Styrene - PSL1

2.0 Summary of Information Critical to Assessment of "Toxic"

2.1 Identity, Properties, Production, and Uses

Styrene is a monoaromatic hydrocarbon (CAS No. 100-42-5). Synonyms for styrene include vinylbenzene, vinylbenzol, phenylethylene, styrolene, styrol, styrole, ethenylbenzene, cinnamene, cinnamenol and cinnamol (Bond, 1989; Sax and Lewis, 1989; CCOHS, 1990). The structural formula of styrene is C₆H₅CH = CH₂ . Styrene is a colourless liquid at room temperature, with a vapour pressure of 667 Pa @ 20°C (Verschueren, 1983), a water solubility of 300 mg/L @ 20°C (Verschueren, 1983), a log octanol/water partition coefficient of 2.95 (U.S. EPA, 1984), an estimated organic carbon partition coefficient (K_oc) using the method proposed by Mackay et al. (1992) of approximately 370, and a Henry's Law Constant of 284.72 Pa-m³/mol (Howard, 1989). In air, 1 ppm styrene equals 4.33 µg/m³ (Verschueren, 1983). Styrene absorbs infrared radiation in the 7 to 13 mm wavelength region (Sadtler Research Laboratories, 1982).

Analysis of styrene in air generally includes preconcentration in a cryogenic trap and detection by gas chromatography with a flame ionization or electron capture detector (detection limits range from 0.05 to 0.1 µg/m³ ) [Dann, 1991]. Styrene in surface water has also been analyzed by a purge-and-trap technique with gas chromatography/mass spectrometry (detection limit 0.1 mg/L) [Otson, 1987].

In 1990, Canadian production of styrene was reported to total 718 kt/year; of this amount approximately 490 kt were exported while negligible amounts of styrene were imported (Camford Information Services, 1991a). Industrial uses of styrene in Canada include the manufacture of polystyrene (147 kt/year), styrene-butadiene (SB) latex preparations (34 kt/year), acrylonitrile-butadiene-styrene (ABS) resins (28 kt/year), unsaturated polyester resins (11 kt/year) and SB rubber (5 kt/year). End uses of styrene or styrene-containing materials include foam materials, synthetic rubber products such as automobile tires, plastics, waxes, paints and varnishes, adhesives, metal cleaners and fibrous glass products (Howard, 1989; Camford Information Services, 1991a, 1991b).

2.2 Entry into the Environment

Styrene can be released into the Canadian environment from any stage in its production, storage, transport, use and disposal. The substance can also be released from similar activities involving styrene-containing materials.

There are three plants in Canada producing styrene - two in Ontario and one in Alberta (Environment Canada, 1990). Of the seven polyester production plants in Canada, three are located in Ontario, three in Quebec and one in Alberta. Plants producing other derivatives of styrene, such as acrylonitrile-butadiene-styrene (ABS) resin, styrene-butadiene (SB) rubbers and latex preparations, polyester resins and plastics, are mostly located in Quebec and Ontario, with a few located in Alberta, British Columbia and Nova Scotia.

Data on emissions of styrene in Canada are limited. Estimates based on available emission factors for various industrial activities are summarized in Table 1.

Table 1 Summary of Estimated Emissions of Styrene Resulting from Industrial Activities in Canada (tonnes/year).

1. Environment Canada (1990)
2. Estimated using emission factors presented in U.S. EPA (1990)
3. Edgecombe (1989)

Under the Municipal and Industrial Strategy for Abatement (MISA) program of the Ontario Ministry of the Environment, styrene was detected in the effluent of six industrial sites from the "Organic Chemical Manufacturing sector" (OME, 1992). The highest average concentration of styrene reported for the period October 1979 to September 1990 was 71.1 mg/L; this represents a loading value of 0.511 kg/day for this source.

As well, styrene was detected in 9 of 274 samples of raw sewage; the mean concentration was 21.4 mg/L (OME, 1988). Styrene was detected in 1 of 51 samples of raw sludge (at a concentration of 6 011 mg/kg wet weight) from Ontario municipal water pollution control plants, and in 2 of 262 samples of primary and secondary effluents (at concentrations of 15 and 13 mg/L) [OME, 1988]. It was not detected in treated sludge from 34 monitored plants. The detection limits for this study were 40 and 3 mg/L in raw sewage and effluents, respectively; the detection limit in sludge was not stated.

Other unquantified sources of styrene include: combustion products from spark-ignition engines, oxy-acetylene flames, cigarettes, pyrolysis and cracking of petroleum derivatives, bituminous-coal and shale-oil tars (Royal Society of Chemistry, 1989); emissions from waste incineration (Junk and Ford, 1980); spills (OME, 1991; NATES, 1992); and natural sources, including by-products of fungal and microbial metabolism (Clifford et al., 1969; Harada and Mino, 1973; Shirai and Hisatsuka, 1979; Sato et al., 1988; Shimada et al., 1992).

2.3 Exposure-related Information

2.3.1 Fate

The major processes affecting the fate of styrene in the environment are photo-oxidation, volatilization and biotransformation. Most industrial emissions containing styrene are released directly to the atmosphere.

