This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Commonwealth
Download 5.77 Mb.
|
- Navigate this page:
- Acute toxicity
- Irritation
- Sensitisation
- Repeat dose toxicity
- Genotoxicity
- Carcinogenicity
- Reproductive effects
- Developmental toxicity
Hazard ClassificationThis section discusses the classification of the health effects of formaldehyde according to the NOHSC Approved Criteria for Classifying Hazardous Substances (the Approved Criteria) (NOHSC, 2004). The Approved Criteria are cited in the NOHSC National Model Regulations for the Control of Workplace Hazardous Substances (NOHSC, 1994c) and provide the mandatory criteria for determining whether a workplace chemical is hazardous or not. Where adequate human data were unavailable and/or inappropriate, the classification for health hazards has been based on experimental studies (animal and in vitro tests). In extrapolating results from experimental studies to humans, consideration was given to relevant issues, such as quality of data, weight of evidence, metabolic and mode of action/mechanistic profiles, inter- and intra- species variability and relevance of exposure levels. Classification of formaldehyde in accordance with the OECD Globally Harmonized System of Classification and Labelling of Chemicals (GHS) (UNSCEGHS, 2005) can be found in Appendix 4. Formaldehyde is currently listed in the OASCC’s Hazardous Substances Information System (DEWR, 2004) with classification of R23/24/25 (toxic by inhalation, in contact with skin, and if swallowed), R34 (causes burns), R43 (may cause sensitisation by skin contact) and R40 (limited evidence of a carcinogenic effect, Category 3 carcinogen).
Although there are old reports of human deaths following ingestion of formaldehyde solution, no reliable quantitative data are available on the doses consumed. Recent cases reported ulceration and damage along the aero-digestive tract, with a feeding jejunostomy performed following ingestion of approximately 700 mg/kg bw of formaldehyde solution, and a tracheostomy and gastrectomy performed following ingestion of an unquantifiable dose. In animal studies, oral LD50 values of 800 and 260 mg/kg bw are available in the rat and guinea-pig, respectively. A dermal LD50 of 270 mg/kg bw in the rabbit, and 4-hour inhalation LC50 values of 480 and 414 ppm (0.578 and 0.497 mg/L) in the rat and mouse, respectively, are also available. The LC50 value in rats, the preferred species, equates to ‘toxic’ by inhalation while the value in mice is almost at the cut-off value for toxic/very toxic. Thus, it is proposed that the classification as ‘toxic’ be retained. The oral LD50 values support classification as ‘harmful’. However, although no deaths occurred in recent cases of ingestion in humans they are considered to represent a potentially lethal dose given the significant toxicity observed, and drastic medical procedures undertaken. Consequently, it is considered appropriate to regard formaldehyde as ‘toxic’ by the oral route and retain its current classification as such. The dermal LD50 value in rabbits supports classification as ‘toxic’. Classification: Based on the human and animal data, formaldehyde meets the Approved Criteria for classification as ‘Toxic by inhalation’ (risk phrase R23), ‘Toxic in contact with skin’ (risk phrase R24) and ‘Toxic if swallowed’ (risk phrase R25).
Skin reactions have been reported in humans, however, because formaldehyde solution is a known skin sensitiser it is difficult to determine whether observed reactions are due to irritation or sensitisation. In animals, although formaldehyde solution is reported to be a primary skin and eye irritant, this is based on old anecdotal evidence rather than robust animal studies. Data are available from a recent rabbit low-volume eye test (LVET) where 10 l of 37% formaldehyde solution produced irritation of the cornea, conjunctiva and iris three hours post-instillation. Additionally, ‘necrosis/loss’ of corneal keratocytes was reported in eyes from animals sacrificed one day post- instillation, and corneal injury was determined to extend at times to 93.2% of corneal thickness. In a repeated dermal study in mice, skin irritation was reported following application of > 0.5% formaldehyde solution, 5 days/week for 3 weeks. A single 6-hour exposure to 15 ppm (18 mg/m3) gaseous formaldehyde produced histological changes to the nasal tract of rats indicative of a direct irritant effect. Data are also available in Alarie assays in mice. Although the reliability of this assay has been questioned (i.e. non-reproducibility of results and species variation in RD50 values) the data supports the histological findings that gaseous formaldehyde causes irritation to the respiratory tract. Thus, there are sufficient data to show formaldehyde is a skin, eye and respiratory irritant. The observations of severe irritation in the rabbit LVET and comprehensive injury to the cornea with 10 l of 37% formaldehyde solution, along with skin irritation at concentrations > 0.5% in a mouse repeat dermal study, raise concerns that corrosivity could be observed if animal studies were conducted to OECD Test Guidelines, i.e., at higher concentrations in skin studies and with 0.1 ml in eye studies. Additionally, corrosive injuries to the oesophagus and stomach were observed in humans following ingestion of formaldehyde solution. Consequently, it is considered appropriate to regard formaldehyde solution as corrosive. Classification: Based on the human and animal data, including observations in cases of human ingestion, formaldehyde meets the Approved Criteria for classification as ‘causes burns’ (risk phrase R34).
