Toxicological Risks of Selected Flame-Retardant Chemicals (2000)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

ANTIMONY TRIOXIDE 229 10 Antimony Trioxide THIS chapter reviews the physical and chemical properties, toxicokinetics, toxicological, epidemiological, and exposure data on antimony trioxide. The subcommittee used that information to characterize the health risk from exposure to antimony trioxide. The subcommittee also identified data gaps and recommended research relevant for determining the health risk from exposure to antimony trioxide. PHYSICAL AND CHEMICAL PROPERTIES The physical and chemical properties of antimony trioxide are summarized in Table 10–1. OCCURRENCE AND USE Antimony trioxide is formed by reacting antimony trichloride (SbCl3) with water. It is used in combination with some brominated flame retardants, and might also be used in conjunction with zinc borate, both within and outside the United States on commercial furniture, draperies, wall coverings, and carpets (R.C.Kidder, Flame Retardant Chemical Association, unpublished material, April 21, 1998). It is also used in enamels, glasses, rubber, plastics, adhesives, textiles, paper, and as a paint pigment (Budavari et al. 1989). 

ANTIMONY TRIOXIDE 230 TABLE 10–1 Physical and Chemical Properties of Antimony Trioxide Characteristic Value Reference Chemical formula Sb2O3 Lide 1991–1992 CAS registry # 1309–64–4 Lide 1991–1992 Synonyms diantimony trioxide, flowers of antimony Budavari et al. 1989 Molecular weight 291.5 g/mol Lide 1991–1992 Physical state Crystal Budavari et al. 1989 Solubility Very slightly soluble in cold water; slightly soluble in hot water; Lide 1991–1992 soluble in KOH, HCl, tartaric acid, and acetic acid Vapor pressure 1×10−10 mm Hg Calculated from Lide 1991–1992 Melting point 656 °C Lide 1991–1992 Boiling point 1,550 °C Lide 1991–1992 Density 5.2 g/cm3 (senarmontite), 5.7 g/cm3 (valentinite) Lide 1991–1992 TOXICOKINETICS Absorption Systemic toxicity and death occurred in rabbits following dermal application of 8g/kg antimony trioxide (Myers et al. 1978), and application of an unspecified dose of antimony trioxide in a paste of “artificial acidic or alkalinic sweat” (Fleming 1938). Both studies indicate that antimony trioxide is absorbed dermally in rabbits. Elevated blood and urine antimony levels were reported in workers occupationally exposed to antimony, suggesting that antimony trioxide is absorbed following inhalation exposure (Cooper et al. 1968; Lüdersdorf et al. 1987; Kim et al. 1997). However, no quantitative correlation was found between the air concentrations of antimony and the antimony concentration measured in urine (Kim et al. 1997). Few quantitative data were found regarding the absorption of antimony trioxide following oral exposure. The International Commission on Radiological Protection (ICRP 1981) has recommended that a 1% absorption rate of antimony compounds (including antimony trioxide) be assumed when estimat 

ANTIMONY TRIOXIDE 231 ing exposure from the gastrointestinal (GI) tract. That recommendation is based on studies of various organic and inorganic antimony compounds. Toxicity is greater following exposure to 7.9 mg antimony trioxide/kg-d in 5% citric acid than to 101 mg antimony trioxide/kg-d in water, suggesting that solubility can affect antimony absorption (Fleming 1938). Distribution No studies were identified on the tissue distribution of antimony trioxide following dermal exposure. Retired workers occupationally exposed by the inhalation route to antimony were reported to have elevated concentrations of antimony in their lung tissue as compared to non-occupationally exposed individuals (Gerhardsson et al. 1982). Following intratracheal instillation of a single dose of 1.52 mg antimony trioxide/kg in Syrian golden hamsters, the highest concentrations of antimony were measured in the lungs and liver, with lower concentrations present in the kidney, stomach, and trachea (Leffler et al. 1984). No information was found on the tissue distribution of antimony in humans following oral exposure. In rats, high concentrations of antimony were measured in the thyroid and GI contents following chronic ingestion of 2% antimony trioxide in the feed (Gross et al. 1955a). Detectable levels were also found in the spleen, kidney, heart, bone, muscle, lungs, liver, and GI tissue. Following continuous treatment of rats for 40 d. Antimony was concentrated in the thyroid, with much lower levels found in the other tissues 40 d after cessation of chronic ingestion of 2% antimony trioxide in the feed (Gross et al. 1955a). Metabolism and Excretion No data were identified on the metabolism or excretion of antimony trioxide following dermal exposure. Intraperitoneal injection of rats with 800 µg antimony chloride/kg did not result in detectable levels of any organic form of antimony in the bile or urine, indicating that antimony is not methylated in vivo. Antimony can form a complex with glutathione in vivo (Bailly et al. 1991). McCallum (1963) reported elevated antimony concentrations in the urine of workers occupationally exposed via inhalation to antimony, indicating that excretion by this pathway occurs in humans. 

