Health Canada
www.hc-sc.gc.ca
Common menu bar links
Institutional links
Priority Substances List Assessment Report- 1,3-Butadiene
3.0 Assessment of "Toxic" under Cepa 1999
3.1 CEPA 1999 64(a): Environment
The environmental risk assessment of a PSL substance is based on the procedures outlined in Environment Canada (1997a). Analysis of exposure pathways and subsequent identification of sensitive receptors are the basis for selection of environmental assessment endpoints (e.g., adverse reproductive effects on sensitive fish species in a community). For each endpoint, a conservative Estimated Exposure Value (EEV) is selected and an Estimated No-Effects Value (ENEV) is determined by dividing a Critical Toxicity Value (CTV) by an application factor. A hyperconservative or conservative quotient (EEV/ENEV) is calculated for each of the assessment endpoints in order to determine whether there is potential ecological risk in Canada. If these quotients are less than one, it can be concluded that the substance poses no significant risk to the environment, and the risk assessment is completed. If, however, the quotient is greater than one for a particular assessment endpoint, then the risk assessment for that endpoint proceeds to an analysis where more realistic assumptions are used and the probability and magnitude of effects are considered. This latter approach involves a more thorough consideration of sources of variability and uncertainty in the risk analysis.
3.1.1 Assessment and measurement endpoints
Butadiene enters the Canadian environment mainly from natural and anthropogenic combustion sources, notably vehicle emissions, and from industrial on-site releases. Almost all releases to the ambient environment are to air, with very small amounts released to water and soil.
Given its physical/chemical properties, butadiene undergoes various degradation processes in air. When released to water or soil,it can undergo various biological and physical degradation processes. In addition, much of the butadiene released onto soil can be expected to volatilize to air. Butadiene is not bioaccumulative or persistent in any compartment of the environment, although its continual release from automotive and other combustion sources can lead to chronic exposure of biota.
Based on the sources and fate of butadiene in the ambient environment, biota are expected to be exposed to butadiene primarily in air. Some exposure in water or in soil is possible. Butadiene does not bioaccumulate, and it should largely be found in the gas phase in air or dissolved phase in water. Therefore, the focus of the environmental risk characterization will be on terrestrial and aquatic organisms exposed directly to ambient butadiene in air and water and on soil organisms exposed to butadiene in soil.
3.1.1.1 Aquatic
Experimental data for toxicity of butadiene to aquatic organisms are not available (Section 2.4.1). Modelled data are available for algae, one species of crustacean (acute and chronic exposures) and six species of fish (acute and chronic exposures). Experimental data are available for the related substances 1,3-pentadiene (algae, crustacean, fish) and isoprene (fish).
Algae are primary producers in aquatic systems, forming the base of the aquatic food chain, while zooplankton, including crustaceans, are key consumers and are themselves consumed by many species of invertebrates and vertebrates. Fish are consumers in aquatic communities and are themselves eaten by piscivorous fish, birds and mammals.
Therefore, although limited, the available studies cover an array of organisms from different taxa and ecological niches and are considered adequate for an assessment of risks to aquatic biota. The single most sensitive response for all of these endpoints is considered as the CTV for the risk characterization for aquatic effects.
3.1.1.2 Terrestrial
Data on terrestrial toxicity are available for six species of plants (experimental) and one soil invertebrate (modelled) (Section 2.4.1). Data pertinent to terrestrial vertebrate wildlife are available from several mammalian toxicology studies (Section 2.4.3).
Terrestrial plants are primary producers, provide food and cover for animals, and provide soil cover to reduce erosion and moisture loss. Invertebrates are an important component of the terrestrial ecosystem, consuming plant and animal matter while serving as forage for other animals. Vertebrate wildlife are key consumers in most terrestrial ecosystems.
