Acetone

Endpoint summary

Description of key information

Skin irritation: no indication of irritating effect from non-guideline study (Anderson et al., 1986)
Eye irritation:
Low-volume eye test with application of 10 µL acetone, rabbit: mild irritative effect, fully reversible within 7 days (Maurer et al., 2001)
OECD Guideline 405, post-observation 7 days: mild irritative effect, partly reversible within 7 days (Märtins et al., 1992)
Eye irritation, Draize test, only 24 hr observation: severe changes with undiluted acetone; up to 30% acetone minimal irritating effects (Kennah et al., 1989)
Sensory irritation of the respiratory tract: nose irritation threshold  10,000 ppm or 25,000 mg/m3; NOAEC 5,000 ppm or 24,000 mg/m3

Key value for chemical safety assessment

Skin irritation / corrosion

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (not irritating)

Eye irritation

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (irritating)

Additional information

Skin irritation

There are no standard irritation studies with guideline conditions available. However, based on the available information (see below) and the experience with widespread use of acetone as solvent with possible skin contact, a further investigation of the skin irritation potential applying standard conditions does not seem to be scientifically justified.

The only reliable information comes from an alternative animal model for assessment of skin irritancy which can be used for a weight of evidence approach. In this test, a reduced volume of 10 µl acetone was applied to 1 cm² of shaved skin of female Dunkin Hartley guinea pigs for 3 days, 3 times daily. The presence of erythema and edema was assessed daily until sacrifice on post-treatment day 1. Punch biopsies were taken and histological sections were were evaluated microscopically for epidermal thickness and dermal inflammatory cells (counts of granulocytes, mononuclear cells, mast cells). Based on any endpoint of this test model, acetone showed no irritant action to the skin of guinea pigs whereas other substances as e.g. methylethyl ketone and trichlorethylene produced irritant responses and demonstrated the validity of the test system (Weight of evidence: Anderson et al., 1986). Under test conditions differing significantly from standard test protocols there was no indication of skin irritation.

Acetone was shown to be a skin defatting agent leading to reversible disturbation of the epidermal permeability barrier and to transepidermal water loss. Further reversible sequelae in the skin of mice or humans were local cellular changes, mainly in the stratum corneum and the stratum spinosum characterized by intracellular edematous lesions, reduction of protein biosynthesis and induction of a short-lasting proliferative response in epidermal cells (for details see IUCLID Section 7.9.3).

Eye irritation

There is no key study available which meets the standards of OECD Guideline 405, including both a test volume of 100 µL and a sufficient post-observation period of 21 days or up to full reversibility of effects. However, in a weight of evidence approach with the reliable studies it is possible to make conclusions on the classification for eye irritation.

The ocular irritation potential of acetone was investigated in rabbits with a test arrangement similar to OECD Guideline 405 except for instillation of a test volume of 10 µL (Low-volume eye test). Macroscopic eye injury was scored according to Draize (individual scores for lesions of the cornea, iris and conjunctivae, and maximum mean total score (MMTS)). Additionally, lesions of the cornea and the eyelids, epithelial and corneal thickness, depth of keratocyte injury, mean surface epithelial cell size, numbers of dead epithelial cells and dead keratocytes were quantified by different microscopic techniques. Time course of effects was followed for up to 35 days. In the standard eye irritation test, observed changes were consistent with a mild irritative effect on rabbit eyes being fully reversible within 7 days based on Draize scores. The maximum mean total score (MMTS) was 19.1 during the observation period beginning on day 1 (maximal possible value would be 110), and reached background on day 7. The microscopic methods revealed slight to mild reversible effects, with significant changes of parameters appearing at single timepoints of the 35-day observation period (Maurer et al., 2001). Based on this study and on the criteria of EU regulation 1272/2008 acetone would be classified into Category 2, irritating to eyes

In an eye irritation test, performed according to OECD Guideline 405 except for treatment of both eyes of two rabbits and with a post-observation limited to 7 days, acetone produced slight irritating effects based on grading scores according to Draize. Cornea opacity and redness of conjunctivae had maximum scores of 2 at 72 hrs which decreased to a maximum score of 1 on day 7. One eye of each animal showed conjunctival redness of grade 1 and one animal had corneal opacity of grade 1 in both eyes. Additionally, increases in corneal thickness, an indicator of eye irritation determined by ultrasonic corneal pachymetry, demonstrated considerable reversibility within 7 days post application (Märtins et al., 1992). Based on the observed reversibility of severity of effects within 4 days, full reversibility is expected to occur within further 14 days. Consequently, based on this study a classification into Category 2, irritating to eyes seems to be justified according to EU Directive 1272/2008.