Formaldehyde solution is a known skin sensitiser and is included in standard series for patch testing. In addition to skin sensitisation being clearly observed in numerous clinical trials and case reports in humans, positive results have been observed in a large number of animal studies in guinea-pigs and mice. When determining whether a chemical is a respiratory sensitiser immunological mechanisms do not have to be demonstrated, and for human evidence it is necessary to take into account the size of the population and the extent of exposure. Although large numbers of people are exposed to gaseous formaldehyde, there are very few reported cases of well-conducted bronchial challenge tests in humans giving a positive response to formaldehyde. Conversely, several studies have reported negative bronchial challenge tests. However, limited evidence indicates that formaldehyde may elicit a respiratory response in some very sensitive individuals with bronchial hyperactivity, probably through irritation of the airways. Additionally, studies determining the effect on lung function following workplace exposure to formaldehyde in air, along with epidemiology studies, do not indicate formaldehyde to be a respiratory sensitiser. There is generally little correlation between the presence of formaldehyde-specific antibodies and respiratory symptoms in humans. Similarly, in animals, the results of immunoglobulin-E tests and cytokine profiles do not provide evidence that formaldehyde can induce respiratory sensitisation, though there is limited evidence available indicating that it may enhance allergic responses to other respiratory sensitisers. Thus, the available human and animal data indicates formaldehyde in air is unlikely to induce respiratory sensitisation. Classification: Based on the human and animal data formaldehyde meets the Approved Criteria for classification as ‘May cause sensitisation by skin contact’ (risk phrase R43) but not for sensitisation by inhalation.
Effects on pulmonary function, histological changes within the nasal epithelium, and neurobehaviour were investigated in populations exposed to gaseous formaldehyde in occupational and/or community environments. Though transient decreases in lung function across a work shift have been observed in some studies, overall, the data do not provide conclusive evidence that formaldehyde exposure induces major changes in pulmonary function. Conflicting results for histological changes within the nasal epithelium have been observed for workers occupationally exposed to formaldehyde. Although histological changes were observed in the most extensive and well conducted study (Holmstrom et al., 1989), the weight of causality is weak, due primarily to the limited number of investigations of relatively small populations that do not permit adequate investigations of exposure response. Additionally, it is not reported whether these studies examined other exposures that may have contributed to the observed histopathological changes. This is also true for the observance of histopathological changes in a community study. Consequently, the histopathological findings cannot be attributed to formaldehyde exposure. Likewise, there is presently no convincing evidence that indicates formaldehyde is neurotoxic. In animals, no evidence of systemic toxicity was seen in rat inhalation and oral studies up to approximately 2 years duration, or in the only dermal study available, a 2- to3-week rat study. Toxicity in response to irritation was restricted to the site of contact: skin irritation in the dermal study, histological changes in the nasal tract in inhalation studies, and stomach in oral studies. Classification: Based on the available human and animal data formaldehyde does not meet the Approved Criteria for classification as causing serious damage to health by prolonged exposure through inhalation, ingestion or dermal contact.
