Registration Dossier

Administrative data

Description of key information

Guideline, GLP skin and eye irritation studies are available for furfural. Overall the data available for skin irritation suggest furfural is a mild irritant to the skin, sufficient to classify the material as a skin irritant. Furfural is also an eye irritant and irritation of the respiratory system of rats, rabbits and hamsters was seen after inhalation exposure; based on these observations furfural is considered a respiratory irritant; a clear NOAEC for local effects after repeated exposure was established at 8 mg/m3.
Respiratory irritation after acute exposure and as relating to the potential for short-term (15 minute) exposure to peak concentrations, it is considered that the weight of evidence indicates a value of 20 mg/m3 will be protective. The rationale is based on a number of factors: in acute inhalation studies no evidence of adverse respiratory symptoms were seen at an exposure concentration of 540 mg/m3 (Merkel, 2003); adverse respiratory symptoms were only seen at concentrations of 80 mg/m3 and above following repeated inhalation exposures (Arts et al, 2004); histopathological effects following 28 daily exposures (Arts et al, 2004; Staal, 2008) are equivocal and considered adaptive in nature at 20 mg/m3. Taken together with the human data reported above that indicates no significant irritancy at 20 mg/m3 these data support this dose level as a realistic NOAEC for local effects following short term exposures.

Key value for chemical safety assessment

Skin irritation / corrosion

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (irritating)

Eye irritation

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (irritating)

Respiratory irritation

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (irritating)

Additional information

Skin irritation

Animal data

A GLP complaint guideline rabbit skin irritation study on furfural was reported by Joseph in 2003(b). Furfural (99.62% purity) was applied to the clipped flanks of three rabbits, a semi- occlusive dressing was applied for 4 hours and then the flanks were assessed for signs of skin irritation for up to 14 days. Erythema was observed in all 3 rabbits until day 7 and oedema was observed in 2 out of 3 animals until 72 hours. The mean score for erythema was 2, and that for oedema was 0.67. All signs of irritation fully resolved within 14 days.

These data are supported by a limited study reported by Woods and Seevers (1955, cited in EU-RAR 2008), where 45-1000 mg/kg bw undiluted furfural was applied to the shaven non-abraded skin of rabbits and allowed to remain in place for 48 hours (occlusive conditions). After another 48 hours, any cutaneous reactions were noted and mild local irritation was observed in rabbits receiving 45-500 mg/kg. No data are available on the extent (e.g., scores) and/or the reversibility of effects.

In another publication (Cocker et al, 1992) with limited detail, furfural was applied to the skin of guinea pigs for 3 days, 4 hours/day. Undiluted furfural was reported to cause intensive but reversible skin irritation. A very mild reaction was produced by application of 5% whereas applications of 1% furfural did not produce any signs of irritation. The same authors report that application of undiluted furfural to intact shaved skin for 4 hours on 20 successive days resulted in hyperplasia, hyperkeratosis and exfoliation of the epidermis. Similar but less severe effects were observed after application of 5 and 1% furfural

Human data

During regular medical examinations of employees at the Jugotanin plant in Slovenia (Medicinski Center, 1989), none of the employees had any skin changes which were considered to have been a consequence of daily contact with furfural.

Eye irritation

Animal data

A GLP complaint guideline rabbit eye irritation study on furfural was reported by Joseph in 2003(c). 0.1mL undiluted furfural (99.62% purity) was applied to one eye of 3 rabbits, the other eye acting at control. Treated and control eyes were washed with distilled water 24 hours post treatment. Evidence of corneal opacity was recorded (mean score = 0.9); in 2 of 3 rabbits, the corneal opacity had completely reversed by the 72 h post-instillation, whereas one rabbit continued to exhibit corneal opacity until Day 14. The mean score for redness was 2.1 and that for chemosis was 1.56. There were no signs of iritis. In summary, furfural is irritating to rabbit eyes.

These data are supported by a limited reported study by Woods and Seevers (1955, cited in EU-RAR 2008); undiluted furfural was instilled in one eye of 15 male adult white rabbits, the eye was gently massaged for up to 40 seconds and the untreated eye acted as control. No individual scores are reported but the authors report slight oedema of the conjunctiva was observed after application of 0.001-0.002 mL; 0.04mL caused marked irritation for 5 days and application of 0.09-1 ml furfural resulted in eyelid spasm for 7 days with gross corneal opacity. The eyes appeared normal at day 9.

