(10.5)
20.6 (10.2)
23.8 (11.6)
4
5
63
199
339
6
ALL
47.6 (14.8)
53.3 (7.8)
49.6 (14.2)
47.4 (7.5)
20.1 (12.1)
35.3 (17.7)
26.9 (14.8)
20.8 (11.3)
233
NoiseChem 234
June 2004
Combining Data from Labs
As can be seen from the Table 1above, there was a possible confounding of exposure
condition by age. Specifically, there was a significant difference in age between the exposure
conditions [F (5, 1970) = 25.7, p < .001; MSE = 124.3]. Post-hoc tests indicated that the noise
group was significantly older than all the other groups with the exception of the solvent mix
group (Tukey’s HSD, p = .001). In addition, the solvent mix + noise group was significantly
older than the control, styrene and styrene and noise groups (p = .001). Since the increase in
age for the noise and solvent mix + noise groups paralleled the increase in the level of noise
exposure, cases were removed from these two groups (criterion of 50 years) in order to
equalise the mean age values.
Data from the six laboratories (Johnsonn, Pawlas, Sliwinska-Kowalska, Starck, Sulkowski
and Prasher) were concatenated into a single database in order to examine the general effect
of noise and solvent exposure on Pure Tone Audiometric (PTA) thresholds. Although overall
there were six exposure conditions, one condition (solvent mix; Pawlas) was excluded from
the analyses because of a small number of subjects (N = 20). This left five conditions
(controls, noise, styrene, styrene + noise and solvent mix + noise; see Table above). Although
the PTA stimulus frequency range was from 0.25 to 16 kHz, only the frequencies that all six
laboratories had in common (1, 2, 4, 6 and 8 kHz) were examined.
Age and Smoking
As a preliminary step, the relationship was examined between several relevant concomitant
variables (age, smoking, body-mass ratio, noise and styrene exposure) with the mean PTA
threshold values (averaged over ears and frequencies). As expected, age was a reliable
predictor of hearing loss [Y = 11.2 + .81 * age (r2 = .33)].
100
A
Mean PTA threshold (dB HL)
A
75
50
25
0
A
A
A
A
A
A
A A
A
AA A
A A
A
A
A
A A
A
A
A
A
A
A
A
A
A
A
AA
AA A
A AAA A AA A A A
A
AA
AA
A
A
A
A
A
A
A
A
A
A
A
AA
A A AA A
A
A AAA A A
AA
A AA
A AA
AA
AA
AA A
AA A A
A
A
A
A
A
A
AA
A A
AA A A
A A
AA A A
A
A A
A
AA
AAA
AA
AAA
AA A A
A
AA A
A AAAAAAAA
AAAA
AA AAAA
A A A AA A A AAA
A
A
AA A A
A
A
A
A
A
A
AA
AA
AA AA AAA
A
AA A A A
AAA
AAA
AA
AA AA A
A
A
A
A
A
AAA
AAAAA
A A
AA
A A A
AAA
AA
A A
A AA
A
AAA
AA A
A A
A A
A
AAAA A
AAA
AA
A AA
A
A
A AAA
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
AAA A
A
A
A
A
A
A
AA
AA AAAA
AA
AAAA
AAAAAA
AA
AA
AAAA
A
A
A
A
A
A
A
A
A
A
A
A
AAA
AA
AA AA AAA A
A A AA
A
AA A
A
AAAA
AA
A
A
AA
A
AAA
AA
A
AA
AA
AA
AA
AA
A AAA AA
A
AA
A
A
A
A
AA
AA
AA
A A A
A
A
A
AA
A
A
A
AA
A
A
AAA
AAA
A
AAA
AA
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
AA
A
AAAA A
AAAA A
AAAA
A
AA
A
AA
A
AA
AAA
AAAA
AA
AA
AAA A
A
AA
AA
A
A
AAAA
AA
AA
AA
A
AA
A
AA
AA
A
AA
AAA A
A
AA
A
A
A
AA
AAA
A
AA
AA
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
AA
AAA
A
AAA
A
AA
AAA
AA
AAAA
A
A AA
AA
AA
AA
A
A
A
A
AA
A
A
A
AA
A
A
A
A
AA
A
A
A
AA
A
A
A
A
AA
A
A
A
A
A
A
A
A
A
A
AA
A
AAA
A
AA
AAAAA
A
A
A
A
AA
A
A
A AA A
A
A
AA
AA
AA
A
A
AA
AA
A
A
AA
AA
AA
AA
AA
AA
AA
A
A
A
AAA
A
A
AAA
AAAA
AA
A
A
A
AA
A
A
A
AA
AA
AAAAA
AA
AA
AA
A
A
AA
A
AA
A
AA
AA
A
A
A
A
A A
A
A
A
AA
A
AA
A
AA A
AA
AA
A
A
A
AAAAA
AA
AA
A
AA
AAA
AA
AA
A
AA
AAA
A
A
AA
A
AAA
A
A
A
A
A
AA
AAA
A
A
A
A
A
A
A
A
A
A
A
A
A
AA AA A A A
A
A
AA
A AAA A
A
A
A
A
20
30
40
50
60
Age (ye ars)
Figure 1: Mean PTA threshold values as a function of exposure group and stimulus
frequency.
234
NoiseChem 235
June 2004
100
A
Mean PTA threshold (db HL)
A
75
50
25
0
AA
A
A
A
A
A
A A
A A
A A
A
A
A A
A AA
A A A
A
AA
A
A
A
A
A
A
A
A
A
A
A
A A
A
A A A
AA
A
AA A
A A AAA
A
A A
A
A A
A
AA A AA
A
A A
AA
AA
A
A
A
A
A
A
A
A
A
AA
A
AA
A AA
AA A A
AAA AA A
A
A
AA A A
A
A
AA A
AA
A
A
AA A
AA
A
A
A
A
A
A
A
A
A A A AA AAA
AA
A
AAA
A AA
A
A
AA A A
A
A
A
AA A
A AA A
AA
AA
A
A AA
A
A
AA AA
AAAA AAA
A AA
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A A
AAA
A AA A
AA
A AA AA
A AAA A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
AA AA AA A
AAA A
A
AA A
AA A A A
AA
A
A
A
AA
A AA
A AA
AA
AA
AA
AA A AA
A
A
A
A
AAA
A AA
A
A
AA
A
AA AA
A AA A
A
A
AA
A
AA
AAA
A AA
AAA
A
AAA
A
A AA
A
A
A A
A
A A
A
AA
AA
AA
AA
AA
AA
AA
A
AA
A
A
A
A
A
A
A
A
AAA
A
A
A
A
A
A
A
A
A
A
AA
A
AA
A
A
A
AA
AA
A
AAA
AA
AAAA
AA
AA
AA
AA A A
A
A
A
A
A
A AA AA
A
AA
A
AA
A
AA AA
AA
A
AA
AA
AA
AAA
A
A A
AA
AA
AA
A
A
A
A
A
A
AAA A
A
AA
A
A A
AA
AA
A A
A A
AA A
AA A
A
A
0
10
20
30
40
50
Smoking (yea rs)
Figure 2: Mean PTA vs Smoking
PTA and Exposures
A one-between one-within ANOVA with factors exposure group and frequency revealed the
main effect of frequency (*F (2.18, 2885.9) = 485.6, p < .001; MSE = 260.2). Figure 3 shows
that the effect was caused by a general increase in threshold values with frequency. The peak
at 6 kHz was followed by a reversal of the trend at 8 kHz. A highly significant main effect of
exposure group (F (1, 1320) = 14.6, p < .001; MSE = 1180.8) could be ascribed to a
noticeable difference between the control and exposed conditions. A post-hoc test confirmed
this (Tukey’s HSD, p < .001) as well as showing a significant difference between the styrene
and styrene + noise groups (p = .011). Interestingly, the interaction between the factors was
also highly significant (F (16, 5280) = 10.5, p < .001; MSE = 142.2).
