Use of this information is subject to copyright laws and may require the permission of the owner of the information, as described in the ECHA Legal Notice.
EC number: 200-001-8
CAS number: 50-00-0
Skin irritation (rabbit, 20h, occlusive): corrosive (similar to OECD TG 404, non-GLP)
Eye irritation: No studies are available, however, as formaldehyde has skin corrosive properties, no testing is required.
Skin irritation in 2 rabbits
Exposure duration 20 h
24 h after initiation
Superficial Ne & B
48 h after initiation
72 h after initiation
Ne (leather like)
Ne (black, hardened)
Ne: necrosis; B: bleeding
In a reliable study comparable to OECD TG 404, the test substance was tested for its potential to cause acute dermal irritation or corrosion by a single topical application for 20 hours to the shaved skin of 2 White Vienna rabbits, using a patch of 2.5 cm x 2.5 cm soaked with 40 % formalin, covered with occlusive dressing. After removal of the patch the application area was washed off. Readings were performed 24, 48, 72 h and 6 and 8 days after application.
Beside edema and erythema superficial necrosis was obvious 24 h after initiation of exposure. In both rabbits full thickness necrosis were found the following days.
A solution of 40 % formalin induced full thickness necrosis in 2 rabbits after 20 h occlusive exposure.
In a study comparable to OECD Guideline 404 (BASF AG, 1973 & 1974) with acceptable restrictions (exposure period 20 h; occlusive dressing; application of a patch soaked with ca. 1 mL formalin; partly limited documentation, no details about the test substance) two White Vienna rabbits were exposed for 20 h under occlusive dressing to a patch (2.5 x 2.5 cm²) soaked with 40% formalin.40% formalin, no further details Readings were performed 24, 48, 72 h and 6 and 8 days after application. Besides edema and erythema, superficial necrosis was obvious 24 h after initiation of exposure. In both rabbits full thickness necrosis were found the following days.
In supporting studies on rats (Sekizawa et al., 1994) authors reported “erosion” of the skin at 7-9% aqueous formaldehyde solution; no effects were seen at this concentration in mice and guinea pigs using a similar experimental design. In rats, however, no skin effects were reported at a concentration of 3% (Sekizawa et al., 1994).
The results from studies on sensitization give information on max. non-irritant concentrations of aqueous solutions in the GPMT. A 2% solution was non-irritant in 4 independent studies (challenge concentration in non-induced controls) and concentration used for topical induction (minimal irritant concentration) was 5 or 10% solution (see Section Skin sensitization; Bayer AG, 1987; Kimber et al., 1991; Hoechst AG, 1983; Hilton et al., 1996). In the study of Hoechst AG (1983) even a concentration of 4% resulted in no irritation in 10 controls (24 h, occlusive; 1st reading). Sensitization studies in humans (see WHO, 1989 in Section Sensitization data (humans)) suggested a non-irritant concentration at 1%; in diagnostic patch testing for sensitization concentrations of formaldehyde not leading to irritation have to be used (i.e. many authors proposed 1%).
A literature search after the last IUCLID update was carried out up to April 20, 2015 and provided the following new information:
Saito et al. (2011, supporting) investigated the reaction of the skin of mice after application of 2, 5, and 10% formaldehyde solutions in acetone, once per week over 5 weeks. 25 µl were painted to the dorsal and ventral surface of the ear lobes. Skin thickness was significantly increased by 10% and 5% and numerically by 2%. Histopathological alterations were observed, like infiltration of inflammatory cells or skin hypertrophy, by the 5 and 10% solutions. The effect was significantly suppressed by capsazeine, a TRPV-1 antagonist, showing that the TRPV-1 nociceptive receptor is involved in the skin response. The involvement of the TRPV-1 receptor was further substantiated by investigating TRPV-1 gene-knockout mice which showed an attenuated response in the ear swelling test as compared to wild type animals (Usuda et al., 2012, supporting). And finally, also Tian et al. (2009, supporting) showed that this nociceptive receptor is involved in vivo in formaldehyde induced pain after injection of a 4% formaldehyde solution into the plantar surface of rat hind paws, as well as in vitro by electrophysiological measurement of the response in dorsal root ganglion cells. The in vivo and in vitro effects, both were attenuated by capsazeine.