Overall, epidemiology data from occupational studies investigating cytogenetic effects in nasal and buccal cells are suggestive of formaldehyde having a weak localised genotoxic activity, while the evidence for a systemic activity, including peripheral lymphocytes, is equivocal. Small group sizes and the often limited details reported, limit the significance that can be attached to the observed effects. The main concern is that there was co-exposure to other chemicals in these studies (e.g. phenol in embalming fluid and resins, and wood dust in paper production) whose contribution to the observed effects cannot be precluded. Consequently, no reliable conclusions can be drawn from human data on the genotoxic potential of formaldehyde. In vitro, formaldehyde was clearly genotoxic in bacterial and mammalian cells: Ames test (+/- S9); gene mutation (-S9); chromosome aberration (+/-S9); SCE (+/-S9); and produced DNA single strand breaks and DNA protein cross-links (- S9). In vivo, several ip and inhalation studies are available in rodents investigating the genotoxicity of formaldehyde in somatic cells. Negative results were seen in bone marrow cytogenetic and micronuclei studies conducted to validated test methodology. A statistically significant increase in chromosomal aberrations (chromatid or chromosome breaks) in the bone marrow was reported in a single study that used a prolonged exposure period (4 months) and for which only limited details are available. Similarly, a positive result was seen in only one of several studies investigating tissues other than the bone marrow; a marginal, but statistically significant, increase in chromosomal aberrations (chromatid or chromosome breaks) in pulmonary macrophages. In the only oral study, which used a non-validated test method, a statistically significant increase in the proportion of cells with micronuclei and nuclear anomalies was seen in cells from the stomach, duodenum, ileum and colon of rats. However, the observed effects clearly correlated with severe local irritation (hyperaemia and haemorrhage), and are thus considered a likely consequence of cytotoxicity. Formaldehyde exposure did induce DPX in the nasal tract of rats and monkeys. In ip studies in germ cells in vivo, effects on sperm morphology and dominant lethal findings were seen in a single study that employed a 5-day exposure period. Although negative studies for germ cells used only a single administration, much higher dose levels were employed. Thus, the limited positive results in somatic cells in vivo are from cytogenetic studies that employed non-validated test methodology and, as such, neither study is considered to provide conclusive evidence of genotoxicity as uncertainty exists in interpreting the reliability of the data. In contrast, negative findings were observed in several studies conducted to validated test methodology. Similarly, the positive result in a single study in germ cells is not considered to provide conclusive evidence that formaldehyde is a germ cell genotoxicant, as negative results were seen in other studies at higher dose levels. The only other finding was the formation of DPX in the nasal tract following inhalation. Formaldehyde is genotoxic in vitro, and it appears that the chemical is weakly genotoxic at the site of contact in vivo. The relevance of the finding that formaldehyde is capable of producing DPX formation is unclear. Classification: Based on the human and animal data formaldehyde dose not meet the Approved Criteria for classification as a mutagenic substance.
There are a large number of epidemiology studies available (case-control and cohort) in industrial workers and professionals, investigating the incidence of cancers in the nasal tract, pharynx or lungs. Conflicting results have been observed in these studies. To consolidate the findings, meta-analysis of the data was conducted by Blair et al. (1990), Partanen (1993) and Collins et al. (1997). No association was seen in any meta-analysis for gaseous formaldehyde exposure and lung cancer. In contrast to earlier meta-analyses, the most comprehensive evaluation of the data by Collins et al. (1997) found no association (all studies combined) between sinonasal cancers and exposure to formaldehyde. An association was observed for nasopaharyngeal cancers in this meta-analysis, however, this was considered to be due to non-reporting of expected numbers in some industrial cohort studies. Following an adjustment for non-reporting of expected numbers, a non-significant increased risk was observed for nasopharyngeal cancers. Mixed results (i.e. occasional associations) have been observed for nasopharyngeal cancers in recent (post-1997) case-control and cohort studies. Consequently, it is considered that although the human data do not provide strong evidence of a causal association, it is acknowledged that there is some human evidence that occupational exposure to gaseous formaldehyde may result in the development of nasopharyngeal cancer. Increased risks of various non-respiratory cancers have occasionally been seen in some studies, with the most evidence being for leukaemia, particularly myeloid leukaemia. A recent