Figure 3: Combined PTA Thresholds for different exposures
PTA combined data
(Laboratories 1, 2, 3, 4 and 5)
Mean PTA threshold (dB HL)
40
30
EXPOSURE
styrene
20
noise
control
10
styrene + noise
solvent mix + noise
0
1
2
4
6
8
Frequency (kHz)
235
NoiseChem 236
June 2004
Combined data for all exposure groups across labs shows that styrene+noise had the worst
thresholds in comparison with all other groups. Styrene on its own, noise and solvent
mix+noise were very similar in their outcome in terms of the threshold shift. They all showed
a very similar pattern across frequencies.
Although the general tendency was the same for all exposure groups with respect to stimulus
frequency (flat curve at 1 and 2 kHz followed by an inverted U trend at higher frequencies),
exposure groups diverged at higher frequencies. This was especially true of the styrene +
noise group, whose threshold values were particularly high at 6 and 8 kHz.
PTA and Styrene
In order to examine the relationship between individual exposure conditions, styrene and
styrene plus noise groups were compared separately, as were noise and solvent mix plus noise
groups. Figure 4 illustrates the former. A mixed-design ANOVA confirmed that the
introduction of styrene into a noisy working environment considerably increases the damage
to the hearing system [the main effect of exposure; F (1, 479) = 9.2; p = .003; MSE = 1338.1].
Importantly, and despite a highly significant interaction [F (4, 1916) = 9.7, p < .001: MSE =
170.4], the effect of styrene appeared to be additive. In other words, the frequency-contingent
threshold function retained its shape under both types of exposure.
Upon consideration of the styrene with and without noise, it is clear from Figure 4 that
styrene in combination with noise worsens the threshold.
PTA combined data
50
Mean PTA threshold (dB HL)
40
30
20
EXPOSURE
10
styrene
0
styrene + noise
1
2
4
6
8
Frequency (kHz)
Figure 4: PTA Threshold comparison of styrene and styrene+noise groups
PTA and Solvent Mix
A somewhat different relationship was observed between noise and solvent mix plus noise
conditions. Specifically (see Figure 5), there main effect of exposure group was not
significant (p > .1) suggesting that in absolute terms there was no difference in threshold
values between the groups. However, a highly significant interaction [F (4, 3204) = 9.8, p <
.001; MSE = 109.5] confirmed that the groups produced substantially different threshold
functions. Specifically, the noise condition was associated with the increase in thresholds at
the higher end of frequency range (between 4 and 8 kHz), while the introduction of solvent
mixture resulted in a relative threshold increase at 1 and 2 kHz.
236
NoiseChem 237
June 2004
PTA combined data
Mean PTA threshold (dB HL)
30
20
EXPOSURE
noise
10
solvent mix + noise
1
2
4
6
8
Frequency (kHz)
Figure 5: PTA Threshold comparison between noise and solvents mix+noise
Prevalence of Hearing Loss by Lab
The prevalence of hearing loss for individual laboratories is given in Table 2. It can be seen
that samples in labs 1 and 6 show a noticeably high proportion of subjects with mean
threshold above 20 dB (over 50%). By contrast, the number of such subjects was relatively
small in lab 2.
Table 2: Prevalence by Labs
LAB
1
2
3
4
5
6
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Frequency
107
177
260
28
310
205
226
75
158
91
146
227
Percent
36.9
61.0
90.3
9.7
60.2
39.8
73.9
24.5
60.3
34.7
38.5
59.9
237
NoiseChem 238
June 2004
Prevalence of Hearing Loss by Exposure Group
The prevalence of abnormality with respect to exposure group is given in Table3
Table 3: Prevalence by Exposure Group
Exposure group
control
noise
styrene
styrene + noise
solvent mix
solvent mix + noise
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Normal
Abnormal
Frequency
239
64
283
291
216
128
87
62
59
27
323
231
Percent
78.4
21.0
48.6
50.0
62.4
37.0
58.0
41.3
61.5
28.1
57.6
41.2
The next step involved calculating the odds ratios for the four exposed conditions.
Table 4: Odd ratios
Exposure group
Odds ratio based on mean
threshold > 20 dB
Control
1
Noise
2.82
Styrene
2.56
Styrene + noise
3.09
Solvent mix + noise
2.74
Table 4: Odds ratios for hearing loss in the five exposure groups [PTA thresholds are
averaged over frequencies and based on the cut off point of 20 dB (< normal > loss)].
Prevalence vs. severity of loss
It can be seen that the highest risk of hearing loss is associated with styrene + noise and
solvent mix groups. While the former group’s association with a high risk of hearing loss is
corroborated by the ANOVA results, the solvent mix group could not be distinguished from
styrene and noise groups. This suggests that the hearing loss in the solvent mix group was
widespread yet less extensive in terms of threshold values when compared with the styrene +
noise group. Viewing the hearing loss effect as consisting of two relatively independent
factors (extent of damage and prevalence in the population) could represent a useful tool in
the analysis of agent-specific damage to the hearing system.
Transient Otoacoustic Emissions and Exposures
TOAE S/N ratio
An omnibus ANOVA was carried out on the TOAE data from labs 2, 3, 5 and 6 (Figure 6).
The solvent mix group was excluded from the analysis because of its small size (19). The
main effect of frequency was highly significant [*F (2.3, 1743.2) = 184.8, p < .001; MSE =
238
NoiseChem 239
June 2004
21.1] as was the main effect of exposure group [F (4, 764) = 33.5, p < .001; MSE = 117.4].
Finally, the significant interaction [F (12, 2292) = 1.8, p = .040; MSE = 16] mainly reflected
the deviant pattern in the solvent mix + noise group. The Transient emissions show
deterioration with frequency as expected. The worst amplitude of TOAE was observed for the
noise group followed by solvent mix + noise, styrene and then styrene+noise. It is interesting
to note that styrene is worse than in combination with noise whereas all other solvent mix
groups show greater loss with noise. A differential effect between PTA and TOAE is seen for
the styrene groups.
Figure 6: TOAE: Transient Otoacoustic Emission levels by Exposure
Omnibus TOAE S/N ratio
12
10
TOAE amplitude (dB SPL)
8
EXPOSURE
6
control
noise
4
styrene
2
styrene + noise
solvent mix + noise
0
1
2
3
4
Frequency (kHz)
239
NoiseChem 240
June 2004
Thresholds by Lab by Exposure
Figure 7: PTA Thresholds by Exposure for Lab1
PTA Lab 1
60
50
Estimated Marginal Means
40
EXPOSURE
30
styrene
20
noise
10
control
styrene + noise
0
1
2
4
6
8
Frequency (kHz)
As can be seen in Figure 7 for Lab 1, there were lab specific differences depending on the
particular exposure conditions and levels associated with each lab. For Lab1, the worst
thresholds were observed for styrene group and the controls also exhibited threshold shift
which was significant. This was largely explained on the basis of previous noise exposure.
These results indicate the difficulty of finding a suitable uncontaminated control group.