Human data

Overall, no substantive information is available. Data on the irritating properties of furfural vapour in humans are reported by ACGIH (2001). A study by the NIOSH in a grinding wheel plant revealed a general higher incidence of eye and respiratory tract irritation which was attributed to furfural vapour concentrations of 20-63 mg/m3. 11 of 15 workers reported symptoms of eye irritation which included itching, burning, tearing and/or redness. However, the authors state the influence of other hazardous substances and factors cannot be excluded, and are unknown; therefore, these data cannot be considered for risk assessment purposes.

Respiratory irritation

Animal data

The sensory irritation potential of furfural was investigated in two strains of mice by Steinhagen and Barrow, 1984. Sensory irritation was quantified by measuring respiratory rate depression during 10 minutes inhalation exposures, followed by a 5 minute recovery period and the RD50 value was defined as the concentration eliciting a 50% decrease in respiration rate. The authors reported exposure of furfural vapours in both strains of mice caused a rapid decrease in respiration rate at the onset of exposure which was sustained throughout the exposure period. RD50 values for reduced respiration rate were reported as 234 ppm (95% confidence limits: 174-327 ppm) equivalent to 920 mg/m3 (95% confidence limits: 684-1285 mg/m3) and 287 ppm (95% confidence limits: 216-402 ppm) equivalent to 1128 mg/m3 (849-1580 mg/m3) for B6C3F1 and Swiss-Webster mice, respectively. There were no differences in strain sensitivity.

Repeat dose inhalation studies have been conducted in rats, rabbits and hamsters and in several of these respiratory tract irritation has been observed; the key studies are considered to be the 28 day inhalation studies in rat (Staal et al., 2008; Arts et al, 2004, see detail in Section 5.6: Repeat dose toxicity).

In the study reported by Arts et al, 2004, evidence of nasal irritation was seen at all exposure concentrations and the NOAEC for local effects was <20 mg/m3 (LOAEC 20 mg/m3) based on metaplasia and atypical hyperplasia of the transitional respiratory epithelium in the anterior part of the nose, and epithelial disarrangement of the olfactory epithelium. In the later study (Staal et al., 2008), the exposure concentrations were selected in order to determine a NOAEC for nasal irritation. At 20 mg/m3 respiratory epithelial hyperplasia and inflammatory cell infiltration was seen in both males and females at the end of the study, but only mononuclear cell infiltration in a single male following 4 weeks recovery. A clear NOAEC for local effects was established at 8 mg/m3.

In the study reported by Arts et al, 2004, evidence of nasal irritation was seen at all exposure concentrations and the NOAEC for local effects was <20 mg/m3 (LOAEC 20 mg/m3) based on metaplasia and atypical hyperplasia of the transitional respiratory epithelium in the anterior part of the nose, and epithelial disarrangement of the olfactory epithelium. The effects at the lowest dose 20 mg/m3 were confined to the anterior part of the nose. The treatment related effects are described as:

(1) transitional respiratory squamous metaplasia

(2) transitional respiratory epithelial atypical hyperplasia

(3) respiratory epithelial squamous metaplasia

(4) respiratory epithelial atypical hyperplasia

(5) respiratory epithelial disarrangement

At 20 mg/m3, (1) and (2) were increased compared to controls at low severity; (3) no change from controls; (4) no effect in males, marginal increase in females and (5) similar to controls( 1 versus 3 in males and 2 versus 4 in females). Although the authors use the term “LOAEC”, it is considered that the effects seen are minimal and should be considered to be adaptive (ECETOC, 2002).

In the later study (Staal et al., 2008), the exposure concentrations were selected in order to determine a NOAEC for nasal irritation. At 20 mg/m3 respiratory epithelial hyperplasia and inflammatory cell infiltration was seen in both males and females at the end of the study, but only mononuclear cell infiltration in a single male following 4 weeks recovery. A clear NOAEC for local effects was established at 8 mg/m3.