The irritant potential of undiluted acetone and of dilutions (in polyethylene glycol) ranging from 1 to 30 % was investigated 24 hrs after instillation of 0.1 mL volumes into one eye of New Zealand rabbits (N = 4 - 6/group). Draize eye scores were compared with changes of corneal thickness obtained using a slit-lamp pachometer. Undiluted acetone was rated as severely irritating to rabbit eyes both based on a mean maximal total Draize score of 66 and on corneal swelling. At 30 % and lower concentrations, eye effects were minimal (Kennah et al., 1989).

Respiratory irritation

Human experience

Possible interactions between odour and respiratory irritation of acetone were investigated in well-designed studies from one working group. Odour and irritation thresholds were determined in age- and gender-matched groups (N=32) of previously unexposed subjects (controls) and in occupationally exposed workers by objective measures after a very short nose-only exposure by sniffing. Additionally, subjects were exposed to three different concentrations of acetone, determined by each individual's detection and lateralization thresholds and asked for subjective ratings of certain perceived sensations (Wysocki et al., 1997). Influences of occupational exposure and cognitive bias on subjective ratings of perceived odour and irritation, and on reported health symptoms were the subject of two further studies with a 20 min experimental exposure to acetone (800 ppm = 1900 mg/m³) in an exposure chamber. Immediately before and after exposure odour detection thresholds were measured. Test persons were also exposed to phenylethyl alcohol (PEA 200 ppm) as a non-irritant odorous control substance with the responses to PEA being treated as an irritant baseline. Intensities of perceived odour and irritation were rated every minute during exposure, and those of health symptoms immediately after termination of exposure applying a computerized line scale, the validated Labeled Magnitude Scale (LMS) (Dalton et al., 1997a, b). One study compared two age- and gender matched groups of acetone-exposed workers and controls without previous acetone exposure (N=27, 4 m, 23 f, age 23-65 yrs, 14 smokers). Workers had an exposure history with a median duration of 12 yrs to an average concentration of 479 ppm acetone (range of average 114-643 ppm) (Dalton et al., 1997).

Odour and irritation thresholds were determined by a two-alternative, forced-choice procedure applying a modified up-down-staircase method (binary dilution steps) with very short exposure by sniffing of odorant containing atmospheres via a nose-piece. Odour was detected by differentiation between blank and acetone atmospheres. Nasal irritation was assessed using the lateralization method by identification of the nostril exposed to acetone.

A comparison of odour detection thresholds (see Table) from the three studies shows that the range of the median thresholds is 41 to 84 ppm for previously unexposed subjects, and 86 to 855 ppm for acetone-exposed workers with high interindividual variances. Nasal irritation detection thresholds were only determined in a single study with median values of 15,758 and 36,669 ppm in unexposed subjects and exposed workers, respectively, and a lower interindividual variation.There was no significant effect of cognitive bias on odour detection thresholds of acetone.Adaptation due to occupational exposure was indicated both for the odour and the sensory irritation threshold of acetone by the higher values in workers compared to unexposed subjects (see Table). In subjects without previous acetone exposure adaptation already occurred during a 20 min experimental exposure to acetone as indicated by an increase of the odour detection thresholds (see Table; no data for irritation thresholds).Exposure to acetone did not affect sensitivity to butanol

Table: Overview on odour and nasal irritation thresholds in the different studies

Sensory quality	Study group	Detection threshold (ppm)				Reference
		Variable	Basal value	Value after 20 min experimental exposure	Comment
Odour	Unexposed control^a(N=27)	Median	84	278		Dalton et al. 1997
	Acetone-exposed workers (N=27)	Median	86	89
	Unexposed positive bias (N=30)	Median	136	124		Dalton et al. 1997
		Mean	264	521
	Unexposed neutral bias^a(N=30)	Median	84	278
	Unexposed negative bias (N=30)	Median	53	131
	Unexposed control (N=32)	Median	41	n.i.	interindividual variance: ratio 75th/25th percentile 26-fold	Wysocki et al. 1997
	Acetone-exposed workers (N=32)	Median	855	n.i.	interindividual variance: ratio 75th/25th percentile 10-fold
Nasal irritation	Unexposed control (N=32)	Median	15,758	n.i.	interindividual variance: ratio 75th/25th percentile 3-fold
	Acetone-exposed workers (N=32)	Median	36,669	n.i.