PTA Lab 2
60
50
Estimated Marginal Means
40
30
EXPOSURE
20
styrene
10
control
styrene + noise
0
1
2
4
6
8
Frequency (kHz)
Figure 8: PTA Thresholds by Exposure for Lab2
240
NoiseChem 241
June 2004
Controls vs. styrene PTA (composite) N (styrene) = 343, N (controls) = 266. From lab2 a
comparison of the styrene groups with controls shows a signficant difference between
controls and exposed groups at 4, 6 and 8kHz.. This lab again shows a difference compared to
the composite data in that styrene and styrene+noise show no signficant difference but
combined data shows a difference.
Figure 9: PTA Thresholds by exposure
PTA
30
Mean threshold (dB HL)
20
10
EXPOSURE
styrene
0
control
1
2
4
6
8
Frequency (kHz)
The overall comparison of thresholds of styrene exposed group with controls shows a
significant difference from 2kHz to 8kHz.
The main effect of frequency was highly significant (*F (2.16, 1314) = 193.9, p < .001; MSE
= 238.6), as was the main effect of exposure (F (1, 607) = 24, p < .001; MSE = 1034.4. The
interaction was also highly significant (F (4, 2428) = 12.7, p < .001; MSE = 129.15.
241
NoiseChem 242
June 2004
Key Findings from NoiseChem Study
Animal Studies:
• Histological data from rats show a risk of potentiation of noise induced hearing
loss by styrene even at concentrations as low as 400ppm (equivalent of 40ppm
for Humans). Current threshold limit values across Europe vary from 20 to
100ppm averaged over an 8-hour workday.
•
Aged rats are more vulnerable to noise than young rats, the opposite is true for
styrene, for which young rats are more vulnerable. The implication is that age
may be a significant factor in risk assessment for ototoxicity.
•
Styrene causes greater hair cell loss than noise whereas the pure tone threshold is
worse for noise than for styrene. The implication is that the mechanisms of
hearing loss due to styrene and noise are different.
•
Styrene causes chemical poisoning of outer hair cells whereas noise causes
mechanical damage. Styrene exposure results in disorganisation of membranous
structures of haircells whereas noise induced hearing loss can be due to
stereocilia pathology.
•
Distortion product emissions are as sensitive as evoked potential audiometry in
the evaluation of the ototoxic effects of styrene.
•
Central compensation for cochlea damage depends on the ototoxic agents.
•
Combined exposure to noise and toluene in rats shows potentiation of effect
when toluene exposure is close to the lowest observable adverse effect level
(LOAEL).
•
Addition of Carbon monoxide to the combined exposure of noise and toluene
reduces the level of toluene exposure necessary to observe potentiation. This
implies an increased risk in smokers.
•
Impulse noise causes greater hearing loss than wideband continuous noise in rats.
242
NoiseChem 243
June 2004
Human Studies:
•
Occupational exposure to styrene at levels below the lowest accepted threshold
limit value in the world of 20ppm is shown to affect the peripheral and central
auditory function.
•
Cochlear damage observed for a mean occupational exposure level of 13ppm
styrene.
•
Low level styrene exposure (mean 13pppm) shows subclinical balance
disturbance revealed as an increased sway when proprioceptive input is reduced
and visual clues are eliminated.
•
Styrene exposed workers have poorer thresholds than an age-matched
otologically unscreened population.
•
Compared to controls and those exposed only to noise, there is a) higher
prevalence of high frequency loss in groups exposed to styrene and combined
styrene and noise, b)have poorer mean thresholds at 2, 3 4 and 6kHz, and c)
lower Distortion Product emissions.
•
Odds ratio of hearing loss increase to 5.2 fold with styrene exposure and 13.1fold
with styrene and toluene combined.
•
Combined styrene and noise produces worse pure tone audiometric thresholds
particularly at 6 and 8 kHz compared to those exposed to styrene alone.
•
Central auditory effect in terms of a prolongation of the frequency-glide-induced
cortical response latency is observed in those exposed to styrene compared to
controls.
•
There is a potential risk of chemically induced hearing loss in workers not
exposed to noise.
•
Exposure to a mixture of solvents damages hearing and disturbs balance function.
•
Exposure to a mixture of solvents in the presence of noise worsens hearing in the
low frequency region (1-2kHz).
•
Exposure to a mixture of solvents affects peripheral and central auditory function.
•
Some aircraft maintenance workers exhibit changes in acoustic reflexes, auditory
brainstem responses, eye tracking and postural sway compared to controls.
•
Auditory brainstem responses and cognitive potentials were unaffected by styrene
exposure.
243
June 2004
NoiseChem 244
Exploitation and Dissemination of results
There is a clear need to disseminate the findings of the study to create awareness of the hazard
of industrial chemicals to both hearing and balance function in workers.
The setting of industrial exposure limits need to take into account the effects on hearing and
balance function and these need to be tested using the most sensitive means available. This is
currently not being done.
There is a clear need to bring researchers in the field of Audiology and toxicology together to
assess the effects of combined exposures which are in different domains such as chemical and
physical hazards. The effects observed in the work environment may also be useful material
for those concerned with the general environment where chemicals in the air and noise
interact, for example with traffic pollution. In this regard, the NoiseChem group have had
meetings with the index group of the EC Physical and Chemical hazards unit at Ispra, Italy
with a view to sharing information and have submitted a joint paper to an international
conference on the subject.
The results of the study will be circulated widely through peer-reviewed scientific journal
publications, and other educational material but there are no foreseeable patents of any results
envisaged. The impact of the research may include scientific achievements, exchange of data
and scientific information leading to improved measures of assessment and protection from
the effects of noise and chemicals. The goal has been to create harmonisation of approaches
and methodologies used in the different national programmes so that a more concerted effort
may be made against noise and chemicals and their detrimental effects. Clearly Governmental
action would be sought based on scientific data to establish a Pan-European programme to
reduce the effects of mixed chemicals and noise.
The scientific and technical progress has been reported at international conferences and will
be widely disseminated through publications of the proceedings, worker educational material
and electronic means. Major findings of the group, which have implications for policy or are
of general interest to the public, will be released for the press to cover in the newspapers. This
has a number of advantages in education and raising awareness among the general public of
the hazards of chemical exposure and may also bring important information to the attention of
the policy makers. Reporting of scientific progress has been the prime purpose of the
meetings but we will continue to work towards consensus development, improvements in
research protocols and an increase in understanding of the mechanisms of action of the
industrial chemicals.
It is important that in the exploitation plan there is a need to reform the hearing conservation
programmes as new evidence becomes available. Employers need to be persuaded to take
action on the new findings. For example there is a need to take in to account the risk to
244
June 2004
NoiseChem 245
hearing from chemical agents either acting alone or in combination with noise which pose a
threat to hearing even with hearing protection.
There is a need to establish whether the current audiometric surveillance in industry protects
the hearing of individual workers or is it merely monitoring the decline in hearing over the
years. Educational awareness needs to be created such that the workers’ deterioration in
hearing is noticed by the monitoring system before the worker is aware that there is any
damage.
It is important that steps are taken to set up multi-disciplinary forums where many
stakeholders may bring their views and concerns so that action may follow from such groups.
At present the industries where chemicals and noise are hazards are extremely reluctant to
participate in research studies for fear of litigation. Much work needs to be done to persuade
companies to take part in open research which will benefit their workers and their practices in
the long-term.