In summary, irritant effects are expected at concentrations > 3%. These effects are mediated by the TRPV-1 nocireceptor in the skin.
In support of this recommendation, a recent study on microvascular leakage of rat skin after dermal application for 40 minutes showed that skin damage was demonstrated at concentrations >= 2.5% formaldehyde (increases in the amount of Evans blue dye extravagated into the skin), the NOAEC was 1% (Futamura et al., 2009).
No data are available according to current guidelines. Studies on skin irritation/corrosion brought evidence for corrosive properties (no testing on eye irritation required). Data presented by Carpenter & Smith (1946, limited validity) suggested corrosive properties of formaldehyde solutions (5µL of 15% solution) after exposure to the rabbit eye. Additionally, in the study of Sekizawa et al. (1994), only 0.01 mL of 7-9% formaldehyde solution in water resulted in opaque cornea (other parameters of eye irritation were not scored).
Therefore, corrosive properties of formaldehyde solution in the eye at concentration in the range of 7% are to be expected.
Lai et al. (2013, supporting) studies the effect of aqueous formaldehyde solutions on rabbit corneal cells in vivo and in vitro in a non-guideline study. In vivo, the eyes were exposed to small filter discs soaked with 0, 20, 100, 200, or 300 ppm of formaldehyde in water for 5 min. 1, 3, 7 and 10 days after exposure filter strips were introduced into the rabbit’s eye to measure moisture according to the Schirmer’s test. In vitro rabbit corneal cell preparations were exposed to aqueous solutions containing 5-600 ppm formaldehyde over 3-5 min and cell morphology, death and proliferation were determined as well as changes of mitochondria. The authors concluded that in vitro already 5 ppm induced damage to the corneal cells that became apparent only after prolonged observation period. Higher concentrations led to effects already after shorter periods. In vivo, tear production was increased at all exposure concentrations and observation times.
Irritation in humans exposed to gaseous formaldehyde
The exposure to gaseous formaldehyde resulted in irritation of eyes, nose and throat. There is evidence that eye irritation in humans is the most sensitive endpoint. Slight discomfort due to irritation (mainly eyes) was noted in some individuals even at a concentration of 0.3 mg/m³ (0.25 ppm; variation at all dose levels within +-20% target concentration). Although subjects were asked for rating of discomfort due to the degree of airway irritation in the results section only ratings for discomfort are presented without differentiation for irritation to the eyes or respiratory tract. Even at the highest concentration tested (2 mg/m³) the average discomfort never exceeded the middle of the “slight discomfort” range (Andersen & Molhave, 1983). A main drawback of this study is the lack of a sham exposure group.
In another study with 10 volunteers no eye irritation occurred at 0.5 ppm (analytical and target concentration very close together) but odor was detected; dose dependent eye irritation was reported at ≥ 1 ppm (Kulle, 1993, exposure period 3 h,).
There are numerous other studies available with similar findings. A detailed documentation and evaluation of ocular and respiratory irritation in humans are presented in reviews by Paustenbach et al. (1997) and Arts et al. (2006). A summary of selected studies on this endpoint is tabulated in Section Exposure related observations in humans - Direct observations, (Summary chemosensory irritation).
Both reviews (Paustenbach et al., 1997; Arts et al., 2006) primarily relied on human volunteer studies, since according to the authors investigations on subjective irritant effects at the workplace may be compromised by two major problems:
- Fluctuations of the exposure concentrations with higher peak exposures that may lead to an important recall bias for subjective reporting of irritant effects;
- Presence of contaminants at the workplace other than formaldehyde.
The analysis by Paustenbach et al.(1997) was based on the construction of a concentration effect curve for eye irritation. In the review of Arts et al. (2006) a benchmark analysis was carried out in addition for the studies of Andersen and Mölhave (1983) and Kulle (Kulle, 1993; Kulle et al., 1987) taking into account the possible non-zero response at 0 ppm exposures.
Even when concentrating only on controlled volunteer studies both review groups nevertheless noted for all studies one or more of the following deficiencies:
– reliance only on subjective scorings without objective measurements for irritation (apart from Weber-Tschopp, 1977)
– such subjective reporting do not reliably differentiate between irritation and odor perception.