Therefore, although quite limited, the available toxicity studies include organisms from different taxa and ecological niches and are considered adequate for an assessment of risks to terrestrial biota using hyperconservative exposure scenarios. The most sensitive responses for plants, invertebrates and vertebrates (acute and chronic exposure) will be used as CTVs for the risk characterization for terrestrial effects.
3.1.2 Environmental risk characterization
3.1.2.1 Aquatic effects
The lowest estimated concentration associated with aquatic toxicity for butadiene is 2.2 mg/L (2200 µg/L), based on a 16-day EC50 for Daphnia reproduction, derived through QSAR modelling; this value is used as the CTV. To calculate an ENEV, an application factor of 100 has been selected to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability, extrapolation from an EC50 to a no-effect concentration and the fact that, although there are chronic and acute data for a variety of aquatic organisms and corroborating data for other substances, all the information has been obtained by QSAR modelling rather than experimentation. This yields an ENEV of:
Since no empirical data are available for concentrations of butadiene in ambient waters in Canada, the assessment was based on predicted concentrations. For the industrial site having the highest reported air concentration, the equilibrium concentration in water was predicted to be 9.3 × 10-3 µg/L. This concentration can be used as the EEV in estimating the likelihood of risk in water.
A hyperconservative quotient can thus be calculated as follows:
Because the hyperconservative quotient is less than 1, this substance is unlikely to cause a harmful effect on populations of organisms in the ambient aquatic environment.
Extensive effluent monitoring data are available for the major producer and user of butadiene in Canada. Concentrations in raw, undiluted effluents were almost always below the detection limit (2839 samples analyzed with a detection limit of 1 µg/L, and 789 samples analyzed with a detection limit of 50 µg/L). Only two samples of undiluted effluents had measurable concentrations (80 and 130 µg/L) that exceeded the ENEV. It can therefore be expected that risks to aquatic organisms in receiving waters are low.
3.1.2.2 Terrestrial effects
3.1.2.2.1 Plants
The lowest concentrations of airborne butadiene associated with toxicity are a 21-day No- Observed-Effect Concentration (NOEC) of 22.1 mg/m3 and a 21-day Lowest-Observed-Effect Concentration (LOEC) of 221 mg/m3. These levels of exposure caused no effect on cotton, tomato and coleus plants and slight effects in cotton and tomato plants exposed for 21 days, respectively.
An ENEV for terrestrial plants can be derived from the LOEC of 221 mg/m3. An application factor of 100 has been selected to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability, extrapolation from a LOEC to a no-effect concentration and the fact that, although there are data for several plant species, there is only one laboratory study confirming these effects. This yields an ENEV of:
A similar ENEV could be obtained using the no-effect level of 22.1 mg/m3 and an application factor of 10 (no extrapolation to a no-effect concentration required).
At this time, the highest reported or predicted gas-phase concentration in outdoor air in Canada is 28 µg/m3 (reported at one site near the major producer and user of butadiene).
A hyperconservative quotient can thus be calculated as follows:
Because the hyperconservative quotient is less than 1, it is unlikely that butadiene is causing harm to populations of terrestrial plants through exposure in ambient air.
3.1.2.2.2 Soil invertebrates
The only soil toxicity concentration for butadiene is 335 mg/kg dry soil weight, which is a chronic 14-day LC50 earthworm survival value estimated using QSAR. An application factor of 1000 has been selected to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability, extrapolation from an LC50 to a no-effect concentration and the fact that there is only one value available, derived by QSAR. This yields an ENEV of:
The highest estimated soil concentration is 7.5 × 10-6 mg/kg (dry weight), yielding a hyperconservative quotient of:
Because the hyperconservative quotient is less than 1, this substance is unlikely to cause a harmful effect on populations of soil organisms in the terrestrial environment.
3.1.2.2.3 Mammalian wildlife
Since concentrations of butadiene are highest in urban areas and around industrial sites, citydwelling terrestrial organisms are considered to have the greatest potential for exposure. Small mammals such as deer mice are likely to have the highest exposure due to their rapid respiration rate and high metabolism. Although no data have been identified for wild animals, effects data are available for surrogates such as laboratory mammals.