Hydroxyl radicals and tropospheric ozone are major reactants that rapidly degrade styrene in the atmosphere (Atkinson et al., 1982; Alexander, 1990). Atmospheric half-lives ranging between 3.5 and 9 hours have been reported (Howard, 1989; Alexander, 1990). Styrene absorbs wavelengths of sunlight reaching the Earth's surface poorly (Alexander, 1990); therefore, degradation of styrene by direct photolysis is unlikely. The physical removal of airborne styrene by processes such as wet and dry deposition is thought to be relatively minor, and long-range transport of styrene is considered insignificant, based on its short atmospheric half-life (Alexander, 1990).

Styrene is rapidly lost from surface waters by volatilization, with reported half-lives for this medium ranging from 1 to 60 hours, depending on depth of the water body and degree of turbulence (Santodonato et al., 1980; Zoeteman et al., 1980; U.S. EPA, 1987a, 1987b; Howard, 1989; Fu and Alexander, 1992). The U.S. EPA (1984) estimated, on the basis of computer models, that half-lives of styrene were 3 days in a pond and 13 days in an oligotrophic lake. Longer half-life values are expected for stagnant deep water where volatilization does not occur (Zoeteman et al., 1980; Alexander, 1990). Under laboratory conditions, Fu and Alexander (1992) demonstrated that the concentration of styrene in open flasks of distilled and lake water declined by about 95%, from 4 mg/L to about 0.2 mg/L, after 24 hours. They attributed this loss to volatilization.

Styrene can be biodegraded quite readily in water under aerobic conditions. The biodegradation half-life of styrene in water was estimated to be less than 5 days (Price et al., 1974). Based on the results of studies in which the rate of mineralization of styrene was proportional to its concentration in water, Fu and Alexander (1992) suggested that styrene could persist in low concentrations in water. Wilson et al. (1983) reported that styrene would degrade more slowly in groundwater than in surface waters.

The half-life for volatilization of styrene from soil surfaces was estimated to be approximately 1 minute, with the rate of volatilization decreasing with increasing depth (U.S. EPA, 1987a). Based on its estimated organic carbon partition coefficient (K_oc) of 370, the mobility of styrene in soil is considered to be moderate (McCall et al. 1981). Roberts et al. (1980) found the rate of movement of styrene in a sand aquifer to be approximately 80 times slower than that of a non-adsorbing tracer.

A bioconcentration factor (BCF) in fish of 64 was estimated for styrene using the method presented by Veith et al. (1979).

2.3.2 Concentrations

Styrene has been detected at low concentrations in ambient air, indoor air and drinking water across Canada, and in surface water in the Great Lakes area. It was also detected but not quantified in biota and sediment in Canada. Styrene has also been detected in a range of foods, although available data are limited. Data on concentrations in the marine environment, groundwater, rain, snow, dry deposition, plants, birds, wild mammals and breast milk were not identified.

In extensive studies of ambient air, levels of styrene were determined in 586 samples taken from 1988 to 1990 at 18 mostly urban sites across Canada. Mean concentrations in 24-h samples ranged from 0.09 to 2.35 µg/m³ (limit of detection approximately ≤ 0.1 µg/m³ ) with an overall mean from all sites of 0.59 µg/m³ . Daily maximum concentrations were highest in industrial areas of Toronto (10.19 µg/m³ ) and Vancouver (34.20 µg/m³ ). The highest concentration of styrene reported in rural air in Canada was 3.2 µg/m³ , at Walpole Island, Ontario (Dann, 1990).

The most extensive data on concentrations of styrene in indoor air in Canada are from a national pilot study, conducted in 1991, of 757 single-family dwellings and apartments selected to represent a probability sample of the general population. The 24-h concentrations of styrene across all homes ranged from non-detected (detection limits = 0.48 µg/m³ ) to 128.93 µg/m³ , and averaged 0.28 µg/m³ (Concord Environmental, 1992). Slightly higher (several µg/m³ ) mean levels have been detected in a number of other more limited studies of Canadian residences (Chan et al., 1990; Bell et al., 1991; Otson and Benoit, 1985) in urban and urban-industrial communities. The concentrations in indoor air may reflect releases from such household products as carpet glues and from cigarette smoke (Wallace et al., 1987a, 1987b, 1989), although the contribution of such products to levels in indoor air has not been examined in studies conducted in Canada.

In surveys of Canadian water supplies, styrene is generally present at concentrations of less than 1 mg/L. For example, in the Ontario Drinking Water Surveillance Program, between 1988 and 1990, styrene was detected in 90 of > 3 000 samples from 86 sources of drinking water; mean concentrations in the individual sources in 1990 ranged from not detected (detection limit = 0.050 mg/L) to 0.250 mg/L in treated drinking waters (Lachmaniuk, 1991). Concentrations of styrene in surface waters and municipal water supplies reported in the ENVIRODAT data base for the period 1985 to 1991 were all below the detection limit (0.5 and 1.0 mg/L, respectively) [ENVIRODAT, 1992]. Otson (1987) conducted a survey of raw and treated water from 10 municipalities around the Great Lakes between July 1982 and May 1983, and reported mean concentrations of styrene of £ 0.5 mg/L (detection limit = 0.1 mg/L). The maximum concentration measured in raw water during that survey was 1.7 mg/L, in Cornwall (Otson, 1992).