– often zero exposure groups are missing to allow for an estimate of the background of subjective reporting for irritancy (up to 15-20 background incidence)
– missing criteria for severity of effects to differentiate between perception / annoyance / adversity
Both of these reviews basically came to the same conclusions:
– irritation to eyes is more sensitive than that to nose and throat
– concurrent control groups show that 15-20% of the exposed volunteers will report slight subjective eye irritation even at zero exposure (sham exposure data measured in 7 controlled studies, Paustenbach et al. (1997)
– subjective eye irritation at low exposure levels rapidly subsides after termination of exposure
– appreciable / moderate / annoying eye irritation will start at or above 1 ppm
– between 0 and 0.3 ppm there is no increase in eye irritation above a background level of 10- 20% of volunteers reporting effects at zero exposure
– reports of irritation below 0.3-0.5 ppm are too unreliable to attribute the irritation solely to formaldehyde
– finally, Arts et al. (2006) concluded that sensory irritation in humans can be strongly influenced by subjective feelings and odor may confound reporting of irritation.(Dalton et al. cited in Arts et al., 2006)
An important shortcoming of all studies (but one) reviewed by Paustenbach et al. (1997) and Arts et al. (2006) was that only subjective reporting for irritation were given. The only “old” study investigating objective eye irritation parameters found an increase in eye-blinking frequency at a high concentration of 1.7 ppm (Weber-Tschopp et al., 1977 as referenced by Paustenbach et al., 1997).
If only subjective ratings on irritation are available a careful interpretation is necessary to avoid misinterpretation due to bias or confounding factors. There are several important factors that need to be considered before subjective findings can be interpreted as true adverse health effects (Dalton, 2002; cited in Arts et al., 2006):
- exposure history: workers in comparison to naïve volunteers;
- expectations and beliefs: prior information obtained for the chemical;
- bias from odor perception: problems in differentiation between irritation and odor;
- social factors: influence by behaviour of bystanders or study director;
- personality variables: influence on expectation by positive or negative affectivity.
In the case of only subjective symptoms being reported clear to moderate subjective eye irritation should be selected as POD rather than very slight to slight irritation as this level of response is often already is reported at near zero exposure (Paustenbachet al, 1997; Artset al, 2006b).
The deficiencies of the former studies on local sensory irritation were overcome by a recent investigation of Lang et al. (2008). In contrast to many of the former investigations this study included objective measures for sensory irritancy besides subjective ratings, peak exposures, masking of olfactory effects by a strong odorant (ethyl acetate) and physical activity. Further, the evaluation of subjective symptoms took into account positive / negative affectivity traits of the participants as determined by the PANAS (Positive And Negative Affect Schedule) questionnaire. For example a negative affectivity trait is indicated if a subject claims to be “distressed” or “irritable”, while the positive affectivity counterpart would be “active” or “proud”.
In short, the following results were obtained:
- Blinking frequency and conjunctival redness as objective symptoms for eye irritation, ranging from slight to moderate, were significantly increased by short-term peak exposures of 1.0 ppm on a baseline exposure of 0.5 ppm. The NOEL for these objective parameters was 0.3 ppm with 15 min peaks of 0.6 ppm formaldehyde or a constant exposure to 0.5 ppm.
- Subjective ratings of eye and respiratory irritation and olfactory symptoms occurred already at concentrations of 0.3 ppm, but not at 0.15 ppm; however, exposure to ethyl acetate only was also (wrongly) perceived as irritating.
-Volunteers who rated their personality as “anxious” by PANAS tended to report more pronounced complaints. When “negative affectivity” was used as covariate, the level of 0.3 ppm was no longer an effect level, but 0.5 ppm with peaks of 1 ppm was. This corresponds to the effect level obtained by objective parameters.
- Whenever increased symptom reporting occurred, these scores were reversed 16 h after the end of exposure.
These results clearly show that subjective reporting on irritation is strongly influenced by
- the interaction with odor perception as demonstrated by co-exposure to ethyl acetate and
- personal affectivity traits as demonstrated by using PANAS as covariate.
Therefore, taking into account the variety of possible confounders on subjective symptoms preferential it is suggested that objective findings like eye blinking frequency or conjunctival redness should be used as reliable indicators for irritation to the eyes.