Acute exposure
The most sensitive single-exposure lethality study identified for laboratory mammalian studies was a 2-hour inhalation LC50 for mouse of 268 g/m3 (Shugaev, 1969) (256 g/m3 for exposure of unknown duration; Batinka, 1966). This will be used as the CTV for the acute exposure of terrestrial wildlife to butadiene in air. An application factor of 10 has been selected to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability and extrapolation from an LC50 to a no-effect concentration. This yields an ENEV of:
A worst-case quotient is calculated by dividing the acute EEV of 28 µg/m3 (the highest airborne concentration of butadiene measured in Canada) by the ENEV. The resulting hyperconservative quotient (acute) is:
Biochemical changes were reported in mice after exposure for 7 hours at concentrations of 100 ppm (221 mg/m3) or more (Deutschmann and Laib, 1989). An application factor of 10 has been selected to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability and extrapolation from a LOEL to a no-effect concentration. This yields an ENEV of:
A worst-case quotient is calculated by dividing the acute EEV of 28 µg/m3 (the highest air concentration of butadiene measured in Canada) by the ENEV. The resulting hyperconservative quotient (acute) is:
Because the hyperconservative quotients are less than 1, it is unlikely that butadiene emissions will cause acute adverse effects on populations of terrestrial wildlife in Canada.
Chronic exposure
The most sensitive study identified for laboratory mammals was a chronic study in which mice were exposed to butadiene for 2 years (6 hours per day for 5 days per week). NTP (1993) reported that the LOEL was 6.25 ppm (13.8 mg/m3). Butadiene induced adverse toxic effects, including ovarian atrophy, in mice at this concentration. Mice were much more sensitive than other mammals to exposure to butadiene.
The ENEV is derived by dividing the CTV by an application factor of 10 to account for extrapolation from laboratory to field conditions, inter- and intraspecies variability and extrapolation from a LOEL to a no-effect concentration. This yields an ENEV of:
A worst-case quotient is calculated by dividing the chronic EEV of 1.0 µg/m3 (the 95th percentile for extensive air monitoring data under the NAPS program) by the ENEV. The resulting hyperconservative quotient (chronic) is:
Because the hyperconservative quotient is less than 1, it is unlikely that butadiene emissions will cause chronic adverse effects on populations of terrestrial wildlife in Canada.
A summary of the values used in the environmental risk characterization of butadiene is presented in Table 6.
3.1.2.3 Discussion of uncertainty
There are a number of potential sources of uncertainty in this environmental risk assessment. Regarding effects of butadiene on terrestrial and aquatic organisms, there is uncertainty concerning the extrapolation from available toxicity data to potential ecosystem effects. To account for this uncertainty, application factors were used in the environmental risk analysis to derive ENEVs.
All data for effects on aquatic and soil organisms are derived through QSAR modelling, and there is only one experimental study for terrestrial plants. Nonetheless, corroborating data are available for structurally and functionally similar substances, lending support to the modelled data.
Regarding environmental exposure, there could be concentrations of butadiene in Canada that are higher than those identified and used in this assessment. However, the air measurements used in this assessment are considered acceptable because they were selected from an extensive set of recent air monitoring data of urban and other sites, including key industrial manufacturing sites. Thus, available data on atmospheric concentrations are considered representative of the highest concentrations likely to be encountered in air in Canada.
Concentrations in water are expected to be low because of the limited releases to this medium and the limited partitioning of butadiene from air into water. Since no measurements of butadiene are available in ambient water, concentrations were predicted by modelling. Extensive monitoring data are available for the key industrial plant producing and using butadiene and indicate that butadiene is unlikely to occur in concentrations of concern even in undiluted effluents.
Despite some data gaps regarding the environmental effects of and exposure to butadiene, the data available at this time are considered adequate for making a conclusion on the environmental risk of butadiene in Canada.