Based on limited data, background concentrations of styrene in soil are very low. In a survey of organic compounds in soils in uncontaminated urban areas west of Toronto, styrene was detected in 3 of 5 soil samples at Port Credit, Ontario at concentrations of up to 0.2 mg/kg (detection limit = 0.05 mg/kg), and was not detected in any of 8 samples from Oakville/Burlington, Ontario (detection limit = 5 to 10 mg/kg) [Golder Associates, 1987].

Data on concentrations of styrene in Canadian sediments are sparse. Samoiloff et al. (1983) detected, but could not quantify, styrene in sediments from Tobin Lake, Saskatchewan.

Whole body concentrations of styrene ranging between 15 and 100 mg/kg were measured in "Splake", a cross of brook trout and lake trout, and in walleye (Stizostedion vitreum) caught in the St. Clair River (Bonner and Meresz, 1981). Styrene was also detected, but not quantified, in the tissues of several other fish (emerald shiner [Notropis atherinoides]; black crappie [Pomoxis nigromaculatus]; bluegill [Lepomis macrochirus]; pumpkinseed [Lepomis gibbosus]; and walleye [Stizostedion vitreum]) from the St. Clair River. Edible shellfish from Atlantic Canada contained < 10.0 mg styrene/kg (Zenon, 1989). In both of these reports, it was not indicated whether the results were expressed on a wet weight or a dry weight basis.

Styrene has also been found in a wide range of foods, but is generally present at levels of less than 10 mg/kg. In some instances, its presence is the result of the migration from food packaging manufactured from styrene-based polymers or copolymers. There is also evidence that styrene occurs in some foods as a natural constituent: it has been detected, sometimes at extremely high concentrations, in food products with no apparent source of contamination (TNO, 1992; MRI, 1992). In early Canadian studies of foods packaged in polystyrene, the concentration of styrene monomer in the food containers ranged from 809 to 3 019 mg/kg by weight. The mean content of styrene monomer in foods ranged from low values in plain yogurt (trace to 13.0 mg/kg; detection limit = 0.73 mg/kg) to the highest values in sour cream (143.3 to 245.9 mg/kg; detection limit = 13.4 mg/kg) [Withey and Collins, 1978]. In a recent Canadian survey of a much wider range of foods, samples of 34 food groups (each a composite of individual food items, combined in approximate proportion to their consumption in the Nutrition Canada Survey), were collected from retail outlets in Windsor in 1992. Styrene was not detected in any of the 34 food groups (which together approximated an average Canadian diet) by purge-and-trap gas chromatography/mass spectrometry (detection limits were 1.0 mg/L for liquids and 0.005 mg/g [5 mg/kg] for solids) [ETL, 1992].

No data on concentrations of styrene in the breast milk of Canadian women were identified. Styrene was present, but not quantified, however, in all 8 samples of breast milk taken from women living in the urban areas of Bridgeville, Pennsylvania, Bayonne, New Jersey, Jersey City, New Jersey and Baton Rouge, Louisiana (Pellizzari et al., 1982).

Tobacco smoke contributes to indoor air levels of styrene and the body burdens of smokers (Wallace et al., 1986, 1987a). No studies of the styrene content of smoke from Canadian cigarettes were identified, but the U.S. Department of Health and Human Services (1989) reported that styrene was present in mainstream cigarette smoke in the amount of 10 mg/cigarette.

2.4 Kinetics and Metabolism

The toxicokinetics of styrene has been reviewed extensively (HSE, 1981; U.S. EPA, 1988; Bond, 1989). In rats exposed via the oral route, and in rats and humans exposed through inhalation, absorption of styrene is rapid and virtually complete. Dermal absorption is not expected to be an important route of exposure in non-occupational settings. In animals exposed by various routes, styrene is distributed initially to well-perfused organs (particularly the kidney, liver, pancreas and brain), and is then cleared rapidly. Subsequent accumulation in adipose tissue occurs in both animals and humans, but does not persist for extended periods. Styrene is transferred to offspring in animals and humans both transplacentally and via the milk of the mother.

Metabolism of styrene is extensive in both animals and humans, and is initially catalysed by microsomal NADPH-cytochrome P450-dependent monooxygenases, generating reactive epoxides. This step is saturable in rats at airborne concentrations of approximately 200 ppm (866 µg/m³ ) or intraperitoneal doses of 250 mg/kg bw. The major epoxide, styrene-7,8-oxide, is subsequently hydrolysed to styrene glycol, which is further metabolised to mandelic and phenylglyoxylic acids, the principal urinary metabolites in both animals and man. The major site of styrene metabolism in a variety of species is the liver. Styrene-7,8-oxide has been detected at low levels in the blood of humans exposed to styrene (Wigaeus et al., 1983; Löf et al., 1986a, 1986b), although recent preliminary findings (from in vitro studies in tissue samples from 5 individuals) indicate that humans have a lower capacity to form this compound, and also have a higher capacity to metabolize the epoxide, once formed, than either rats or mice (Mendrala et al., 1993).