In conclusion, this study provides two well-founded NOELs for eye irritation being the most sensitive effect parameter for formaldehyde, namely 0.3 ppm plus 15 min. peaks of 0.6 ppm or 0.5 ppm continuously.
Recently a new study on chemosensory effects in human volunteers became available from the same institute (Mueller et al., 2013). In comparison to Lang et al. (2008) the new investigation included the following new features:
- increase of the number of participants from 21 (Lang et al., 2008) to 41.
- Refinement of the NOAEC/LOAEC by using the following exposure concentrations: 0 ppm; 0.3 ppm + peaks of 0.6 ppm; 0.4 ppm + peaks of 0.8 ppm; 0.5 ppm; 0.7 ppm each over 4 h.
- An important additional aspect was allocation of the participants according to their sensitivity for trigeminal nasal irritation (CO2threshold) into subgroups of hyper- and hyposensitives (20 persons each) and extremely hyper- and hyposensitives (10 persons each).
Eye blinking frequency and subjective ratings of symptoms and complaints (SPES) were determined before and within the last 15 min of exposure, and the following parameters were examined before and shortly after exposure:
- conjunctival redness
- tearfilm break-up time
- nasal resisstance and flow
- olfactory functions (n-butanol threshold, sniffing sticks for odor discrimination and identification).
Similar to the study of Lang et al. (2008) personal affectivity traits were taken into account by PANAS.
None of the objective parameters showed an effect indicative for sensory irritation up to the highest exposure concentrations of 0.7 ppm or 0.4 ppm + peaks of 0.8 ppm:
- conjunctival redness. In comparison, Lang et al. (2008) found an increased conjunctival redness during peak exposure to 1.0 ppm. The findings of both studies together indicate that the NOAEL is in the range between 0.7 – 1.0 ppm and any effect caused by peak exposures around 1 ppm will rapidly subside.
- Eye blinking frequency even tended to decrease in the total group as well as in the different sensitivity subgroups. This leads to the same conclusion as for conjunctival redness when comparing the present data to those of Lang et al. (2008).
- For nasal flow and resistance no exposure related effect was found.
- Tearfilm break-up time rather showed a trend for an increase but not for a decrease as would have to be expected for substances leading to eye irritation.
- No exposure related effect was observed for olfactory function (n-butanol threshold, odor discrimination or identification.
For subjective symptoms (subdivided into the subgroups of“eye irritation”, “nasal irritation”, “olfactory symptoms” and “perception of impure air”) the following results were obtained:
- there were no statistically significant changes of subjective symptoms or dose related trends for the categories “eye irritation” and “nasal irritation” in all subgroups of participants.
- for “olfactory symptoms” there was a statistically significant increase for the total group at >= 0.3 ppm with 0.6 ppm peaks, when compared with the pre-exposure examination, and at >=0.4 ppm with 0.8 ppm peaks, when compared to 0 ppm exposure. However, there was no clear dose-response relationship. Generally, hypersensitive volunteers reported more often, partly significantly, “olfactory symptoms” than hyposensitive subjects, especially in the extreme subgroups.
- there was a statistically significant increase for the SPES item “perception of impure air” in the total group and in “hypersensitive” and “extremely hypersensitive” volunteers compared to “hyposensitive” participants. However, there was no significant change of “perception of impure air” in comparison with the control (zero concentration), since volunteers exposed to zero concentration did also report an increase of their “perception of impure air”.
The findings for the symptom subgroups “olfactory symptoms” and “perception of impure air” cannot be taken as adverse effects in the sense of subjective sensory irritation, but they are rather indicative for annoyance. Furthermore, they point to the difficulty to clearly differentiate between olfaction and sensory irritation.
The findings for eye and nasal irritation are in contrast to results of Lang et al. (2008). By statistical analysis taking into account negative personality traits obtained by PANAS questionnaire, Lang et al. (2008) gave strong evidence that these effects were related to personality factors but were not due to clear subjective sensory irritation. This is strongly supported by Mueller et al. (2013): they compared the PANAS scores of both studies and found that the participants of the Lang study had a statistically significantly higher negative affectivity than those of the Mueller study. Thus the subjective scoring for sensory irritation was driven by personality traits in the Lang study.