Meetings
1. Group Meeting, Nancy, France 28th April 2001
2. Group Meeting, Bordeaux, France 20-21 September 2001
3. European Solvents VOC Co-ordination Group, Brussels, Belgium 24 October 2001
4. Best Practices Workshop: Combined Effects of Chemicals and noise on Hearing,
Cincinnati, Ohio USA, Originally scheduled 4-5 October 2001, April 11-12, 2002
5. International Conference: NOPHER 2002-Nordic Noise II, 24-27 Oct 2002,
Stockholm, Sweden. Workshop on Noise and Chemicals
6. Best Practices Workshop: Combined Effects of Chemicals and Noise on Hearing, 1112 April, 2002, Cincinnati, USA
7. International Workshop: WHO/NOPHER Meeting: Combined Environmental
Pollutants and Health, 16-18 June 2003, Bonn, Germany
8. Mid-Term Review Meeting for NoiseChem Project, 12-13 December 2003, London,
UK
9. NoiseChem Group Meeting – Analysis of Results, 05-08 February 2004, Funchal,
Madeira
Publications
Morata, TC, Interaction between Noise and Asphyxiants: A Concern for Toxicology and
Occupational Health. Toxicological Sciences, 66:1-3, 2002.
Morata T.C.; Campo P. Ototoxic effects of styrene alone or in concert with other agents: A
review. Noise and Health, Jan - Mar 2002, vol. 4, no. 14, pp. 15-24(10).
Teixeira C.F.; Giraldo da Silva Augusto L.; Morata T.C. Occupational exposure to
insecticides and their effects on the auditory system. Noise and Health, Jan - Mar 2002, vol. 4,
no. 14, pp. 31-39(9)
Prasher D.; Morata T.C.; Campo P.; Fechter L.; Johnson A.C.; Lund S.P.; Pawlas K.; Starck
J.; Sliwinska Kowalska M.; Sulkowski W. NoiseChem: An European Commission research
project on the effects of exposure to noise and industrial chemicals on hearing and balance.
Noise and Health, Jan - Mar 2002, vol. 4, no. 14, pp. 41-48(8).
Morata T.C.; Little M.B. Suggested guidelines for studying the combined effects of
occupational exposure to noise and chemicals on hearing. Noise and Health, Jan - Mar 2002,
vol. 4, no. 14, pp. 73-87(15).
245
June 2004
NoiseChem 246
Morata T.C., Johnson A-C., Nylen P.R., Svensson E., Cheng J., Krieg E.F., Lindblad, A-C.,
Ernstg rd L., Franks J. Audiometric findings in workers exposed to low levels of styrene
and noise. Accepted for publication at the Journal of Occupational and Environmental
Medicine.
Morata, T.C., Lemasters, G. (2001). Considerações epidemiológicas para o estudo de perdas
auditivas ocupacionais. In: Nudelmann AA, Costa EA, Seligman J, Ibañez RN. PAIR - Perda
auditiva induzida pelo ruído. Vol.II. Livraria e Editora Revinter, Rio de Janeiro. pp.1-16.
Demange V, Chouanière D, Loquet G, Perrin P, Johnson A-C, Planeau V, Baudin V, Toamain
J-P, Morata T (2001). How to explore the otoneurotoxicity of solvents in the framework of
epidemiological studies in the work environment (in French), Les Notes Scientifiques et
Techniques de I.N.R.S. (Institut National de Recherche et de Sécurité), NS 0202: 1-35.
Henderson D, Morata TC, Hamernik R(2001). Considerations on assessing the risk of workrelated hearing loss. Noise and Health, 3(10): 63-75.
Morata, T (2001). Industrial Chemicals and Tinnitus. Tinnitus Today 26(4): 12.
Morata, TC, Campo P (2001). Auditory function after single or combined
exposure to styrene: a review. In Henderson D; Prasher D; Kopke R, Salvi R, and Hamernik
R. Noise-Induced Hearing Loss: Basic Mechanisms, Prevention and Control. NRN
Publications, London, 293-304.
Fischer, F.M., Morata, T.C., Latorre, M.R., Krieg, E.F., Fiorini, A.F., Colacioppo, S,
Gozzoli, L., Padrão, M.A., Zavariz, C., Lieber, R., Wallingford, K., Cesar, L.C. Effects of
environmental and organizational factors on the health of shiftworkers of a printing industry,
Journal of Occupational and Environmental Medicine, 43(10):882-889.
K. Pawlas: Neurotoxic exposure in Poland/Upper Silesia., Neurotoxnews 1(2001), 5
Witecki K., Pawlas K., Powązka E., Czapla A., Kidoń Z Osobowościowe uwarunkowania
homeostazy układu równowagi w aspekcie bezpieczeństwa pracy, IV Krajowa Konferencja
Polskiego Towarzystwa Medycyny Środowiskowej 19 – 20.06.2001, p 97 (in Polish)
Witecki K., Pawlas K., Czapla A., Kidoń Wyniki posturografii komputerowej w aspekcie
zmian organicznych ośrodkowego układu nerwowego ( Results of computer posturography
and oragnic chamges of central nervous system)Streszczenia VIII Sympozjum nt. Zagrożenia
Zdrowotne w Środowisku Pracy, Łódź 30 maja 1 czerwca 2001,p.146 ( in Polish)
Kidoń Z., Tkacz E., Czapla A., Pawlas K., Witecki K Frequency Analysis of Postural Sway
Segment Function. Procceed of VI Inter. Confernece SYMBIOSIS 2001. Szczyrk 1113.09.2001, pp 239-241
Prasher D., Morata T., Campo P., Fechter L., Johnson A-C., Lund SP., Pawlas K., Starck J.,
Sułkowski W., Śliwińska – Kowalska M Noisechem: An European Commission Reaserch
Project on the effetcs of exposure to noise and industrial chemiclas on hearing and balance.
Intern. J. Occup. Med.&Environ. Health 15.1.pp 5-11, 2002
K. Pawlas Stan Słuchu Dzieci W Zależności Od Stężenia Ołowiu We Krwi, ( Children’s
hearing in dependence on blood lead levels ) Materiały X Międzynarodowej Konferencji
„Uwarunkowania środowiskowe zdrowia dzieci” 1 – 2.06 2001 Legnica, 28-31 (Proceedings
of the Confernece – in Polish)
246
June 2004
NoiseChem 247
Pawlas K., Kmiecik-Małecka E., małecki A., PawlasN. Dependence of posture sway upon
blood lead cocnentration in Children Bok of Abstracts of VII Internationla Symposium on
Inorganic Biochemistry: Metals and Neurodegenerative Diseases, Wroclaw 20-23 September
2001 p 45
Pawlas K., Bronder A., Powązka E Czynniki kształtujące próg słuchu zdrowych dzieci w
wieku 4- 13 lat, Książka Streszczeń XI Sympozjum Audiologicznego Wrocław 2001, p.72
Bok of Abstracts of the Symposium – in Polish)
Powązka E., Pawlas K., Bronder A. Parametry słuchu w grupie zdrowych dzieci w wieku 4 –
14 lat wyznaczone za pomocą różnych metod badania, Książka Streszczeń XI Sympozjum
Audiologicznego Wrocław 2001, p, 45 Bok of Abstracts of the Symposium – in Polish)
Stefania Rzymełka, Krystyna Pawlas Hałas w szkole i związane z nim zagrożenia zdrowotne
(School noise and health effects ) Materiały V Konferencji Hałas- Profilaktyka - Zdrowie
2001, Kołobrzeg 8 – 11.12.2001, pp. 23- 30 (Proceedings of the Confernece – in Polish)
Krystyna Pawlas Środowisko akustyczne dzieci i młodzieży (Acoustical environment of
children and adolescents) Materiały V Konferencji Hałas- Profilaktyka - Zdrowie 2001,
Kołobrzeg 8 – 11.12. 2001, pp 19 – 22 (Proceedings of the Confernece – in Polish)
Morata T.C., Johnson A-C., Nylen P.R., Svensson E., Cheng J., Krieg E.F., Lindblad, A-C.,
Ernstgård L., Franks J. Audiometric findings in workers exposed to low levels of styrene and
noise. Accepted for publication at the Journal of Occupational and Environmental Medicine.
