Toluene

Specific investigations: other studies

Description of key information

Toluene is ototoxic in rats producing loss of cochlea hair cells which is considered to be an irreversible effect. Hearing loss has been reported in humans, especially when toluene exposure is associated with high exposure concentrations and a noisy environment.  In addition there have been reports of disturbances of colour vision.  For both these effects epidemiology studies (Schaper et al, 2003; 2004; 2008) have demonstrated that adverse changes do not occur when exposures are maintained below the current indicative occupational exposure limit of 50 ppm (192 mg/m3).

Additional information

Ototoxicity and colour vision

Non-human information

Effects on hearing

Toluene exposure causes a functional hearing deficit in rats, which can be measured as a lack of behavioural response to sound, by electrophysiological changes, and by morphological damage to the outer hair cells of the inner ear. In what is considered to be the key study the relationship between toluene concentration, exposure pattern, and hearing loss in rats was investigated in a series of experiments in male Fischer rats (Pryor et al, 1984). Hearing loss was measured using behavioural and electrophysiological methods. Two weeks of exposure to 1000 ppm (3768 mg/m³) toluene 14 h/day caused hearing loss. Lower concentrations (400 (1507 mg/m³) and 700 ppm (2638 mg/m³)) were without effect even after 16 weeks of exposure. Three-day exposures to 1500 ppm (5653 mg/m³) for 14 h/day or to 2000 ppm (7537 mg/m³) for 8 h/day were ototoxic. Single exposures to 4000 ppm (15074 mg/m³) for 4 h or to 2000 ppm for 8 h were without effect. Intermittent exposure to 3000 ppm (11306 mg/m³) for 30 minutes every hour for 8 h/day caused hearing loss within 2 weeks, but a similar exposure schedule for 4 h/day was ineffective even after 9 weeks. Similar effects on auditory function were seen in the supporting study by Brandt-Lassen et al (2000). Male rats exposed 6 h/day for 10 days at 1500 or 2000 ppm toluene developed mid-frequency hearing loss, whereas rats exposed to 500 or 1000 ppm did not.

For comparative purposes, the sub-chronic NOAEC for systemic toxicity of toluene in the rat is 625 ppm (2,344 mg/m³) and the chronic NOAEC 300 ppm (1,125 mg/m³).

Sub-acute oral administration of toluene (8.47 mmol/kg bw/d; 780 mg/kg bw/d) to male rats was reported to lead to a reduction in numbers of outer hair cells present in the three rows of the cochlea, with greatest losses in the third row and the lowest losses in the first row (Gagnaire and Langlais, 2005). The findings indicate that toluene is ototoxic after sub-acute, high level oral exposure, however limitations in study design (specifically the use of a single treatment level and omission of a control group) limit the value of the data for risk assessment purposes.

Rats exposed to toluene up to 500 ppm plus noise did not show any increase in hearing impairment compared to the rats exposed to noise only, with no synergistic or additive interactions present (Lund and Kristiansen, 2008).

Auditory dysfunction also occurs in guinea pigs although the effect appears to be transient in this species (McWilliam et al, 2000). Effects were assessed by cubic distortion product otoacoustic emission (CDP) and histology of hair cells in guinea pigs exposed to 250, 500 or 1000 ppm toluene for 8 h/day, 5 day/week for 1 and 4 weeks. All concentrations of toluene were able to temporarily disrupt auditory function acutely in the guinea pig, and 500 and 1000 ppm toluene produced greater acute dysfunction. Although the auditory dysfunction progressed between 1 and 4 weeks of exposure, recovery was seen in all animals within 3 days. A permanent loss did not develop and hair cell death was not seen. 