In conclusion, the study of Mueller et al. (2013) showed that
IARC (2006) presented a summary on toxic effects in animals (see Section Additional toxicological information) as well as a summary on toxic effects in humans including data on local irritation.
Mueller et al. (2013, key) further refined the results obtained by Lang et al. (2008) (both study entries are available in IUCLID section 7.10.3). 41 male volunteers (non-smokers, age ±9.9 years) were exposed in a randomised schedule to 0, 0.5, 0.7 ppm and to 0.3 ppm with 4 15 min peaks of 0.6 ppm and to 0.4 ppm with peaks of 0.8 ppm. During exposure 4 cycle ergometer units at 80 W were performed for 15 min. Subjective pain perception induced by nasal application of carbon dioxide served as indicator for sensitivity to sensory nasal irritation to define subjects hyper- and hyposensitive for irritation. The following parameters were examined before and after exposure: subjective rating of symptoms and complaints (Swedish Performance Evaluation System), conjunctival redness, eye-blinking frequency, self-reported tear film break-up time and nasal flow rates. In addition, the influence of personality factors on the volunteer’s subjective scoring was examined (Positive And Negative Affect Schedule; PANAS). No indications for subjective or objective indications of sensory irritation were obtained under these exposure conditions. There was a statistically significant differences for olfactory symptoms, especially for the ‘perception of impure air’ when comparing the subjective symptoms under formaldehyde exposure with zero exposure. These subjective complaints were more pronounced in hypersensitive subjects. When comparing the studies of Lang and Mueller, Lang et al. (2008) observed subjective symptoms of eye irritation already at 0.3 ppm while these effects were not found by Mueller et al. (2013) even at higher exposures. This is explained by Mueller et al. (2013) by the fact, that negative affectivity was significantly higher in the subjects exposed in the Lang study as compared to those of Mueller. The increased ‘perception of impure air’ was attributed by the authors impairment of well-being caused by situational and climatical conditions in the exposure chamber, because a statistically significant difference in symptom scores between FA exposures and control condition was missing, and hypersensitive subjects reported statistically significantly higher complaints even after exposure to 0 ppm. The NOAEC for sensory irritation was 0.7 ppm over 4 h and 0.4 ppm with peaks of 0.8 ppm.
The importance to clearly differentiate between sensory irritation and olfaction was demonstrated by Berglund et al. (2012, key). 31 subjects (18–35 years old) were exposed to formaldehyde at concentrations varying between 6.36 and 1000 ppm corresponding to the Swedish TLV. Exposure was carried out in a hood exposure system and the volunteers took one sniff of the atmosphere over 3 sec (3 sniffs/min).P50absolute thresholds (50% correct detection of odor) were for formaldehyde odor 110 ppb (range 23–505). For sensory irritation the P50could not be calculated because too few subjects were studied and the exposure was limited to 1000 ppb. But all thresholds for irritation were higher than for odor.
In a comprehensive review about factors that may influence olfaction Greenberg et al. (2013, key) concluded that perception of odor cannot be used as a surrogate marker for chemical exposure. Odor perception is affected by the psychological state and bias because odor is often negatives biased by association with health-related symptom.
Some further studies in humans on self-reported symptoms during prolonged work with formaldehyde are mentioned here, although not related to single exposures. As far as subjective symptoms were recorded these studies are less reliable than those in volunteers under controlled exposure conditions. The major problems stem from factors like exposure to mixtures of unknown composition, peak exposures that were not controlled for, or recall bias to former exposures. The main subjective symptoms were reported by students of a gross anatomy dissection course using a more detailed questionnaire (Mori et al., 2013, supporting). The symptoms were reversible 6 months after the course. Apart from subjective reportings, again the problem is exposure assessment. Exposure was reported to be about 0.2 ppm, but this was measured by area and not by personal sampling for 20 min after start of the course. Thereby personal peak exposures could not be accounted for.