Baudin V, Toamain J-P, Morata T (2001). How to explore the otoneurotoxicity of solvents in
the framework of epidemiological studies in the work environment (in French), Les Notes
Scientifiques et Techniques de I.N.R.S. (Institut National de Recherche et de Sécurité), NS
0202: 1-35.
Nylén, P, Johnson, AC, Englund, A. Chemical and Physical Agent Interaction. In Bingham E,
Chorssen B, Powell CH (Eds). Patty’s Toxicology Fifth Edition, Vol 8, 669-698, Wiley &
Sons Inc. New York, 2001
Ilmari Pyykkö, Jukka Starck, Esko Toppila and Ann-Christin Johnson. Methodology and
Value of Databases. In Henderson D, Prasher D, Kopke R, Salvi R, Hamernik R (Eds). Noise
Induced hearing Loss: Basic Mechanisms, Prevention and Control.nRn Publications. London,
2001.
Campo P., Lataye R., Loquet G., Bonnet G. Styrene-induced hearing loss: a membrane insult.
Hearing Research (2001) 154 : 170-180.
Pouyatos, B.; Campo P.; Lataye R. Use of DPOAEs for assessing hearing losses caused by
styrene in the rat. Hearing Research (2002) 165 : 156- 164.
Robert L., Campo P., Pouyatos B., Cossec B., Morel. G. (April 2002) Solvent Ototoxicity in
Rats and Guinea pigs. Neurotox & Teratology. Submitted
Pierre Campo, Benoît Pouyatos, Robert Lataye, Georges Morel. Is the old ear more
susceptible to noise or styrene damage than the young ear? Hearing Research. Submitted
D. Prasher et al.: NoiseChem-An EC research project on the effects of exposure to noise and
industrial chemicals on hearing and balance. IJOMEH, vol. 15, No 1, 5-11, 2002
247
June 2004
NoiseChem 248
W.J.Sulkowski, S. Kowalska, W.Guzek, W. Wesolowski: Effects of occupational exposure to
a mixture of solvents on the inner ear – a field study, XXVIII International Congress of
Otoneurology, Alghero, Italy, 4-6 May 2001
W.J. Sulkowski, M.Sward-Matyja, W.Matyja: Audiobus-the first Polish audiological mobile
unit, NHCA/NIOSH Best Practices Workshop, Cincinnati, USA, 10-13.04.2002
Chen GD, Kong J, Reinhard K and Fechter LD. NMDA receptor blocker protects against
permanent NIHL but not its potentiation by CO. Hearing Research 2001, 154:108-115.
Tawackoli W, Chen GD and Fechter LD. Acute disruption of cochlear potentials by chemical
asphyxiants. Neurotoxicology and Teratology 2001, 23:157-165.
Rao DB, Moore D, Reinke LA and Fechter LD. Free radical generation in the cochlea during
combined exposure to noise and carbon monoxide: an electrophysiological and an EPR study.
Hearing Research 2001, 161:113-122.
Fechter LD, Johnson DL and Chen GD. Potentiation of noise induced hearing loss by low
concentrations of hydrogen cyanide in rats. Toxicological Sciences 2002, 66: 131-138.
Fechter LD, Johnson DL and Lynch RA. The relationship of particle size to olfactory nerve
uptake of a non-soluble form of manganese into brain. Neurotoxicology, in press.
Campo P., Lataye R., Loquet G., Bonnet G. Styrene-induced hearing loss: a membrane insult.
Hearing Research (2001) 154 : 170-180.
Pouyatos, B.; Campo P.; Lataye R. Use of DPOAEs for assessing hearing losses caused by
styrene in the rat. Hearing Research (2002) 165 : 156- 164.
Robert L., Campo P., Pouyatos B., Cossec B., Morel. G. (April 2002) Solvent Ototoxicity in
Rats and Guinea pigs. Neurotox & Teratology.
Campo P., Pouyatos B., Lataye R., Morel G., 2003. Is the old ear more susceptible to
noise or styrene damage than the young ear? Noise and Health 5, 1-19.
Pouyatos B., Morel G., Lambert-Xolin AM, Campo Pierre.. 2003 Consequences of noise- or
styrene-induced cochlear damages on glutamate decarboxylase levels in the rat inferior
colliculus. Hearing Research
Lataye R., Pouyatos B., Campo P., Morel G., 2003. Critical period for solvents in the rat.
Noise and Health.
Campo P., Blachère V., Pouyatos B., Lataye R. 2003. Comparison of predicted uptake and
elimination of toluene between rat and guinea pig blood. Neurotox & teratol.
Laurence D. Fechter Promotion of Noise-Induced Hearing Loss by Chemical Contaminants
Journal of Toxicology and Environmental Health, Part A, 67:727–740, 2004
D. Schwela, S. Kephalopoulos and D. Prasher. Air pollutants and other stressors as
confounding variables in noise epidemiological studies. Internoise 2004 Conference, Czech
Republic.
248
June 2004
NoiseChem 249
Policy Related Benefits
•
Currently there are no guidelines or standards for combined exposures. Current
national occupational exposure limit values for solvents do not account for possible
solvent ototoxicity. They should take into account the effects on hearing and balance
in setting the limits. Ototoxic properties and combined exposures must be taken into
consideration in preventive practices
•
Workers in chemical exposure areas need hearing protection. Noise is generally
present in the occupational environment where solvent exposures occur and damage
to hearing is attributed to the exposure to noise alone. The NoiseChem studies
indicate that solvent exposure alone can damage hearing and therefore protection
measures and threshold limit values need to take in to account the effects on hearing
and balance. This has implications for policy on hearing conservation. Hearing loss
prevention strategies should not be limited to noise exposures.
•
Air pollution due to road traffic by it very nature also produces noise and chemical
pollution. It is necessary to consider the joint effect of air and noise pollution.
•
The first steps need to be taken to raise awareness among occupational health
professionals that solvents pose a risk to hearing and balance and exposure to solvents
in the presence of noise increases the risk further.
•
Animal experiments which are critical in setting exposure limits should evaluate
damage to hearing using new sensitive measures rather than the currently used startle
reflex method which is a very crude indicator.
•
The NoiseChem studies indicate a need to examine further the interaction of physical
and chemical agents. The availability of more information should be used in the
development of guidelines on improved hearing loss prevention strategies, providing
exposed workers, investigators, and occupational health professionals with tools to
address the risk
•
Long-term aim must be to reduce the risk of work related hearing loss from all
causes.
249
June 2004
NoiseChem 250
Literature cited
ATSDR, Agency for Toxic Substances and Disease Registry (1990) Public Health Statement:Xylene.
Barregard, L., Axelsson, A. (1984) Is there an ototraumatic interaction between noise and solvents?
Scandinavian Audiology, 13, 151-155.
Bergstrom, B., Nystrom, B. (1986) Development of hearing loss during long-term exposure to
occupational noise, Scandinavian Audiology, 15, 227-234.