The EU RAR (2003) concluded that toluene produced permanent hearing loss in the rat following exposure to concentrations of 1,000-1,400 ppm (3,800-5,320 mg/m³) for 2-8 weeks, with risk characterisation based on a NOAEC of 700 ppm (2,660 mg/m³) reported for animals exposed 14 hours/day for up to 16 weeks. The RAR also noted that results of studies conducted in the guinea showed transient auditory system impairment in the guinea pig at a toluene concentration of 250 ppm (938 mg/m³), however the effects were not permanent and no hair cell death was seen in the cochlear.

Effects on colour vision

No animal data were located, with findings in human populations discussed below.

Human information

Effects on hearing

The available human data were reviewed in the EU RAR (2003). New data are available in what is considered to be the key study (Schaper et al, 2003). The ototoxicity of occupational exposure to toluene below 50 ppm (mean exposure 26 ppm / 98 mg/m³) was investigated in a longitudinal study over 5 years with four repeated examinations starting with 333 male workers from rotogravure printing plants. The auditory thresholds were measured with pure tone audiometry. Repeated measurement analyses (grouping factors: toluene intensity, exposure duration and noise intensity) and logistic regressions did not reveal significant effects of toluene intensity, of exposure duration and of interactions between toluene intensity and noise intensity. In a follow-up study (Schaper et al., 2008), the effects of occupational exposure to toluene plus noise was assessed 192 employees from the Schaper et al (2003) study during the period 1996-2001. Recent individual exposures were measured 10 times during the study and past lifetime weighted average exposures (LWAE) to toluene and noise determined from individual work histories. Mean LWAE exposures: 45 ± 17 ppm toluene plus 82 ± 7 dB(A) noise for printers (high toluene intensity); 10 ± 7 ppm toluene plus 82 ± 4 dB(A) noise for end-processors (low toluene intensity). Mean current exposures: 26 ± 20 ppm toluene plus 81 ± 4 dB(A) noise for printers; 3 ± 3 ppm toluene plus 82 ± 4 dB(A) noise for end-processors. There were no significant effects of toluene intensity, of exposure duration or of interactions between toluene intensity and noise intensity. Based on current exposure 26 ppm (93 mg/m3) was the human NOAEC for developing hearing loss as a result of occupational toluene exposure.

Regarding effects of toluene on hearing in humans, the EU RAR (2003) noted that studies indicating that occupational exposure to toluene at high concentrations and noise may increase the risk of developing mild high frequency hearing loss were not appropriate for determining a LOAEC / NOAEC, and that other human studies involving lower levels of toluene exposure had found no association between toluene exposure intensity and age-corrected auditory thresholds.

Effects on colour vision

The strength of association between occupational exposure to organic solvents and impaired colour discrimination was assessed by Lomax et al. (2004) in a review of nine cross-sectional or longitudinal studies involving print workers and other workers exposed mainly to toluene. Exposure-response relationships and reversibility of effect were addressed, including evidence of acute effects of contemporary toluene exposure versus evidence of persistent effects of past exposures. It was concluded that there is good evidence that toluene has no acute effects on colour discrimination for exposures in the region of 300-350 ppm for 30 minutes and 50-150 ppm 8h TWA. It was also concluded that all of the studies investigating whether or not long-term repeated exposure to toluene can cause a persistent impairment of colour discrimination were subject to limitations such that the evidence from the studies was inconclusive. The authors considered that the weight of evidence from the studies tends to suggest that toluene does not cause a persistent impairment, but reassurance of the absence of an effect was lacking.

Information from several human occupational studies investigating possible effects of toluene exposure on colour vision (e.g. Zavalic et al., 1998 a,b,c; Muttray et al., 1999; Cavalleri et al., 2000; Campagna et al., 2001; Paramei et al., 2004) is included in the registration dossier. The investigations were of cross sectional design and compared pre-shift colour vision among groups of toluene-exposed workers with unexposed workers. Colour vision was assessed by the Lanthony D-15 desaturated panel, and colour vision loss was quantitatively established by the Colour Confusion Index (CCI) and classified by type of acquired dyschromatopsia according to Verriest's classification. However while results from these studies are widely cited as supporting an association between toluene exposure and impaired colour vision, aspects of their design and reporting have the potential to confound interpretation (see below) and they have been assigned Klimisch Rel. 4 (not assignable) as a result.