Some studies reported objective parameters associated with formaldehyde exposure. Neghab et al. (2011, supporting) carried out a cross sectional study with 70 workers of a melamine-formaldehyde producing factory including 24 non-exposed referents. In total 7 air samples were taken over 40 min. The frequency of some of the self-reported respiratory symptoms (e.g. cough, chest tightness, episodes of chest illness associated with cold) was significantly higher in the exposed population and pre- and post-shift parameters of pulmonary function showed significant decrements. A recovery of lung function capacity was observed following temporary cessation of exposure. Airborne formaldehyde exposures clearly exceeded current exposure limit values with 0.78 ppm (SD=0.4). Hisamitsu et al. (2011, supporting) investigated serum IgE levels, olfactory tests and nasal sensitivity to histamine in 41 medical students before, during and after an anatomy dissection course. Olfactory anomalities and increased histamine sensitivity were observe red during and immediately after the course but the effects were reversible and disappeared after completion of the course. Formaldehyde concentrations ranged from 0.51-0.97 ppm (mean 0.67) in the centre of the laboratory.
Wieslander and Norbaeck (2010, supporting) studied symptoms of upper respiratory tract irritation in 31 indoor painters. Exposure concentrations over 8 h were measured for formaldehyde, other volatile organic chemicals and microbial organic chemicals. In comparison to controls, painters had increased ocular symptoms, lysozyme and myeloperoxidase in nasal lavage, while tear film break-up time and nasal patency (by acoustic rhinometry) were reduced. As the daily formaldehyde levels were below detection limits (30 µg/m³) and other volatile organic chemicals prevailed, the reported symptoms are not caused by formaldehyde.
A literature search after the last IUCLID update was carried out up to May 2022 and provided the following new information:
In the context of a restriction of formaldehyde and formaldehyde releasers, the sensory irritation studies of Lang et al. (2008) and Müller et al. (2014) were reevaluated (RAC & SEAC, Opinion on an Annex XV dossier proposing restrictions on formaldehyde and formaldehyde releasers, compiled version, adopted 13 March 2020 by RAC and 17 September 2020 by SEAC). In the view of RAC, such small numbers of volunteers and such a high variability of EBF in both studies do not allow a dose-related effect to be detected (unless the effect is sufficiently strong). Under these conditions (small size of samples, males only, very high variability of the test parameter, lack of data on continuous EBF monitoring during exposure) the lack of observed effects cannot be considered as evidence for the absence of dose-related effects. In addition to the uncertainties reported above, the variability of the results obtained might not reflect the variability of the general population, particularly of children, as indicated in a review by the JRC (2005). It is also noted that the respective exposure treatments in the studies by Lang et al. and Müller et al. were rather short (4 h) single exposure events (Lang et al., 2008). It is not known whether the threshold value for EBF would be different (i.e. lower) for formaldehyde, if exposure duration and/or frequency were expanded. Further weaknesses were identified in the Lang study: Data from male and female volunteers were pooled, although increased eye redness in females indicated a gender-specific difference.
In conclusion on the study of Lang et al. (2008) and Mueller et al. (2013), RAC considers that the absence of a sensory irritation effect at formaldehyde concentrations below 1.24 mg/m³ (1 ppm) is uncertain. Due to high variability of the measured effect (large ranges, high standard deviations), the low numbers of volunteers (yielding low statistical power) and the additional uncertainties identified, false negative results at 0.62 mg/m³ (0.5 ppm) or lower cannot be excluded. The studies of Lang et al. (2008) and Mueller et al. (2013), thus, cannot be used to derive a DNEL, because the uncertainties are too high and the lack of observed effects cannot be considered as evidence for the absence of dose-related effects.
In addition to the uncertainties reported above, the variability of the results obtained might not reflect the variability of the general population, particularly of children.
As a consequence, two new sensory irritation studies were initiated with specifically addressing the shortcomings and deficiencies of the studies of Lang et al. and Müller et al..
According to EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008, the substance has to be classified as Skin Corr 1B, H314 and Serious eye damage 1, H318.
Information on Registered Substances comes from registration dossiers which have been assigned a registration number. The assignment of a registration number does however not guarantee that the information in the dossier is correct or that the dossier is compliant with Regulation (EC) No 1907/2006 (the REACH Regulation). This information has not been reviewed or verified by the Agency or any other authority. The content is subject to change without prior notice.Reproduction or further distribution of this information may be subject to copyright protection. Use of the information without obtaining the permission from the owner(s) of the respective information might violate the rights of the owner.
Welcome to the ECHA website. This site is not fully supported in Internet Explorer 7 (and earlier versions). Please upgrade your Internet Explorer to a newer version.
Do not show this message again