^agroups comprise same subjects; n.i. = not investigated

Adaptation is indicated not only from odour and irritation thresholds (see above) but also from the time course of perceived odour intensity in the studies of Dalton et al. (1997a, b). During the 20 min exposure period subjective ratings of odour decreased significantly by 46% in previously unexposed subjects (controls), and by 28% in workers, while ratings of perceived irritation decreased only in the controls. Similar effects were indicated in the study with the different bias groups (see below).

Dalton et al. (1997) analysed three groups of volunteers (N=30/group, age 23-65 yrs, no previous acetone exposure) with different cognitive bias towards the 20 min odorant exposure.The positive bias group (POS) was told to be exposed to natural extracts that were often used in aromatherapy und purported to have beneficial effects on mood and health. The negative bias group (NEG) was told to be exposed to industrial solvents that purportedly caused health effects and cognitive problems following long-term exposures, whereas the neutral bias group (NEU) was told to be exposed to standard odorants being commonly used and approved for olfactory research. While odour intensity ratings(very strong to strong)were similar for all groups during the first 10 min,odour intensities varied among groups with POS < NEU < NEG (all in the range of strong)in the second half of the exposure.For perceived irritation significantdifferences were found (POS < NEG < NEU; rating strong to moderate) during the first 10 min of exposure, while ratings were similar for the negative and neutral bias group and distinctly lower for the positive bias group (all values in the range of moderate) during the second 10 min of exposure(Dalton et al., 1997).

Questionnaires on irritation qualities after short sniffing exposures yielded an evaluation of a type of irritation that is different from sensory nasal irritation measured with the lateralization method. It seemed, that concentrations of acetone well above the detection threshold but below the irritation threshold could be "irritating" or annoying to some subjects simply because they do not recognize the odour and erroneously attribute a toxicity to the compound. Generally, acetone-exposed workers tended to treat the acetone stimulus simply as non-irritating odour, whereas the unexposed subjects ascribed irritating qualities to the stimuli. In conclusion, subjective, categorical, qualitative reports of intranasal irritation do not yield results that are comparable with those obtained from quantitative, objective measures (Wysocki et al., 1997).

The subjective assessment of seven regional irritation symptoms, typical for solvent exposure, of 14 somatic or control symptoms, and of the current perceived irritation, following a 20 min exposure applying the LMS, showed that there was a significant higher frequency and rating of all types of symptoms in previously unexposed control subjects than in workers. Concerning solvent-associated symptoms, controls showed significantly higher ratings for throat irritation, nasal irritation, lightheadedness, drowsiness (all rated as weak), headache, nausea and, and for current irritation.(Dalton et al., 1997). The type of reported solvent-associated symptoms differed significantly as a function of bias condition. Mean ratings for throat irritation, nasal irritation, lightheadedness, nausea and drowsiness as well perceived current irritation and somatic control symptom were significantly less intense in the positive bias group compared to the other groups (Dalton et al., 1997). Prinicipally, in both studies the majority of ratings was in the range of barely detectable.

In conclusion, odour was the most significant factor mediating the degree of perceived irritation from acetone both for subjects without or with previous acetone exposure. Occupational exposure to acetone appeared to produce a very specific change in the perceived intensities of irritation and odour but apparently did not induce a general olfactory dysfunction. No correlations were found for individual exposure histories (exposure levels, cumulative exposure of workers). Responses to the non-irritant control substance PEA suggested that the ratings of irritation are not pure measures of irritation independent of odour (Dalton et al., 1997).