Metabolites of styrene are excreted rapidly and almost exclusively in urine in animals and humans exposed to low levels. With levels that saturate metabolism in animals, increased amounts of unchanged styrene are excreted in the expired air. Pharmacokinetic models, which have been developed on the basis of experiments with animals, have been used to predict the toxicokinetics of styrene and styrene-7,8-oxide in humans.

2.5 Effects-related Information

2.5.1 Experimental Animals and In Vitro

A wide range of effects has been reported in experimental animals following exposure to styrene. The emphasis in this summary is on those studies for which the lowest effect levels have been reported.

The acute toxicity of styrene is weak in experimental animals following inhalation or ingestion (U.S. EPA, 1988; Bond, 1989). In short-term or subchronic repeated-dose experiments, the effects that are evident at the lowest levels are those on neurological development following inhalation, and on the immune system following administration by the oral route. Thus, Kishi et al. (1992a, 1992b) reported that Wistar rat offspring exposed to airborne styrene in utero (60 ppm [260 µg/m³ ] for 6 hours/day during days 7 to 21 of gestation) had significantly reduced pup weights at day 1, delayed development of righting reflex and auditory startle reflex, and nonsignificantly decreased levels of serotonin and its metabolite 5-hydroxy-indoleacetic acid (5-HT) in the cerebrum. Exposure to 293 ppm (1 270 µg/m³ ) significantly reduced pup body weight and the levels of serotonin and 5-HT, and induced alterations in a wider range of behavioural measures. Shigeta et al. (1989) reported that the development of exploratory and avoidance behaviour, as well as the age at which developmental milestones (pinna detachment, incisor eruption) were achieved, was retarded in THA rats by exposure to 50 ppm (217 µg/m³ ) styrene, 7 hours/day, 6 days/week from birth to 48 days of age. The same developmental milestones, but not performance in behavioural tests, were also affected at a concentration of 25 ppm (108 µg/m³ ). (It is possible, however, that the neonates may not have been at a comparable stage of development at the start of the experiment, as differences in body weight between exposed and control groups were evident on the first day of exposure.) Exposure to 27 to 30 mg/kg bw/day styrene via the oral route for a period of 4 weeks was associated with increased mortality from an oncogenic virus encephalomyocarditis and a rodent strain of malaria (there was also significantly increased infection from malaria at 18 mg/kg bw/day) [Dogra et al., 1992], and cell-mediated, humoral, and macrophage functional responses were altered in mice exposed to as little as 20 mg/kg bw/day for 5 days (Dogra et al., 1989). However, there has not been confirmation of these immunotoxic effects in longer-term studies by additional investigators.

With higher exposures, a wide range of effects has been observed in rodents exposed to styrene levels of between 150 and 450 ppm (650 to 1 950 µg/m³ ) in short-term and subchronic studies, including glutathione depletion in lung and liver (Vainio et al., 1979; Elovaara et al., 1990), microsomal enzyme induction in liver and kidney (Sandell et al., 1978; Vainio et al., 1979), histopathological changes in the respiratory epithelium (Morisset et al., 1979; Ohashi et al., 1985, 1986) and liver (Vainio et al., 1979), effects on the haematopoietic system (Seidel et al., 1990), alterations in the levels of neurochemicals and in the cellular composition of the brain (Savolainen and Pfaffli, 1977; Rosengren and Haglid, 1989), and performance in behavioural testing (Kulig, 1988). In short-term and subchronic studies in which styrene was administered orally or intraperitoneally, exposure to between 100 and 450 mg/kg bw/day altered the activities of metabolizing enzymes in the liver, kidney, and brain of rats (Sandell et al., 1978; Dixit et al., 1982; Srivastava et al., 1982; Das et al., 1981, 1983), induced histopathological changes in the liver of rats and dogs (Quast et al., 1979; Srivastava et al., 1982), and affected several neurological end-points in rats, including levels of neurotransmitters in the brain (Husain et al., 1985), dopaminergic function (Agrawal et al., 1982; Zaidi et al., 1985) and performance in behavioural tests (Zaidi et al., 1985; Husain et al., 1985). (Although Ohashi et al. [1985, 1986] reported that exposure to as little as 30 ppm [130 µg/m³ ] of styrene for several weeks produced mild ultrastructural changes in the nasal mucosa of rats, the observed alterations were not clearly compound-related, and were of uncertain clinical significance. Similarly, Fujita et al. [1987] reported reductions in the activity and concentrations of d-amino-levulinate dehydratase in the erythrocytes and bone marrow of rats following continuous exposure to 48 ppm [210 µg/m³ ] styrene for 1 week; the clinical significance of these results is unclear, however, and there were no effects on body weight and liver weight, the only other end-points examined).

The carcinogenicity of styrene has been examined in rats exposed by various routes, and in studies in mice following oral administration. The following discussion is limited principally to those studies that involved an adequate number of animals exposed to styrene for a sufficient length of time. Even these studies have limitations that preclude firm conclusions being drawn with respect to the carcinogenicity of styrene.