Biscaldi, G.P., Mingardi, M., Pollini, G., Moglia, A., Bossi, M.C. (1981) Acute toluene poisoning.
Electroneurophysiological and vestibular investigations, Toxicological European Research 3:6, 271273
Campo, P., Lataye, R., Cossec, B., Placidi, V. (1997) Toluene-Induced Hearing Loss: A MidFrequency Location of the Cochlear Lesions, Neurotoxicology and Teratology, 19:2, 129-140.
Campo, P., Lataye, Loquet, G., Bonnet, P. (2001) Styrene-induced hearing loss: a membrane insult,
Hearing Research, 154, 170-180
Campo, P., Loquet, G., Blachere, V., Roure, M. (1999) Toluene and Styrene Intoxication Route in the
Rat Cochlea, Neurotoxicology and Teratology, 21:4, 427-434.
Cappaert, N.L.M., Klis, S.F.,Baretta, A.B., Muijser, H., Smoorenburg, G.F. (2000) Ethyl Benzeneinduced ototoxicity in rats: a dose dependent mid-frequency hearing loss, Journal of the Association for
Research in Otolaryngology, 1:4, 292-299.
Cappaert, N.L.M., Klis, S.F.L., Muijser, H., Kulig, B.M., Ravensberg, L.C., Smoorenburg, G.F. (2002)
Differential susceptibility of rats and guinea pigs to the ototoxic effects of ethyl benzene,
Neurotoxicology and Teratology, 24, 503-510.
Cappaert, N.L.M., Klis, S.F.L., Muijser, H., de Groot, J.C.M.J., Kulig, B.M., Smoorenburg, G.F.
(1999) The ototoxic effects of ethyl benzene in rats, Hearing Research, 137:1-2, 91-102.
Chang, S-J.,Shih, T-S., Chou, T-C., Chen, C.J., Chang, H-Y., Sung, F-C. (2003) Hearing Loss in
Workers Exposed to Carbon Disulfide and Noise, Environmental Health Perspectives, 111:13, 16201624.
Clerici, W.J., Fetcher, L.D. (1991) Effects of chronic carbon disulfide inhalation on sensory and motor
functions in the rat, Neurotoxicology and Teratology, 13:3, 249-255.
Crofton, K.M., Lassiter, T.L., Rebert, C.S. (1994) Solvent-induced ototoxicity in rats: An atypical
selective mid-frequency hearing deficit, Hearing Research, 80, 25-30.
Crofton, K.M., Zhao, X. (1993) Mid-frequency hearing loss in rats following inhalation exposure to
trichloroethylene: evidence from reflex modification audiometry, Neurotoxicology and Teratology,
15:6, 413-423.
Crofton, K.M., Zhao, X. (1997) The ototoxicity of trichloroethylene: extrapolation and relevance of
high-concentration, short-duration animal exposure data, Fundamental and Applied Toxicology, 38,
101-106
Davis, R.R., Murphy, W.J., Snawder, J.E., Striley, C.A.F., Henderson, D., Khan, A., Krieg, E.F. (2002)
Susceptibility to the ototoxic properties of toluene is species specific, Hearing Research, 166, 24-32.
Fechter, L.D. (1993) Effects of acute styrene and simultaneous noise exposure on auditory function in
the guinea pig, Neurotoxicology and Teratology, 15:3 151-155
Fechter, L.D., Liu, Y., Herr, D.W., Crofton K.M. (1998) Trichloroethylene Ototoxicity: Evidence for a
Cochlear Origin, Toxicological Sciences, 42, 28-35.
Fornazzari, L., Wilkinson, D.A., Kapur, B.M., Carlen, P.L. (1983) Cerebellar, cortical and functional
impairment in toluene abusers, Acta Neurology Scandinavia 67:6, 319-329.
Hyden, D., Larsby, B., Andersson, H., Odkvist,L.M., Liedgren, S.R., Tham, R. (1983) Impairment of
visuo-vestibular interaction in humans exposed to toluene, ORL Journal for oto-rhino-laryngology and
its Related Specialities, 45:5, 262-269
Jacobsen, P., Hein, H.O., Suadicani, P.,Parving, A., Gyntelberg, F. (1993) Mixed solvent exposure and
hearing impairment: an epidemiological study of 3284 men. The Copenhagen male study, Occupational
Medicine, 43:4, 180-184.
250
June 2004
NoiseChem 251
Jaspers, R.M., Muijser, H., Lammers, J.H., Kulig, B.M. (1993) Mid-frequency hearing loss and
reduction of acoustic startle responding in rats following trichloroethylene exposure, Neurotoxicology
and Teratology, 15:6, 407-412.
Johnson, A-C. (1993) The ototoxic effect of toluene and the influence of noise, acetyl salicylic acid, or
genotype. A study in rats and mice, Scandinavian Audiology Supplement, 39,1-40.
Johnson, A-C., Canlon, B. (1994a) Toluene exposure affects the functional activity of the outer hair
cells, Hearing Research, 72, 189-196.
Johnson, A-C., Canlon, B. (1994b) Progressive hair cell loss induced by toluene exposure, Hearing
Research, 75, 201-208.
Johnson, A-C., Hagerman, B., Lindbald, A-C., Morata, T., Nylen, P., Svensson, E.B. (2002)
Epidemiological study of noise and styrene effects using a comprehensive audiological test battery,
Abstract, Nordic Noise, An International Symposium on Noise and Health.
Johnson, A-C., Juntunen, L., Borg, E., Hoglund, G. (1988) Effect of interaction between noise and
toluene on auditory function in the rat, Acta Otolaryngology, 105:1-2, 56-63.
Johnson, A-C. Nylen, P., Borg, E., Hoglund, G. (1990) Sequence of Exposure to Noise and Toluene
Can Determine Loss of Auditory Sensitivity in the Rat, Acta Otolaryngology, 109, 34-40.
Kilburn,K.H., (1999), Neurobehavioural and Respiratory Findings in Jet Engine Repair Workers: A
comparison of Exposed and Unexposed Volunteers, Environmental Research, 80, 244-252.
Kowalska, S., Sulkowski, W., Sinczuk-Walczak, H., (2000) Assessment of the hearing system in
workers chronically exposed to carbon disulfide and noise, Abstract, Medycyna Pracy, 51:2, 123-138.
Larsby, B., Tham, R., Erikksson, B., Odkvist, L.M. (1986) The effect of Toluene on the Vestibulo-andOpto-oculomotor System in Rats, Acta Otolaryngology,101, 422-428.
Larsby, B., Tham, R.,, Odkvist, L.M., Hyden, D., Bunnfors, I., Aschan, G. (1978) Exposure of rabbits
to styrene. Electronystagmographic finding correlated to the styrene levelin blood and cerebrospinal
fluid, 4:1,60-65.
Campo, P. (1997) Combined Effects of Simultaneous Exposure to Noise and Toluene on Hearing
Function, Neurotoxicology and Teratology, 19:5, 373-382.
Lataye, R., Campo, P., Barthelemy, C., Loquet, G., Bonnet, P. (2001) Cochlear pathology induced by
styrene, Neurotoxicology and Teratology, 23:1, 71-79.
Lataye, R., Campo, P., Loquet, G. (1999) Toluene Ototoxicity in Rats: Assessment of the Frequency of
Hearing Deficit by Electrocochleography, Neurotoxicology and Teratology, 21, 267-276.
Lataye, R., Campo, P., Loquet, G. (2000) Combined effects of noise and styrene exposure on hearing
function in the rat. Hearing Research, 139, 86-96.