This study used 10/sex/group, compared with 5 in the previous study. In addition, more sections of nasal cavity were examined. The text refers to exposure related inflammatory changes in the nasal cavity ( level3) in the top dose (20mg/m3) which were not seen at lower levels. This seemed to be different from the earlier study, but on review of the tables, the inflammatory changes occur in a low number of rats and are also present in controls in some levels of the nose. The reason level 3 is highlighted is that there was very slight unilateral inflammatory cell infiltration' is present in 3/10 males and 5/10 females compared with zero incidence in the controls. There is also very slight unilateral transitional respiratory epithelial hyperplasia in 3/10 males and 3/10 females. In summary “the effect” at 20 mg/m3 in this study appear to be less than the previous study and are marginal. There were no changes at lower doses or after 28 days recovery.

In summary, this recent 2008 study is essentially more powerful (larger group sizes, more nasal sections), has a clear NOAEC, but of note are the equivocal results at the higher dose of 20 mg/m3 which could be regarded as adaptive with clear recovery.

Furfural induced histopathological changes in the nose of syrian golden hamsters exposed to vapours at concentrations up to 2165 mg/m3 for 6 hours/day, 5 days/week for 13 weeks. Based on observed histopathological changes in the nose (focal atrophy of the olfactory epithelium, often accompanied by accumulation of sensory cells in the lamina propria and occurrence of cyst-like structures) at 0.448 mg/L, the NOAEC for local effects after 13 week exposure by inhalation of hamsters is reported as 0.077 mg/L (77mg/m3; Feron,1979).

Rabbits were exposed up to 1000 mg/m3 by inhalation, for 4 hours/day, 5 days/week, until death (<80 days). Following exposure to 1000 mg/m3/h rabbits showed signs of irritation of the conjunctiva and the mucosa of the upper respiratory tract; autopsy revealed congestion and oedema of the lungs (Castellino, 1963).

Human data

Data on the irritating properties of furfural vapour in humans are reported by ACGIH (2001). A study by the NIOSH in a grinding wheel plant revealed a general higher incidence of eye and respiratory tract irritation which was attributed to furfural vapour concentrations of 20-63 mg/m3. Ten workers noted symptoms of nasal irritation which included stuffiness, dryness or soreness, and in one case, occasional bloody nasal discharge. Dryness of the mouth or throat was reported by seven workers. A health hazard evaluation was conducted by NIOSH at a manufacturer of aluminium graphite tubes where irritation of the eyes, mouth, throat, and nose, and respiratory symptoms were reported. Eleven of the 15 (73%) time weighed average personal breathing zone furfural concentrations (range 0.3-4.2 ppm) exceeded the threshold limit value of 8 mg/m3 (2 ppm). However, the authors state the influence of other hazardous substances and factors cannot be excluded, and are unknown; therefore, these data cannot be considered for risk assessment purposes. During regular medical examinations of employees at the Jugotanin plant in Slovenia (Medicinski Center, 1989), none of the employees had any changes in lung functions or other evidence of respiratory irritationwhich were considered to have been a consequence of daily contact with furfural.

Data from surveys of occupational exposure practices at a Quakers Oats plant (Bochinski, 1982) and other occupational hygiene data supplied by Quaker Oats indicate that exposures which elicit an irritant response are in excess of 5 ppm (20 mg/m3).

In a compilation of data taken from the industrial hygiene literature and other compilations of odour threshold data (Ruth, 1986), the odour threshold was reported to range from 0.0240 mg/m3 (low) to 20 mg/m3 (high) with the description of odour as “almonds”. The author reported that the irritating concentration of furfural was at 48.00 mg/m3; considerably higher than the odour threshold.


Effects on skin irritation/corrosion: irritating

Effects on eye irritation: irritating

Effects on respiratory irritation: irritating

Justification for classification or non-classification

The skin and eye irritation data reviewed and summarised in the EU RAR is 2008 predates some of the studies summarised in this dossier; however the outcome is concluded to be the same warranting similar classification under DSD and CLP.

Furfural is irritating to the skin, leading to classification with R38 “Irritating to skin” and H315 “Causes skin irritation” according to Dir 67/548/EEC and CLP Reg (EC) No 1272/2008, respectively.

Furfural is irritating for eyes, leading to classification with R36 “Irritating to eyes” and H319 “Causes serious eye irritation” according to Dir 67/548/EEC and CLP Reg (EC) No 1272/2008, respectively.

Irritation of the respiratory system of rats, rabbits and hamsters was seen after inhalation exposure. Based on these observations furfural is a respiratory irritant and is classified as follows: R37 “Irritating to respiratory system” according to Dir 67/548/EEC; H335 “May cause respiratory irritation” under Reg (EC) 1272/2008.