Jersey et al. (1978) reported that in groups of 84 to 86 Sprague-Dawley rats of both sexes exposed to 600 and 1 000 ppm (2 600 and 4 330 µg/m³ ) for 6 hours/day, 5 days/week for 78 to 89 weeks via inhalation, the incidence of leukaemias and lymphosarcomas combined was significantly increased in exposed females compared to historical controls, but not to their concurrent controls; however, it is currently considered inappropriate to combine leukaemias and lymphomas (McConnell et al., 1986), and the incidences of the individual malignancies were not significantly increased. An increased incidence of mammary carcinomas in females was reported, but did not appear to be compound-related, as it was observed only at the low concentration. There were no significant compound-related increases in tumour incidence in males, although these may have been masked by high mortality from murine viral pneumonia in the control and high-dose groups (Jersey et al., 1978).

In a somewhat smaller inhalation study of groups of 30 Sprague-Dawley rats of both sexes exposed to lower concentrations of styrene (up to 300 ppm, or 1 300 µg/m³ , 4 hours/day, 5 days/week for 52 weeks) [Conti et al., 1988], there was an increase in the incidence of malignant mammary tumours in all groups of styrene-exposed female rats, but no clear dose-response relationship. It is not possible to assess whether these tumours were associated with exposure to styrene based on the information included in the report of this inadequately documented study, in which the results of a large number of studies with several compounds were summarized.

NCI (1979) investigated the carcinogenicity of styrene via the oral route in groups of 50 Fischer 344 rats of both sexes with long-term exposure to 500, 1 000 or 2 000 mg/kg bw/day, 5 days per week for 104 to 105 weeks. There were no significant differences in tumour incidence at any site between exposed and control rats. A distinct lack of neoplasia in the high-dose rats was attributed to the early mortality at this dose. The significance of this study is further limited by the small number of controls (20) used.

The incidence of lung tumours has been increased marginally in mice following administration of styrene by gavage in two studies. In one of these investigations (NCI, 1979), groups of 50 B6C3F₁ mice of each sex were exposed to 150 or 300 mg/kg bw styrene, 5 days/week for 78 weeks. There was a significant increase in the incidence of lung adenomas and carcinomas combined in the high-dose males when compared to the concurrent controls, but not to untreated historical controls. Interpretation of this study is limited by the small number (20) of concurrent controls. The authors questioned the significance of this increase, noting the unusual absence of these tumours in the concurrent controls; however, none was also observed in a larger group of 40 historical vehicle controls.

An increased incidence in lung adenomas and carcinomas combined was also reported by Ponomarkov and Tomatis (1978) in groups of 45 male and 39 female 0₂₀ mice progeny exposed to styrene in utero (dams received 1 350 mg/kg bw by gavage on day 17 of gestation), and then by gavage to the same dose weekly for 16 weeks after weaning; this exposure is considered to have exceeded the maximum tolerated dose. The increase was significant in both sexes compared to vehicle controls, and in females compared to untreated controls. Lung tumours reportedly developed earlier in styrene-exposed mice, but it may be that exposed animals, many of which died early in the experiment, were simply being examined earlier than the controls. Interpretation of this experiment is also limited by the fact that the data were not analyzed by litter.

Forestomach papillomas and carcinomas have been consistently observed in rats and mice following oral exposure to styrene-7,8-oxide, the putative genotoxic and carcinogenic metabolite of styrene in mammals (Maltoni et al., 1979; Ponomarkov et al., 1984; Lijinsky, 1986; Conti et al., 1988). The relevance of these findings to humans is uncertain, however; the neoplasms were accompanied by squamous cell hyperplasia and/or hyperkeratosis of the forestomach, suggesting that these tumours may have been associated with tissue damage at the high doses employed in these studies. Mechanistic studies of these neoplasia, including DNA-binding and cellular proliferation, were under way at the time of completion of this assessment (Science and Technology Task Group of the Styrene Information Research Center, 1991).

The genotoxicity of styrene has been examined in a number of in vitro and in vivo assays and has been recently reviewed by Barale (1991) and Preston (1990a, 1990b). In the absence of metabolic activation, styrene has not been genotoxic in in vitro assays for gene mutation (bacterial and mammalian systems), clastogenicity or DNA damage. (This contrasts with the clear genotoxic activity in these assays of the primary metabolite of styrene, styrene-7,8-oxide, and indicates that any genotoxic activity of styrene will be dependent upon appropriate metabolic activation.) With metabolic activation, styrene has been either weakly, or not, mutagenic, likely reflecting variations in the balance between the production and inactivation of metabolically activated species. Despite some inconsistency in results, styrene should be considered an in vitro clastogen capable of inducing chromosomal aberrations and micronuclei in mammalian-cell systems. In addition, styrene has been shown to induce sister chromatid exchanges (SCE) in Chinese hamster ovary cells in the presence of metabolic activation and to bind DNA and induce DNA repair in isolated rat hepatocytes.