3.2 CEPA 1999 64(b): Environment upon which life depends
Butadiene is generally released to air, and its properties largely preclude its partitioning into other compartments. There is thus a potential for butadiene to be involved in critical atmospheric processes. Butadiene does not deplete stratospheric ozone, and its potential contribution to climate change is negligible. Butadiene is more reactive (POCP of 407) than compounds such as ethene that are recognized as important in the formation of ground-level ozone. Given its high reactivity and the concentrations measured in air in Canada, butadiene represented approximately 0.9% of the total volatile organic carbon reactivity, ranking it 26th among nonmethane hydrocarbons and carbonyl compounds contributing to the formation of ground-level ozone (Dann and Summers, 1997). Butadiene may therefore be important in the photochemical formation of ground-level ozone in urban areas.
3.3 CEPA 1999 64(c): Human health
3.3.1 Estimated population exposure
The principal source of environmental exposure to butadiene is air. Although few data were identified regarding levels in drinking water and food, intake of butadiene in these media is expected to be negligible in comparison with that in air because of its physical/chemical properties (e.g., vapour pressure and partition coefficients) and environmental release patterns (i.e., principally atmospheric emissions).
Twenty-four-hour average concentrations of butadiene were measured in 9168 samples of outdoor air between 1989 and 1996 under the NAPS program (Dann, 1997). Sampling sites were located in rural, suburban and urban areas. In the absence of potential indoor sources, and if it is assumed that the general population in Canada is exposed to similar concentrations of butadiene, 50% of the population can be expected to be exposed to 24-hour average concentrations of up to 0.21 µg/m3, while 95% of the population can be expected to be exposed to 24-hour average concentrations of up to 1.0 µg/m3.
The general population in urban areas is exposed to higher concentrations of butadiene on an ongoing basis. The 90th and 95th percentile values of the distribution of concentrations in 2913 samples from nine urban NAPS sites were 0.8 µg/m3 and 1.1 µg/m3, respectively. Butadiene was detected in 98% of 1576 samples from four reasonable worst-case urban NAPS sites, at a mean concentration of 0.5 µg/m3. In the absence of potential indoor sources, and if it is assumed that highly exposed subgroups within the population in Canada are exposed to concentrations of butadiene similar to those at the reasonable worst-case sites, 50% of the population can be expected to be exposed to 24- hour average concentrations of up to 0.40 µg/m3, while 95% of the population can be expected to be exposed to 24-hour average concentrations of up to 1.3 µg/m3.
Exposures from ambient air may be substantially higher for populations in the vicinity of point sources. Concentrations of butadiene were measured at distances between 1 and 3 km downwind of an industrial point source of discharge to the atmosphere in Sarnia, Ontario (MOEE, 1995). If these concentrations can be considered as a worst case of the ongoing exposure of nearby residents, and in the absence of additional potential indoor sources, 50% of the population in the vicinity of this source can be exposed to short-term concentrations of up to 0.62 µg/m3, and 95% of this population can be exposed to short-term concentrations of up to 6.4 µg/m3.
Individuals may also be exposed to butadiene for short durations while at self-service gasoline filling stations or in parking garages. Estimates of average daily intake of butadiene by inhalation for various exposure scenarios indicate that intake is negligible while at self-service stations due to the infrequent occurrence and short duration of these exposures. Higher daily intakes are possible for commuters using personal motor vehicles and parking garages on a regular basis. However, these intakes are still much less than average daily intakes for the general population from inhalation of background concentrations of butadiene in outdoor and indoor air.