In studies of colour discrimination reported by Zavalic and co-workers, it is not clear whether there was overlap between the study group of male rotogravure printers included in Zavalic et al. (1998a) and the largely male group of rotogravure printers assessed by Zavalic et al. (1998b,c); and also whether there was overlap between the group of male electronics workers which form the comparison group in the Zavalic et al. (1998a) study and the similar sized group of male electronics workers included in the comparison group of Zavalic et al. (1998b,c). There are also discrepancies between the two reports of one study (Zavalic et al., 1998b,c): median blood toluene in the shoe-workers group was reported as being 0.01 mg/L in one report and 0.01 mg/g in the other, while in the printers median blood toluene was reported to be 0.042 mg/L and 0.0042 mg/g.

The study by Muttray et al. (1999) was principally an investigation into the acute effects of airborne toluene concentrations in the range 290-360 ppm experienced for 30-40 minutes while clearing printing presses, however the investigation also contained a study of the chronic effects of toluene exposure which used a different comparison group and the measurements of the exposed group that had been performed some months previous.

Exposure to toluene in the study by Cavalleri et al. (2000) was monitored by quantitation of urinary toluene, which is not recognised as an index of toluene exposure and the cumulative exposure indices, which were based on a single sample, may not have accurately reflected past exposures. The CCI scores from this study were also extremely variable.

Campagna et al. (2001) reported an association between toluene exposure and impaired colour discrimination, however the pattern of results obtained raised the possibility that five of the toluene-exposed workers may have had an undiagnosed congenital colour vision deficiency, while a significant difference in colour discrimination between unexposed subjects and the low exposure group (exposure level = 8.5 ppm) did not appear plausible.

Paramei et al. (2004) performed an assessment of colour vision impairments due to occupational exposure to organic solvents (including 11 studies for toluene) by means of a meta-analysis. The approach permitted quantification of the trends covered by the literature by estimating effect sizes for single studies. The analysis yielded a positive effect size indicating an inferior performance of the exposed subjects, although the value did not reach significance at the 5% level. However, this analysis was based on incomplete use of all of the available information (i.e. studies that did not report CCI as mean and standard deviation were disregarded) while there were clear errors made for two of the four effect sizes included in the meta analysis (i.e. standard deviations from Zavalic et al. (1998a) were 10 times higher than reported by the study authors, while an effect calculated for the rotogravure printers in the study by Muttray et al (1999) did not use the results for the exposed workers and comparison subjects from the “chronic” component of the study).

Overall, these cross-sectional studies and a meta-analysis provide weak evidence of effects on colour discrimination below 50 ppm, but some evidence of effects among the two groups of printing workers exposed to median levels of 120 ppm and 132 ppm. No reliable LOAEC/NOAEC can be derived from these studies.

Potential effects of human occupational exposures to toluene on colour vision were investigated in a follow-up study over 4 years with three repeated examinations (Schaper et al, 2004) with good characterisation of current and long-term exposure. Colour vision was measured with the Lanthony desaturated colour vision test D-15d, and the colour confusion index (CCI) was calculated. The mean exposure durations were 23 ± 6 years for "long-time exposed" and 7± 2 years for "short-time exposed" subjects. Repeated analyses (n = 162) and multiple regressions (maximum available n = 267) did not reveal significant effects of toluene on colour vision with respect to intensity or duration of current or long-term exposure. The NOAEC from this study was 26 ppm (98 mg/m³).

It can be concluded based on current exposure that 26 ppm (98 mg/m³) is a NOAEC for developing a hearing loss or effects on colour vision as a result of occupational toluene exposure and this value will be taken forward in the risk characterisation.