update of a major cohort study of industrial workers reported an association for myeloid leukaemia and peak exposures to formaldehyde in air (Hauptmann, 2003). However, a reanalysis of the data, using additional analyses, provided little evidence to support the suggestion of a casual association (Marsh & Youk, 2004). In recent updates of two other major cohort studies of industrial workers, an increased risk of leukaemia was seen in US garment workers (Pinkerton, 2004), while no such increased risk was seen in UK industrial workers (Coggon, 2003). Furthermore, conflicting results were seen in earlier epidemiology studies investigating leukaemia in industrial workers (i.e. a slight increased risk or no risk). Increased risks for leukaemia have been observed in several studies of professional workers (e.g. embalmers), however, data on exposure to formaldehyde is not available for these studies. Overall, the data is considered insufficient to clearly establish an association between formaldehyde exposure and leukaemia. This conclusion is consistent with the present toxicokinetic profile and animal carcinogenicity data for formaldehyde. In inhalation carcinogenicity animal studies, a significantly increased incidence in nasal squamous cell carcinomas was observed in rats at concentrations > 6 ppm formaldehyde. Nasal polyploid adenomas were also observed in a single study at 15 ppm formaldehyde, however, the non-reproducibility of these findings at similar concentrations (14-14.3 ppm) in other studies indicates that they are not treatment related. In contrast, an absence or no significant increased incidence in nasal tumours was observed in mice and hamsters at equivalent or greater exposure concentrations that produced such tumours in rats. In oral carcinogenicity studies, no significant tumour findings were seen in the most comprehensive study available up to the top dose of 82 and 109 mg/kg bw/day in male and female rats, respectively (Til, 1989). Although an increase in ‘haemolymphoreticular tumours’ was seen in male and female rats at the top dose of 75 and 100 mg/kg bw/day, respectively, in another study, this study was criticised for its ‘pooling’ of tumour types whose incidence has been inconsistently reported. Similarly, although an increase in papillomas of the forestomach was seen in an initiation/promotion study where rats were administered 0.5% in drinking water for 32 weeks, the study has been questioned over its histological diagnosis of benign tumours. In contrast, no leukaemias or stomach tumours were seen in the most comprehensive study to date, which employed comparable or higher dose levels of formaldehyde solution. No skin tumours were seen in mouse initiation/promotion studies, the only dermal data available. Therefore, the available data in animals do not support formaldehyde being carcinogenic by the dermal or oral routes. The International Programme on Chemical Safety (IPCS) developed a conceptual framework in 2001 based on the general principles involved in considering the chemical induction of a specific tumour in animals (Sonich-Mullin et al., 2001). The data for nasopharyngeal cancers and leukaemia and formaldehyde exposure have been evaluated using this framework (Appendix 5). The postulated mode of action for nasopharnygeal cancers is that inhalation of formaldehyde causes inhibition of mucociliary clearance, followed by nasal epithelial cell regenerative proliferation resulting from cytotoxicity and DPX that leads to mutation, and consequent tumour formation. By considering the available data in the IPCS framework, it was concluded that the postulated mode of action for formaldehyde-induced tumours in the nose is likely to be relevant to humans, at least qualitatively. In contrast, a mechanism by which formaldehyde may induce leukaemia has not been identified and the framework highlights the low degree of confidence that may be ascribed to the hypothesis that formaldehyde induces leukaemia. Overall, it is considered that the available epidemiology data are not sufficient to establish a casual relationship between formaldehyde exposure and cancer. For nasopharyngeal cancers there are several epidemiological studies that show an increased risk, whereas other studies do not. There is also clear evidence from inhalation studies of nasal squamous cell carcinomas in the rat, though not the mouse and hamster. The postulated mode of action for these tumours is considered likely to be relevant to humans. Therefore, based on the available nasopharyngeal cancer data, formaldehyde should be regarded as if it may be carcinogenic to humans following inhalation exposure. There are also concerns for an increased risk for formaldehyde-induced myeloid leukaemia, however, the available data are not considered sufficient to establish an association and there is currently no postulated mode of action to support such an effect. IARC concluded that formaldehyde is carcinogenic to humans (Group 1), on the basis of sufficient evidence in humans and sufficient evidence in experimental animals. IARC’s conclusion is as follows: ‘Nasopharyngeal cancer mortality was statistically significantly increased in a cohort