In in vivo tests, styrene induced SCE and chromosomal aberrations in bone marrow cells, and SCE in splenocytes, regenerating liver cells and alveolar macrophages of mice following inhalation exposure. Intraperitoneal injection of styrene increased the levels of SCE and micronuclei in bone-marrow cells of mice, and of SCE in splenocytes of mice and rats. No conclusions can be drawn regarding the genotoxic activity of styrene in mice following acute or subchronic oral administration, since both of the available studies were flawed (Loprieno et al., 1978; Sbrana et al., 1983). Styrene has also been weakly active in sperm morphology assays in rats and mice treated by intraperitoneal injection, supporting the conclusion that styrene is genotoxic in vivo.

In the most comprehensive reproduction study (Beliles et al., 1985), the reproductive system of male rats was not affected by chronic exposure to 14 mg/kg bw/day styrene in drinking water, although in other studies, testicular histopathology and biochemistry, and sperm count, have been affected at higher doses; however, Beliles et al. (1985) observed that pup survival or weight was slightly reduced at some times in each of three generations exposed to styrene (250 ppm [mg/L] in drinking water [14 and 21 mg/kg bw/day for males and females, respectively]) over the lifespan. The no-observed-effect-level (NOEL) in this study was 125 mg/L in drinking water (7.7 and 12 mg/kg bw/day for males and females, respectively).

The results of most available studies do not indicate that styrene is a developmental toxicant at concentrations or doses less than those that are maternally toxic; however, in one study rat pups exposed to airborne styrene in utero had reduced body weights and slight alterations in behaviour and neurotransmitter levels in brain at airborne concentrations [60 and 293 ppm (260 and 1 300 µg/m³ ) for 6 hours/day on days 7 to 21 of gestation] that did not affect the dams (Kishi et al., 1992a, 1992b).

2.5.2 Humans

Irritation, prenarcotic symptoms, and altered coordination have been commonly reported in workers and volunteers exposed to between 10 and 100 ppm (43 to 433 µg/m³ ) and more styrene in air. There are also indications of increased prevalence of abnormal EEG patterns, reduced peripheral nerve conduction velocity (especially in sensory nerves) and effects on the neuroendocrine system in workers exposed to atmospheric concentrations of 50 to 100 ppm (217 to 433 µg/m³ ) styrene (Seppäläinen and Härkönen, 1976; Rosén et al., 1978; Mutti et al., 1984a; Cherry and Gautrin, 1990; Murata et al., 1991). In neuropsychological studies, the principal observations are a slowing of reaction time, especially in workers who have been exposed for more than a few weeks to concentrations of styrene of 50 to 100 ppm (217 to 433 µg/m³ ) [Gamberale et al., 1976; Cherry et al., 1980, 1981; Cherry and Gautrin, 1990]. More subtle effects, such as reductions in visuomotor accuracy and verbal learning skills, and subclinical effects on colour vision, appear to occur at lower concentrations (i.e., 25 to 50 ppm [108 to 217 µg/m³ ]) in some studies (Härkonen, 1978; Mutti et al., 1984b; Gobba et al., 1991; Fallas et al., 1992), although the extent of historical exposures and exposures to other substances is uncertain. The neurological effects observed are characteristic of occupational exposure to other organic solvents such as toluene or white spirit, which does not involve lasting nerve damage, does not lead to progressive impairment of nervous system function and is generally reversible on cessation of exposure.

The potential carcinogenicity of styrene has been investigated in a number of historical cohort and case-control studies in populations exposed to styrene and related chemicals in the manufacture of styrene-butadiene rubber, styrene and/or polystyrene, or fibrous glass products. Mortality from cancers of the lymphatic and haematopoietic system has been of particular interest, based on an early report of a significantly elevated risk ratio for this family of malignancies (4 cases, RR = 6.2, 95% confidence interval (CI) 4.1 to 12.5) among workers in the synthetic rubber plant at a United States tire manufacturing factory (McMichael et al., 1976).

In the most sensitive study, Matanoski et al. (1990) reported the mortality experience in a cohort of 12 110 male workers involved in styrene-butadiene polymer production for at least 1 year. The vital status was known for 96.6% of this cohort, who had an average duration of follow-up of 20.8 years. (In addition to styrene and butadiene, compounds to which workers were exposed would potentially have included thiocarbamates, diphenylamines, mercaptans, hydroquinones, extender oils and carbon black.) Mortality from the major causes of death (2 441 deaths in all) and for most cancer sites was generally lower than expected for the general population. Deaths from lymphopoietic cancers were not significantly elevated in the cohort as a whole (55 observed, Standardized Mortality Ratio (SMR) = 0.97, 95% CI 0.73 to 1.26), but occurred more frequently than expected among production workers (19 observed, SMR = 1.46, 95% CI 0.88 to 2.27), a difference due primarily to a significant excess in Black production workers (6 observed, SMR = 5.07, 95% CI 1.87 to 11.07). Significant excesses were observed in the subcategories of leukaemia in Black production workers and in other lymphatic cancers (non-Hodgkin's lymphoma and multiple myeloma) in production workers as a whole. The SMRs were not clearly related to either duration of employment or years since first employment. In a nested case-control study of 59 lymphopoietic cancer cases, there was a significantly increased risk of mortality from leukaemia associated with an index of cumulative exposure to butadiene (26 cases, odds ratio (OR) = 9.36, 95% CI 2.05 to 22.9), and a nonsignificant increase for styrene (OR 3.13, 95% CI = 0.84 to 11.2). There was a significant correlation between log cumulated exposure to butadiene and to styrene; however, based on conditional logistic regression analysis, only butadiene was associated with an increased risk of leukaemia (Matanoski et al., 1989; Santos-Burgoa et al., 1992).