Laukli, E., Hansen, P.W. (1995) An Audiometric Test Battery for the Evaluation of Occupational
Exposure to Industrial Solvents, Acta Otolaryngology, 115, 162-164.
Ledin, T., Jansson, E., Moller, C. Odkvist, L.M.(1991) Chronic Toxic Encephalopathy Investigated
Using Dynamic Posturography, American Journal of Otolaryngology, 12, 96-100
Ledin, T., Odkvist, L.M., Moller, C. (1989) Posturography findings in workers exposed to industrial
solvents, Acta Otolaryngology (Stockholm), 107, 357-361.
Loquet, G., Campo, P., Lataye, R. (1999) Comparison of Toluene-Induced and Styrene-Induced
Hearing Losses, Neurotoxicology and Teratology, 21:6, 689-697.
Loquet, G., Campo, P., Lataye, R. (2000) combined effects of an exposure to styrene and ethanol on
the auditory function in the rat, Hearing Research, 148: 173-180.
Lui, Y., Fetcher, L.D. (1997) Toluene disrupts outer hair cell morphometry and intracellular calcium
homeostasis in cochlear cells of guinea pigs, Toxicology and Applied Pharmacology, 142:2, 270-277.
McWilliams, M.L., Chen, G.D., Fetcher, L.D. (2000) Low-level toluene disrupts auditory function in
guinea pigs. Toxicology and Applied Pharmacology, 167:1, 18-29.
Moller, C., Odkvist, L., Larsby, B., Tham, R,. Ledin, T., Berghltz, L. (1990) Otoneurological findings
in workers exposed to styrene, Scandinavian Journal of Work and Environmental Health, 16:3, 189194
251
June 2004
NoiseChem 252
Moller, C., Odkvist, L., Thell, J., Larsby, B., Hyden, D, Berghltz, L., Tham, R. (1989) Otoneurological
findings in psycho-organic syndrome caused by industrial solvent exposure, Acta Otolaryngology,
107:1-2, 5-12
Morata, T.C. (1989) Study of the effects of simultaneous exposure to noise and carbon disulfide on
workers’ hearing, Scandinavian Audiology, 18, 53-58.
Morata, T.C. (1998) Assessing Occupational Hearing Loss: Beyond Noise, Scandinavian Audiology,
27, (supplement 48) 111-116.
Morata, T.C., Campo, P. (2002) Ototoxic effects of styrene alone or in concert with other agents: A
review, Noise and Health, 4:14, 15-24.
Morata, T.C., Engel, T., Durao, A., Costa, T.R.S., Krieg, E.F., Dunn, D.E., Lozano, M.A. (1997)
Hearing Loss from Combined Exposures among Petroleum Refinery Workers, Scandinavian
Audiology, 26, 141-149
Morioka, I., Kuroda, M., Miyashita, K., Takeda, S. (1999) Evaluation of Organic Solvent Ototoxicity
by the Upper Limit of Hearing, Archives of Environmental Health, 54:5, 341-346
Morioka, I., Miyai, N., Yamamoto, H.,Miyashita, K., (2000) Evaluation of Combined Effect of Organic
Solvents and Noise by the Upper Limit of Hearing, Industrial Health, 38, 252-257
Muijser, H., Lammers, J.H.C.M., Kulig, B.M., (2000) Effects of exposure to trichloroethylene and
noise on hearing in rats, Noise and Health, 6, 57-66.
Niklasson, M., Arlinger, S., Ledin, T., Moller, C., Odkvist, L., Flodin, U., Tham, R. (1998)
Audiological Disturbances Caused by Long-term Exposure to Industrial Solvents. Relation to the
Diagnosis of Toxic Encephalopathy, Scandinavian Audiology, 27, 131-136.
Niklasson, M., Moller, C., Odkvist, L. M., Ekberg K.,Flodin, U., Dige, N., Skoldestig, A. (1997) Are
deficits in the equilibrium system relevant to the clinical investigation of solvent-induced
neurotoxicity? Scandinavian Journal of Work and Environmental Health, 23:3, 206-213.
Niklasson, M., Tham, R., Larsby, B., Eriksson, B. (1993) Effects of toluene, styrene, trichloroethylene,
and trichloroethane on the vestibulo-and opto-oculo motor systems in rats, Neurotoxicology and
Teratology, 15:5, 327-324.
NIOSH, National Institute for Occupational Safety and Health (1987) Organic Solvent Neurotoxicity,
Current Intelligence Bulletin *-48
NIOSH, National Institute for Occupational Safety and Health (1996) IDLH documentation,
Trichloroethylene.
NIOSH, National Institute for Occupational Safety and Health (2002) International Chemical Safety
Cards:0073
NIOSH, National Institute for Occupational Safety and Health (2003) Federal Register 68:167, 5178651787
Nylen, P., Larsby B., Johnson, AC., Erikksson, B., Hoglund, B., Tham, R. (1991) Vestibularoculomotor, opto-oculomotor and visual function in the rat after long-term inhalation exposure to
toluene, Acta Otolaryngology, 111:1, 36-43.
Nylen, P., Hagman, M. (1994) Function of the auditory and visual systems, and of peripheral nerve, in
rats after long-term combined exposure to n-hexane and methylated benzene derivatives. II. Xylene,
Pharmacology and Toxicology, 74:2, 124-129.
Odkvist, L.M., Arlinger, S.D., Edling, C., Larsby, B., Bergholtz, L.M. (1987) Audiological and
vestibulo-oculomotor findings in workers exposed to solvents and jet fuel. Scandinavian Audiology,
16, 75-81.
Odkvist, L.M., Larsby, B., Frederickson, J.M.F., Liedgren, S.R.C., Tham, R. (1980) Vestibular and
oculomotor disturbances caused by industrial solvents, The Journal of Otolaryngology 9:1, 53-59.
Odkvist, L.M., Larsby, B., Tham, R., Ahlfelt, H., Andersson, B., Eriksson, B., Liedgren, S.R. (1982)
Vestibulo-oculomotor disturbances in humans exposed to styrene, Acta Otolaryngology, 94:5-6, 487493
Odkvist, L.M., Larsby, B., Tham, R., Hyden, D. (1983) Vestibulo- oculomotor disturbances caused by
industrial solvents, Otolaryngology – Head and Neck Surgery, 91, 537-539.
Odkvist, L.M., Moller, C., Thuomas, K-A. (1992) Otoneurologic disturbances caused by solvent
pollution, Otolaryngology – Head and Neck Surgery, 106, 687-692
252
June 2004
NoiseChem 253
Pollastrini, L., Abramo, A., Cristalli, G., Baretti, F., Greco, A. (1994) Abstract, Early signs of
occupational ototoxicity caused by inhalation of benzene derivative industrial solvents. Acta
Otorhinolaryngologica Italica, 14:5, 503-512.
Prasher, D., Morata, T., Campo, P., Fetcher, L., Johnson, A-C., Lund, S.P., Pawlas, K., Starck, J.,
Sliwinska-Kowalska, M., Sulkowski, W. (2002) NoiseChem: An European Commission research
project on the effects of exposure to noise and industrial chemicals on hearing and balance, Noise and
Health 4:14, 41-48.