Although available Canadian data indicate that butadiene is detected with greater frequency in indoor air than in outdoor air, there are insufficient data to characterize the distributions of concentrations of butadiene in various indoor environments. In general, butadiene is detected more frequently and at higher concentrations in indoor environments contaminated by ETS than in areas where smoking does not occur. In nonsmoking indoor areas in Canada, the distributions of concentrations of butadiene are likely to be similar to the distributions of concentrations in the outdoor air samples from the NAPS program. Non-smokers who spend a considerable proportion of their time in indoor environments where ETS is present can be exposed to concentrations of butadiene that are an order of magnitude higher than the average levels in the outdoor air. Moderate tobacco use (e.g., 20 cigarettes per day) can increase the daily intake of butadiene by smokers by five times over the daily intake by non-smokers in ETS-contaminated indoor locations. The daily intake of butadiene by smokers can be 100 times greater than the daily intake of non-smokers who are not exposed to ETS.
3.3.2 Hazard characterization
As discussed in Section 2.4.2, although metabolism of butadiene appears to be qualitatively similar across species, there are extensive data that indicate that the putatively active epoxide metabolites are formed to a greater degree in mice than in rats. Similarly, although in vivo data are limited, humans appear to metabolize butadiene to the mono- and diepoxide metabolites to a much lesser extent than mice. However, based on the observed variability in the formation of adducts of hemoglobin with butadiene metabolites in occupationally exposed human populations, there appears to be interindividual variation in humans, which is likely related to polymorphism for genes that code for enzymes involved in the metabolism of butadiene. The weight of evidence for the carcinogenicity, genotoxicity and non-neoplastic effects of butadiene needs to be considered, therefore, in the context of these interspecies and interindividual variations.
3.3.2.1 Weight of evidence for carcinogenicity and genotoxicity
Data supporting the interspecies differences in production of active epoxide metabolites are in concordance with the observed difference in sensitivity between mice and rats (at least for the few strains investigated) to butadiene-induced carcinogenicity, in that the substance appears to be much more potent in mice than in rats. Although butadiene was a multi-site carcinogen in both mice and rats at all exposure levels tested (Hazleton Laboratories Europe Ltd., 1981a; NTP, 1984, 1993; Irons et al., 1989), the concentrations that induced tumours in the only study available in rats were much greater than those that were tumorigenic in mice (i.e., ≥1000 ppm versus ≥6.25 ppm).
Species differences in sensitivity to genetic effects induced by butadiene have also been observed. Although mutagenic in somatic cells of both mice and rats, the mutagenic potency of butadiene was greater in mice. Other genotoxic endpoints (chromosomal aberrations, sister chromatid exchanges and micronuclei) were noted in somatic cells of mice but not in those of rats exposed to much higher concentrations. Butadiene was genotoxic in germ cells of male mice in multiple assays, while negative results were obtained in the single dominant lethal study in rats. Unlike the observations with the parent compound, however, there is little evidence that there are species differences in the sensitivity to genotoxic effects induced by the epoxide metabolites of butadiene (EB, DEB and EBdiol), although there was some indication of interstrain variability. These data suggest that interspecies differences in sensitivity to butadiene-induced genotoxicity are related to quantitative differences in the formation of active metabolites.
There is also limited evidence of the genotoxicity of butadiene in exposed workers; although data are not completely consistent, increased frequencies of chromosomal aberrations, sister chromatid exchanges, and hprt - mutations and decreased DNA repair capability have been reported in some studies of workers in the monomer and/or styrene-butadiene rubber manufacturing industries (Legator et al., 1993; J.B. Ward et al., 1994, 1996; Au et al., 1995; Tates et al., 1996; Hallberg et al., 1997; Ward, 1997a, 1997b; Srám et al., 1998). The discrepancy in the results may be due to the use of different methods for the detection of mutations or differences in exposure levels. In addition, since sensitivity to induction of genetic effects by butadiene and its metabolites has been linked to genotype for glutathione-S-transferase enzymes in several in vitro and a few in vivo studies, interpretation of the inconsistent observations in the available database is complicated by the lack of information on genotype for most of the small populations examined.