For 2 756 workers in two styrene-butadiene rubber plants in the U.S., Meinhardt et al. (1978, 1982) reported that mortality for the cohort as a whole was similar to that for the U.S. general population. (In addition to styrene, the workers were also exposed to butadiene and benzene.) A slight excess of mortality was observed, however, for cancers of the lymphatic and haematopoietic system (9 observed, SMR = 212, 0.05 < p < 0.10), and, in the subset of this category, leukaemia and aleukaemia (5 observed, SMR = 278, 0.05 < p < 0.10) in 600 workers employed during a period when "hot temperature batch polymerization" was used to produce styrene-butadiene rubber.

Bond et al. (1992) conducted a historical cohort study of the mortality of 2 904 male workers (99.6% traced, average duration of follow-up = 30.9 years) employed in the development and manufacture of styrene products at United States Dow Chemical plants. When mortality for the cohort (687 deaths in all) was compared to that of a reference group of Michigan-based Dow employees, there was a significant excess of multiple myeloma for the cohort (Observed:Expected (O:E) = 7:2.9, relative risk 2.45, 95% CI 1.07 to 5.65). There was a significant increase in mortality from lymphatic and haematopoietic cancer among workers exposed to an estimated 8-h time-weighted average (TWA) concentration of 1 to 4 ppm styrene and ethylbenzene (O:E = 12:5.1, SMR = 236, 95% CI 122 to 411), but this occurred primarily in workers with less than 5 years exposure. Most of these neoplasms occurred in workers involved in polymerization, colouring and extrusion, who would have been exposed to extrusion fumes, solvents (styrene, ethylbenzene, acrylonitrile) and colourants.

In a study of 622 workers in a styrene-production and polymerization plant in the U.K., among the 34 deaths from all causes, three deaths from non-Hodgkins lymphoma were reported, compared to 0.56 expected for the general population (p = 0.02), and one subject whose cause of death was reported as heart disease also had leukaemia (Hodgson and Jones, 1985). There was no apparent association between length of service in styrene-exposed jobs and the incidence of lymphatic and haematopoietic cancer. The extent of exposure to styrene and other chemicals is also unknown.

In two earlier studies of workers involved in the manufacture of styrene/polystyrene, there were only single deaths from individual lymphopoietic malignancies in a cohort of 560 workers (83 deaths in all) at a United States plant (Nicholson et al., 1978), and in 1 960 employees (73 deaths in all) at a plant in the Federal Republic of Germany (Frentzel-Beyme et al., 1978).

Some of the highest occupational exposures to styrene occur in jobs that involve the manufacture of glass-reinforced plastics; to date, however, insufficient time may have elapsed for the development of cancers in studies in this relatively new industry. Okun et al. (1985) reported that there were no significant excesses of death (176 in total) from any specific cause in workers with either "high" or "minimal" exposure, and no deaths from any malignant neoplasm of the lymphatic or haematopoietic system, in a cohort of 5 021 workers at two boat-building facilities in the United States; however, the average length of follow-up was only 8.2 years, and the power of the study to detect a two-fold excess risk was only 14% for leukaemia and 15% for lymphoma.

In a more extensive historical cohort study of workers exposed to styrene in the manufacture of glass-reinforced plastics, Coggon et al. (1987) examined the mortality of 7 949 men and women (96% traced) employed at 8 British companies. At the 7 companies where follow-up was reasonably complete, the overall mortality (637 deaths) was less than that for the general population of the U.K. Deaths from lymphatic and haematopoietic cancers occurred less frequently than expected (O:E = 6:14.9; there were, in addition, 8 cases who died from other causes or were alive at the end of follow-up), and only 1 death from this family of neoplasms occurred in a subject with high exposure. The authors cautioned that the study had only limited power to detect cancer with a long latency.

Wong (1990) carried out an historical cohort study of 15 908 men and women who had potentially been exposed to styrene in the reinforced plastics industry. There was no significant excess in mortality from any cause (452 deaths). When the cohort was divided into groups according to their estimated exposure to styrene (separate analyses for max TWA > 20 ppm (> 87 µg/m³ ) vs < 20 ppm, or average TWA > 12 ppm (> 52 µg/m³ ) vs < 12 ppm), the incidence of leukaemia and aleukaemia was higher in the high-exposure group than in the low-exposure group, but the difference between the two groups was not statistically significant, and was based on only 5 deaths in all. Excesses in respiratory cancers were observed in some subsets of the cohort, but in a nested case-control study of 40 respiratory-cancer cases, there was a significant association with smoking, and not with manufacturing-process type or direct exposure to styrene. It should be noted that there was a high loss to follow-up (16.1%), the number of observed deaths from several causes was quite small, the average duration of follow-up was short (7.7 years), and the participants were quite young and were employed for only a short while. The results of a recent update of this study, sponsored by the Styrene Information and Research Center, were not available at the time of completion of this assessment.