Pryor,G.T., Dickenson, J., Howd, R.A., Rebert, C.S., (1983) Transient cognitive deficits and highfrequency hearing loss in weanling rats exposed to toluene , Neurobehavioural Toxicology and
Teratology 5:1, 53-57
Pryor,G.T., Dickenson, J., Feeney, E., Rebert, C.S., (1984a) Hearing loss in rats first exposed to
toluene as weanlings or as young adults, Neurobehavioural Toxicology and Teratology 6:2, 111-119
Pryor,G.T., Howd, R.A., (1986) Toluene-induced ototoxicity by subcutaneous administration,
Neurobehavioural Toxicology and Teratology 8:1, 103-104
Pryor,G.T., Rebert, C.S., Dickenson, J., Feeney, E.M.., (1984b) factors affecting toluene-induced
ototoxicity in rats, Neurobehavioural Toxicology and Teratology 6:3, 223-238
Pryor, G.T., Rebert, C.S., Howd, R.A., (1987) Hearing loss in rats caused by inhalation of mixed
xylenes and styrene, Journal of Applied Toxicology, 7:1, 55-61
Rebert, C.S., Becker, E. (1986) Effects of inhaled carbon disulfide on sensory-evoked potentials of
Long-Evans rats. Neurobehavioural Toxicology and Teratology, 8:5, 533-541.
Rebert, C.S., Day, V.L., Matteucci, M.J., Pryor, G.T. (1991) sensory-evoked potentials in rats
chronically exposed to trichloroethylene: predominant auditory dysfunction, Neurotoxicology and
Teratology, 13:1, 83-90.
Rebert, C.S., Sorenson, S.S., Howd, R.A., Pryor, G.T., (1983) Toluene-induced hearing loss in rats
evidenced by the brainstem auditory-evoked response, Neurobehavioural Toxicology and Teratology
5:1 59-62
Rebert, C.S., Sorenson, S.S., Pryor, G.T. (1986) Effects of intraperitoneal carbon disulfide on sensoryevoked potentials of Fischer-344 rats, Neurobehavioural Toxicology and Teratology, 8:5, 543-549.
Rybak, L.P. (1992) Hearing:The effects of chemicals, Otolaryngology-Head and Neck Surgery, 106,
677-686.
Savolainen, K., Linnavuo, M., (1979) Effects of m-xylene on human equilibrium measured with a
quantative method, Acta Pharmacology and Toxicology, 44:4, 315-318
Savolainen, K., Riihimaki, V., Vaheri, E., Linnoila, M. (1980) Effects of xylene and alcohol on
vestibular and visual functions in man, Scandinavian Journal of Work and Environmental Health, 6:2,
94-103.
Savolainen, K., Riihimaki, V., Luukkonen, R., Muona, O. (1985) Changes in the sense of balance
correlate with concentrations of m-xylene in venous blood, British Journal of Industrial Medicine,
42:11, 765-769
Sass-Kortsak, A.M., Corey, P.N., Robertson, J. Mc.D. (1995) An Investigation of the Association
between Exposure to Styrene and Hearing Loss, Annals of Epidemiology, 5:1, 15-24.
Schaper, M., Demmes, P., Zupanic, M., Blaszkewicz, M., Seeber, A., (2003) Occupational Toluene
Exposure and Auditory Function: Results from a Follow-up Study, Annals of Occupational Hygiene,
47:6, 493-502
Sliwinska-Kowalska,M., Bilski, B., Zamyslowska-Szmytke, E., Kotylo, P., Fiszer, M., Wesolowski,
W., Pawlaczyk-Luszczynnska, M., Kucharska, M., Dudarewicz, A. (2001a) Hearing impairment in the
plastics industry workers exposed to styrene and noise, Abstract , Medycyna Pracy, 52:5, 297-303
Sliwinska-Kowalska,M., Zamyslowska-Szmytke, E., Syzymczak, W., Kotylo, P., Bak, M.,
Wesolowski, W., Fiszer, M., Dudarewicz, A., Pawlaczyk-Luszczynnska, (2002) Epidemiological study
on the association between occupational exposure to organic solvent and hearing, Abstract, Nordic
Noise, An International Symposium on Noise and Health.
Sliwinska-Kowalska,M., Zamyslowska-Szmytke, E., Syzymczak, W., Kotylo, P., Fiszer, M.,
Dudarewicz, A., Wesolowski, W., Pawlaczyk-Luszczynnska, M., Stolarek, R., (2001b) Hearing loss
among workers exposed to moderate concentrations of solvents, Scandinavian Journal of Work and
Environmental Health, 27:5,335-342.
253
June 2004
NoiseChem 254
Sliwinska-Kowalska,M., Zamyslowska-Szmytke, E., Kotylo, P., Wesolowski, W., Dudarewicz, A.,
Fiszer, M., Pawlaczyk-Luszczynnska, M., Politanski. P., Kucharska, M., Bilski, B. (2000) Assessment
of hearing impairment in workers exposed to mixtures of organic solvents in the paint and lacquer
industry, Abstract, Medycyna Pracy 51:1, 1-10.
Smith,L.B., Bhattacharya, A., Lemasters, G., Succop, P., Puhala 2nd E., Medvedovic , M., Joyce, J.
(1997) Effect of chronic low-level exposure to jet fuel on postural balance of US Air Force personnel,
Journal of Occupational and Environmental Medicine, 39:7, 623-632.
Sulkowski, W.J., Kowalska, S., Matyja, W., Guzek, W., Wesolowski, W., Szymczak, W., Kostrzewski,
P. (2002) Effects of occupational exposure to a mixture of solvents on the inner ear. A field study,
Abstract, Nordic Noise, An International Symposium on Noise and Health.
Sulkowski, W.J., Kowalska, S., Sobczak, Z., Jozwiak, Z. (1992) The statokinesiometry in evaluation of
the balance system in persons with chronic carbon disulfide intoxication, Polish Journal of
Occupational Medicine and Environmental Health, 5:3, 265-276
Sullivan,M.J., Rarey, K.E., Connolly, R.B. (1988) Ototoxicity of toluene in rats, Neurotoxicology and
Teratology 10:6 525-530
Tham, R., Bunnfors, I., Eriksson, B., Larsby, B., Lingren, S., Odkvist, L.M. (1984) Vestibulo-ocular
disturbances in rats exposed to organic solvents. Acta Pharmacology and Toxicology, 54:1, 58-63.
Toppila, E., Nyman, P., Pyykko I., Starck, J., Oksa, P., Uitta, J. (2002) Effect of noise and styrene on
hearing, Abstract, Nordic Noise, An International Symposium on Noise and Health.
Ukai, H., Watanabe, T., Nakatsuka, H., Satoh, T., Liu, S-J., Qiao, X., Yin, H., Jin, C., Li, G-L., Ikeda,
M. (1993) Dose-Dependent Increase in Subjective Symptoms among Toluene-Exposed Workers,
Environmental Research, 60, 274-289.
WHO, World Health Organization Regional Office for Europe (2000), Chapter 5.12 Styrene. Air
Quality Guidelines - Second-Edition.
WHO, World Health Organization Regional Office for Europe (2000), Chapter 5.14 Toluene. Air
Quality Guidelines - Second-Edition.
Yano,B.L., Dittenber, D.A., Albee, R.R., Mattsson, J.L. (1992) Abnormal auditory brainstem responses
and cochlear pathology in rats induced by an exaggerated styrene exposure regime, Toxicology and
Pathology 20:1, 1-6
Yokoyama, K., Araki, S., Murata, K., Nishikitani, M., Nakaaki, K., Yokata, J., Ito, A., Sakata, E.
(1997) Postural Sway Frequency Analysis in Workers Exposed to n-Hexane, Xylene, and Toluene:
Assessment of Subclinical Cerebellar Dysfunction, Environmental Research, 74, 110-115.
254