study of United States (US) industrial workers exposed to formaldehyde, and was also increased in two other US and Danish cohort studies. Five of seven case-control studies also found elevated risk for formaldehyde exposure. The Working Group considered it was “improbable that all of the positive findings…could be explained by bias or by unrecognised confounding effects” and concluded that there is sufficient evidence in humans that formaldehyde causes nasopharyngeal cancer. Leukaemia mortality, primarily myeloid-type, was increased in six of seven cohorts of embalmers, funeral-parlour workers, pathologists, and anatomists. These findings had previously been discounted because an increased incidence of leukaemia had not been seen in industrial workers. Recent updates, however, report a greater incidence of leukaemia in two cohorts of US industrial workers and US garment workers, but not in a third cohort of United Kingdom chemical workers. The Working Group concluded that there is “strong but not sufficient evidence for a causal association between leukaemia and occupational exposure to formaldehyde”. Several case-control studies have associated exposure to formaldehyde with sinonasal adenocarcinoma and squamous- cell carcinoma. However, no excess of sinonasal cancer was reported in the updated cohort studies. The Working Group concluded that there is limited evidence in humans that formaldehyde causes sinonasal cancer. In rats, several inhalation studies have shown that formaldehyde induces squamous-cell carcinoma of the nasal cavity. Four drinking-water studies gave mixed results. Formaldehyde also shows cocarcinogenic effects when inhaled, ingested, or applied to the skin of rodents. Formaldehyde is genotoxic in in-vitro models, animals and humans. Increased numbers of DNA–protein crosslinks have been found in peripheral blood lymphocytes of exposed workers, in the upper respiratory tract of monkeys, and in the rat nasal mucosa. Cell proliferation increases substantially at formaldehyde concentrations higher than six parts per million in rats, amplifying the genotoxic effects. The Working Group concluded that, “both genotoxicity and cytotoxicity have important roles in the carcinogenesis of formaldehyde in nasal tissues”. By contrast, the Working Group could not identify a mechanism for leukaemia induction, and this tempered their interpretation of the epidemiological evidence.’ (IARC, 2004b). The available data do not support formaldehyde being carcinogenic by the dermal or oral routes. Classification: Based on the above, formaldehyde meets the Approved Criteria for classification as a Category 2 carcinogen with risk phrase R49 ‘May cause cancer by inhalation’. This is a different category with the IARC classification which is Category 1, (known human carcinogen), principally due to differences in the carcinogen classification criteria and also consideration of the weight of evidence.
Only a few epidemiology studies are available. A retrospective investigation of fertility reported a significant increase in the time to pregnancy (i.e. decrease in the fecundability density ratio) in female workers exposed to formaldehyde. However, limitations in study design prevent any reliable conclusions being drawn from the data. Similarly, in cross-sectional studies, although no difference was seen in female fertility or male sperm count and morphology between formaldehyde exposed workers and controls, study limitations restrict the significance that can be attached to the data. In the only fertility study available in animals, formaldehyde did not produce an adverse effect on fertility in minks, though there are concerns that formaldehyde was not robustly tested in this oral study. No effect on epididymal sperm morphology was seen in an oral mouse study at the only dose tested, and no effects on the testes have been reported in rodents in a chronic repeat oral study and chronic inhalation studies. In contrast, although effects have been seen on epididymal sperm following intraperitoneal administration this is not a relevant route of human exposure. Classification: Based on the human and animal data formaldehyde does not meet the Approved Criteria for classification as a reprotoxicant.
There is no human evidence to indicate occupational exposure to formaldehyde is associated with low birth weight or malformations. For studies investigating spontaneous abortions, the inconsistent findings observed in epidemiological studies and limitations in study design, including the potential for recall and publication bias, mean the findings cannot be attributed to occupational exposure to formaldehyde. In animal studies, the only effect observed following inhalation was a reduction in foetal body weight that was a secondary non-specific consequence of severe maternal toxicity. No effects on development were seen in an oral study though dose levels were not maximised. No robust dermal study is available that allows the developmental toxicity of formaldehyde to be reliably determined. Classification: Based on the human and animal data formaldehyde does not meet the Approved Criteria for classification as a developmental toxicant.
|