The perceived sensory attributes of acetone, i.e. subjective ratings of odour and irritation, can be modulated by a cognitive factor such as information about the consequences of long-term exposure to the chemical. Subjects given a positive characterization perceived significantly less odour and irritation during exposure than subjects given a negative or neutral characterization. However, no influence of bias on the odour detection threshold as an objective measure was observed. Consequently, in published studies, much of the extreme variability in reported irritation and health symptoms from exposure to acetone may stem not from differences in sensitivity, but from differences in cognitive expectations concerning the exposure, which may differ between naive and occupationally exposed individuals. In the absence of any characterizing information (cognitive bias) subjects are guided by a primary quality such as odour intensity to provide an index of irritancy. The correlation between odour and irritancy was higher in the neutral than in the other two groups. There was no influence of demographic factors, as age, gender, or smoking habit, on any variable (Dalton et al., 1997).

A critical review (Arts et al., 2002) on odour detection and irritative effects of acetone in humans, published until 2002, compared studies using subjective (neuro)behavioural methods (questionnaires) and studies using objective measurements to detect odour and irritation thresholds. This review revealed large differences in reported acetone concentrations found to be irritative to the respiratory tract and the eyes.

With respect to odour, the variation in odour threshold levels (total range of 0.5 to 11,600 ppm) was assumed to be due to differences in measuring or calculating exposure concentrations and, to a lesser degree, in test methodology or in characteristics of the test population. Levels in the range of 20 to 400 ppm (ca. 50 to 1000 mg/m³) were evaluated to be scientifically valid considering experimental protocols and measurement data (based on studies of Dalton et al., 1997a and b; Wysocki et al., 1997).

With respect to sensory irritation, measured as nasal pungency (data of Wysocki et al., 1997) levels between 10,000 to 40,000 ppm (ca. 25,000 – 100,000 mg/m³) were reported for subjects without and with occupational exposure. Consequently, the threshold for sensory irritation is about 100 to 400 times higher than the odor threshold levels in subjects without previous acetone exposure, and about 40 to 80 times higher than those in acetone-exposed workers. This tendency is confirmed by data of Commetto-Muniz et al.(1993) showing a 10fold higher nasal irritation threshold level in anosmic subjects compared to the odour detection level in normosmic subjects, and also by data on sensory eye irritation thresholds (Commetto-Muniz et al.,1995).

In contrast, the evaluation of irritant effects in the available neurobehavioral studies is based on questionnaires to establish occurrence and/or intensity of subjective health symptoms in relation to acetone exposure. Here, subjective symptoms such as eye, nose, or throat irritation are reported at exposure levels below 1,000 ppm (e.g. studies of Satoh et al., 1996; Kiesswetter et al., 1994; Seeber et al., 1997; Dalton et al., 1997a and b; Wysocki et al., 1997). However, the investigations of Dalton et al. (1997a, b) and Wysocki et al. (1997), which are presented above, call for caution in the interpretation of effects reported in other studies (e.g. Matshushita et al., 1969a, b; Seeber et al., 1992; Kiesswetter et al., 1994) without adequate characterization and control of cognitions and odour as factors biasing sensory irritation. In summary, irritation thresholds of acetone based on subjective ratings, are not likely to be a reflection of true sensory irritation, and, additionally, are not acceptable for establishing the irritation threshold of acetone.

Based on the available valid data, the LOAEC for sensory irritation of the respiratory tract, indicated by the nose irritation threshold, is set at 10,000 ppm acetone or 25,000 mg/m³ for subjects without occupational exposure and at 40,000 ppm or 100,000 mg/m³ for acetone-exposed workers. Based on the study design with binary dilution steps the corresponding NOAECs are 20,000 ppm or 50,000 mg/m³.

Effects on eye irritation: irritating

Justification for classification or non-classification

Skin irritation: No indication of skin irritation in non-guideline studies, defatting properties were shown. Based on the criteria of EU regulation 1272/2008 there is no classification of acetone for skin irritation.

Labelling according to DSD: R66, Repeated exposure may cause skin dryness or cracking

Eye irritation: Based on weight of evidence approach acetone shows mild reversible irritative effects.

Based on the criteria of EU regulation 1272/2008 acetone is be classified into Category 2, irritating to eyes.

Labelling according to DSD: Xi, R36, Irritating to eyes

Labelling according to CLP: H319 Respiratory tract irritation: Based on the available data classification for respiratory tract irritation is not warranted.