There have been several epidemiological investigations of the carcinogenicity of butadiene that serve as a basis for assessment of the weight of evidence for causality based on traditional criteria. In the most recent cohort study (Delzell et al., 1995), which is also the largest and most comprehensive investigation conducted to date and that in which exposure was most extensively characterized, an association between exposure to butadiene in the styrene-butadiene rubber industry and leukemia was observed (i.e., there was a quantifiable exposure-response relationship). SMRs for leukemia were elevated for the overall cohort of workers from eight plants; the strength of this association was generally greater when specific subgroups with greater potential for exposure were considered. In addition, there was an increase in the RR for leukemia with increased cumulative exposure to butadiene in workers from the six plants for which exposure was best characterized. The association between leukemia and exposure to butadiene remained when the potential role of two other substances present in the work environment (i.e., styrene and benzene) was considered. Although further refinement of the estimates of exposure at one of these plants resulted in increases for several job categories (Macaluso et al., 1997), it is unlikely that these changes would affect the relative ranking of the categories and analyses in which exposed workers were compared with "non-exposed" workers (Gerin and Siemiatycki, 1998); therefore, these results are not inconsistent with the association observed by Delzell et al. (1995).
However, no increase in mortality due to leukemia was observed in studies of workers involved in the production of butadiene monomer who were not concomitantly exposed to the other substances present in the styrene-butadiene rubber industry (E.M. Ward et al., 1995, 1996; Divine and Hartman, 1996). Although there was some evidence of increased mortality due to lymphosarcoma and reticulosarcoma in the subgroup of workers potentially exposed to the highest concentrations of butadiene in the largest of these investigations, there was no association with duration of employment or estimated cumulative exposure (based on qualitative ranking of potential for exposure). Although mortality due to lymphosarcoma was non-significantly elevated in some process groups in the styrene-butadiene rubber cohort (Delzell et al., 1995), there were no consistent patterns (other than for leukemia), even when currently accepted terminology for lymphohematopoietic cancers was used (Sathiakumar et al., 1998).
The traditional criterion of consistency for the observed association between exposure to butadiene and leukemia is fulfilled, at least in part, in that similar excesses were observed among plants in the large cohort study of styrenebutadiene rubber workers (Delzell et al., 1995); i.e., there is internal consistency. A similar exposure-response was also noted in an independant nested case-control study of mostly the same population in which different exposure assessment methodology was employed (Matanoski et al., 1997). Observation of external consistency with results of other cohort studies of styrene-butadiene rubber workers is largely precluded, in view of the scope of the large epidemiological cohort study that included a large proportion of all of the styrene-butadiene rubber workers in North America. Indeed, it is difficult to envisage additional studies in this occupational group that would contribute meaningfully to weight of evidence for consistency of the observed association.
One criterion for causality of observed associations in epidemiological studies, namely coherence, may not have been adequately fulfilled, in view of the difference in the specific form of lymphohematopoietic cancer in excess in available investigations for the two principal types of populations of workers studied. Indeed, increases in lymphosarcoma and reticulosarcoma have been observed in monomer production workers, whereas increases in leukemia have been observed in styrene-butadiene rubber workers. Although it is plausible that this difference may be related to variation in the extent of information available for characterization of exposure or to the nature of exposures in the two industries, this has not been systematically investigated. There is also the possibility of misclassification of cause of death on death certificates (although Sathiakumar et al. [1998] did not observe an association with forms of lymphohematopoietic cancer other than leukemia in the large cohort of styrene-butadiene rubber workers when causes of death were examined using current terminology). The potential for transformation of one form of lymphohematopoietic cancer to another (e.g., non-Hodgkin's lymphoma to leukemia) has also been noted (Sathiakumar et al., 1998). In addition, available data for the large study of styrenebutadiene rubber workers were insufficient to determine if butadiene was causally associated with a specific form of leukemia. Moreover, it is noteworthy that these different tumours observed in styrene-butadiene rubber workers and monomer production workers are of the same organ system, and perhaps even share the same pluripotential stem cell.