In some community-based and nested case-control studies, nonsignificant excesses of various lymphatic and haematopoietic cancers have been associated with exposure to styrene (Flodin et al., 1986; Ott et al., 1989; Siemiatycki, 1991), although these results were all based on very small numbers of cases for any specific type of malignancy (i.e., between 1 and 4).

Cytogenetic studies of workers exposed to styrene and related chemicals have been reviewed by Preston (1990b) and Barale (1991). The available data indicate that exposure to high levels of styrene (approximately 50 to 100 ppm (217 to 433 µg/m³ ) and other chemicals in the workplace may be associated with chromosomal alterations (most often aberrations) in the peripheral lymphocytes of exposed workers. There are also three independent reports that the prevalence of micronuclei is increased in workers exposed to about 10 to 25 ppm styrene (43 to 108 µg/m³ ) [Högstedt et al., 1983; Nordenson and Beckman, 1984; Brenner et al., 1991]. A number of other studies have failed to find evidence of cytogenetic effects at these levels. All of the available studies are limited by one or more of the following factors: small worker populations, uncertainty with respect to the relation between current exposure and exposure at the time when the genetic damage may have occurred (this may account for the frequent lack of dose-response), the high and variable background levels of chromosome aberrations (de Jong et al., 1988) and other potentially confounding factors (such as smoking) that may have contributed to observed effects.

In well-documented studies in the United States and in Finland (Härkönen and Holmberg, 1982; Lemasters et al., 1985), there was no association between exposure to styrene and menstrual disorders, in contrast to the results of earlier Russian studies, which were inadequately reported (Brown, 1991). An increased prevalence of abnormal sperm was reported in workers exposed to styrene (Jelnes, 1988), although these results were based on a small number of subjects, an inadequate number of samples per worker, and questionable controls (i.e., from an infertility clinic). There was no significant association of paternal exposure to styrene and reproductive outcome in a case-control study of men exposed to solvents in the workplace (Taskinen et al., 1989), but the numbers of styrene-exposed cases were small. A slightly increased risk of spontaneous abortion was reported in Canadian women employed in the processing of polystyrene, but this observation was based on small numbers of cases whose exposures were poorly characterized (McDonald et al., 1988). Initial indications that exposure to styrene was associated with CNS defects and spontaneous abortion in Finnish workers (Holmberg, 1977; Hemminki et al., 1980) have not been confirmed in either follow-up or independent studies (Holmberg, 1979; Härkönen and Holmberg, 1982; Holmberg et al., 1982; Härkönen et al., 1984; Hemminki et al., 1984; Lindbohm et al., 1985; Holmberg et al., 1986).

A number of reports indicate that exposure to styrene in the workplace is associated with effects on the liver, kidney, blood system, and lung (reviewed in U.S. EPA, 1988). There is no clear evidence from these studies, however, that styrene has adverse effects on these organs; the effects reported are generally mild and are often not consistent among studies, and possible confounders have not generally been taken into account.

2.5.3 Ecotoxicology

The information that was identified on the toxicity of styrene to aquatic organisms was restricted to acute studies for a number of trophic levels from bacteria and algae through to fish. All studies were conducted under open and static conditions, and the results were expressed in terms of nominal or initial concentrations. Because of its high volatility, styrene disappears rapidly from open systems and, therefore, the concentrations at which effects occurred would have been considerably less than those reported. Studies on the effects of chronic exposure to styrene on aquatic organisms, amphibians and terrestrial wildlife have not been identified.

The lowest concentration of styrene reported to cause an adverse effect in microorganisms was 5.4 mg/L (a 5-min EC₅₀ for reduction of light emitted) for the bacterium Photobacterium phosphoreum (Qureshi et al., 1982). Levels at which adverse effects were observed in algae, bacteria and protozoan species exposed to styrene in solution ranged between 67 and > 256 mg/L (Bringmann and Kühn, 1978, 1980; Bringmann et al., 1980).

Acute 24-h LC₅₀ values identified for rainbow trout (Oncorhynchus mykiss), and the marine sheepshead minnow (Cyprinodon variegatus), were 2.5 mg/L (Qureshi et al., 1982) and 9.1 mg/L (Heitmuller et al., 1981), respectively. Acute 24-96-h LC₅₀ values for five other species of fish ranged from 25 mg/L for the goldfish (Carassius auratus) [Jensen, 1978] to 74.83 mg/L for the guppy (Poecilia reticulata) [Pickering and Henderson, 1966].

The lowest no-observed-effect-concentration (NOEC) identified for mortality in aquatic invertebrates was < 6.8 mg/L for daphnids (LeBlanc, 1980); the 48-h LC₅₀ was reported to be 23 mg/L. In a study with the amphipod (Pontoporeia affinis), exposure to concentrations of styrene of between 35 and 45 mg/L caused an immediate cessation of swimming that lasted several days (Lindström and Lindström, 1980).

Only one study concerning the phytotoxicity of styrene to a terrestrial plant has been identified. Linnainmaa et al. (1978) reported that chromosomal abnormalities were observed in root tip cells of onion (Allium cepa), following exposure to 450 mg/L styrene or styrene oxide in distilled water.