An association between exposure to butadiene and the induction of leukemia is also biologically plausible. The hematopoietic system is a target for butadiene-induced effects in rodents (i.e., lymphocytic lymphomas [NTP, 1993], cytogenetic effects in bone marrow [Cunningham et al., 1986; Irons et al., 1986a, 1987; Tice et al., 1987; NTP, 1993; Leavens et al., 1997] and suppression of stem cell differentiation [Irons et al., 1996]). Aneuploidy, which is believed to be associated with leukemia in humans, has been induced in human lymphocytes exposed in vitro to the mono- and diepoxide metabolites of butadiene (Vlachodimitropoulos et al., 1997; Xi et al., 1997). Moreover, the presence of relevant metabolizing enzymes in progenitor cells believed to be important targets for the induction of leukemia in humans (i.e., CD34+ cells) has been demonstrated in studies of the metabolism of benzene (a documented human leukemogen) (Schattenberg et al., 1994; Ross et al., 1996b) (although exposure of human CD34+ cells to EB at "physiologically relevant concentrations" did not alter cytokine-induced clonogenic response, an early change frequently observed in the development of leukemia; Irons et al., 1996; Irons, 1998). Therefore, available data also support the biological plausibility of an association between exposure to butadiene and leukemia observed in humans, although the active metabolite has not been identified.
Therefore, although not completely convincing in their own right, the available epidemiological studies of the association between leukemia and exposure to butadiene in occupationally exposed human populations fulfil several of the traditional criteria for causality, including strength of association (RR of 4.2 in the highest exposure group [based on five cases], which would be considered moderately strong), quantifiable exposure-response relationship, temporal relationship (the critical investigation [i.e., Delzell et al., 1995] is a historical cohort study), biological plausibility and, to some degree, consistency, although the criterion for coherence is not fully satisfied.
Assessment of the weight of evidence for carcinogenicity in human populations should not, however, be considered in isolation from the extensive supporting data on carcinogenicity, genotoxicity and inter- and intraspecies variations in metabolism and response. The association between exposure to butadiene and development of cancer is supported by limited evidence of genetic damage in exposed workers, as well as the wealth of evidence that butadiene is carcinogenic and/or genotoxic in all species of experimental animals tested (mice, rats and hamsters), inducing a wide range of tumours and genetic damage at relatively low concentrations in mice (i.e., within the same order of magnitude as current occupational health limits). Moreover, while there are quantitative differences in the potency of the substance to induce tumours in various species, likely related to observed quantitative differences in metabolism, there are indications of considerable interindividual variations in the metabolism of butadiene in the human population, consistent with expectations for a complex metabolic pathway.
Based on the evidence of an association between exposure in the occupational environment and leukemia that fulfils several of the traditional criteria for causality of associations observed in epidemiological studies, supporting limited data on genotoxicity in human populations and the overwhelming weight of evidence of carcinogenicity and genotoxicity at relatively low concentrations in some species of experimental animals, butadiene is considered highly likely to be carcinogenic in humans.
Although relevant data in humans are limited, the results of in vivo studies in experimental animals indicate that butadiene induces mutations in somatic cells and male germ cells as well as male-mediated heritable clastogenic damage. While most of the studies have been conducted in mice, rats appear to be less sensitive to these effects, which is consistent with species differences in metabolism. However, in view of the likely considerable heterogeneity in the metabolism of butadiene in human populations, butadiene is considered a likely human somatic and germ cell genotoxicant.
3.3.2.2 Weight of evidence for non-neoplastic effects
The available data on effects of butadiene other than carcinogenicity or genotoxicity are limited. Based on the limited data available, species differences in the ability of butadiene to induce other non-neoplastic effects again appear to be consistent with variations in metabolism of butadiene to active metabolites. However, butadiene is of low acute toxicity in both rats and mice, in contrast to its ability to induce cancer and genetic damage at relatively low concentrations in mice.
Hematological effects suggestive of macrocytic anemia have been consistently observed in mice (two strains) following shortterm, subchronic or chronic exposure to butadiene at concentrations similar to or lower than those that induced general toxicity (as indicated by decreased body weight gain and increased organ weights) (Irons et al., 1986a, 1986b; NTP, 1993; Bevan et al., 1996). For example, changes in hematological parameters were noted in mice exposed to ≥62.5 ppm (≥138 mg/m3) butadiene for 9 months or longer in the NTP bioassay. Butadiene also induced effects on bone marrow (including atrophy, decreased cellularity, regeneration and alterations in stem cell development) in mice (Irons et al., 1986a, 1986b; Leiderman et al., 1986; NTP, 1993), although available data are inadequate to assess the potential effects on immune system function. While effects on the blood and bone marrow have not been reported in rats in recent investigations (including the only identified chronic bioassay; Hazleton Laboratories Europe Ltd., 1981a), the database is considerably more limited. In addition, the lack of observation of hematotoxicity in rats may again reflect the species differences in metabolism. Although the available epidemiological studies are too limited to assess the hematotoxicity in humans, available data support the hematopoietic system being a critical target for butadiene-induced toxicity, since the lymphohematopoietic system is a target for butadiene-induced leukemia in humans. However, it has not been established if the non-neoplastic effects observed in animals may be preliminary to, or associated with, the development of lymphohematopoietic cancers.
The reproductive organs are also critical targets of butadiene-induced non-neoplastic effects in mice. Ovarian atrophy, the severity and incidence of which increased with concentration or duration of exposure, was observed at all concentrations (i.e., ≥6.25 ppm [≥13.8 mg/m3]) in the chronic bioassay conducted by the NTP (1993); in all exposure groups, the level of degeneration at 2 years, characterized by lack of oocytes, follicles or corpora lutea, was incompatible with reproductive capacity. Although recent re-examination of some of the tissue samples indicated that the atrophy observed in the ovaries may be related to senile changes (Davis, 1998), it may be that butadiene is exacerbating these changes. It should be noted, though, that the incidence of these lesions was increased as early as 9 months (although the slides from these interim sacrifices have not been re-examined). That butadiene is causally associated with these lesions is also difficult to dismiss on the basis of currently available data, in view of the consistency with the results of other studies, including the earlier NTP (1984) bioassay and a subchronic study at higher concentrations (Bevan et al., 1996) in which such lesions were also observed, the presence of a clear dose-response relationship and biological plausibility. Based on the observation of depletion of ovarian follicles and alkylation with ovarian macromolecules in mice following intraperitoneal administration of the monoepoxide or diepoxide metabolite and in rats administered the diepoxide (Doerr et al., 1995), it is possible that the ovarian toxicity is mediated through generation of the active epoxide metabolites.
Testicular atrophy was noted only in male mice exposed to concentrations greater than those that induced effects in females (NTP, 1993). Consistent with metabolic differences, butadiene did not induce ovarian or testicular toxicity in the limited number of available studies in rats, although, as noted above, the diepoxide metabolite was ovotoxic in both species (Doerr et al., 1995, 1996). Although available data are limited, there is no conclusive evidence that butadiene is teratogenic in mice or rats following maternal or paternal exposure or that it induces significant fetal toxicity at concentrations below those that are maternally toxic. Available epidemiological data are inadequate for evaluation of potential reproductive or developmental toxicity; in fact, none of the identified analytical studies was conducted in women. However, in view of the qualitative similarities in the metabolism of butadiene in mice, rats and humans and the likely variation across the general population associated with genetic polymorphism for the relevant enzymes, and on the basis of the observed ovarian toxicity in butadiene-exposed mice, butadiene is considered to be a possible reproductive toxicant in humans, although additional work to clarify the relevance of these observed effects is clearly desirable.
Available data on other systemic or organ-specific effects are inadequate to determine if such effects might be considered critical.