ANTIMONY TRIOXIDE 232 Gerhardsson et al. (1982) measured post-mortem antimony levels in the lung tissue of workers occupationally exposed via inhalation to antimony. Levels were found to be elevated compared to a control population, even after workers had been retired for up to 20 yr, indicating a long half-life for lung clearance of antimony in humans. Toxicokinetic studies in adult male Syrian golden hamsters given a single, intratracheal instillation of antimony trioxide (1.52 mg/kg body weight) indicate that 20% of the instilled antimony was cleared from the lung in the first 20 hr (Leffler et al. 1984). Biological half-times of about 40 hr for the initial phase and 20–40 d for the second phase were calculated for lung tissue (Leffler et al. 1984). In rats exposed to 119 mg antimony trioxide dust/m3 for 80 hr, the majority of urinary excretion occurred within the first 3 d after exposure (Gross et al. 1955a). Following a single oral dose (200 mg antimony trioxide) of antimony trioxide to rats, 3% of the administered dose was recovered in the urine within 8 d. Only 0.15% was recovered 1 d after treatment, and 3% was recovered between d 2 and 5 post-treatment (Gross et al. 1955a). Following chronic exposure (2% antimony trioxide in the diet; 8 mo), approximately 99% of fecal excretion and the majority of urinary excretion occurred within 7 d after exposure ceased (Gross et al. 1955a). The large amount of antimony excreted in the feces soon after exposure suggests that a substantial portion of the compound is excreted without being absorbed systemically. That is consistent with the low absorption rate (1%) cited by the ICRP (ICRP 1981) (see Absorption section). HAZARD IDENTIFICATION1 Dermal Exposure Irritation Dermatitis was reported in workers occupationally exposed to 0.4–70.7 mg antimony/m3 (Renes 1953; McCallum 1963; Potkonjak and Pavlovich 1983; White et al. 1993). Although antimony trioxide in the work environment was believed to be responsible for the dermatitis, quantitative data on dermal exposure were not available, and the workers were also exposed to other elements such as arsenic. Therefore, the causative agent for the observed dermatitis could not be positively determined. 1In this section, the subcommittee reviewed toxicity data on antimony trioxide, including the toxicity assessment prepared by the U.S. Consumer Product Safety Commission (Hatlelid 1999). 

ANTIMONY TRIOXIDE 233 In a controlled human study (Industrial Bio-Test Laboratories, Inc. 1973), 52 subjects received a series of nine dermal applications of antimony trioxide over a 3-wk period. The antimony trioxide was applied for 24 hr; the dose was not reported. Two wk after the series of applications, a single dose of antimony trioxide was applied. After each application, skin reactions were evaluated. No skin reactions were observed over the course of the study, suggesting that antimony trioxide is neither a skin irritant nor a sensitizer. Dermal exposure to antimony trioxide generally did not cause dermatitis in tested animals. Only mild skin irritation was observed even after repeated or prolonged exposure to large quantities of antimony trioxide (2–25 g antimony trioxide/kg) in rabbits (Gross et al. 1955a; Ebbens 1972). Skin edema was reported in one study in which antimony trioxide was applied to rabbits in corn oil (8 g antimony trioxide/kg for 24 hr) (Myers et al. 1978). However, that study is limited in that there was no solvent control group, and data on severity and number of animals responding was lacking. In a study by Haskell Laboratory (Haskell Laboratory 1970a), a suspension of 12, 31, or 61 mg antimony trioxide/kg in a fat/acetone/dioxane mixture was applied to intact shaved skin (all doses) or abraded skin (31-mg/kg group only) of 10 albino guinea pigs. The exposure duration was not reported. Irritation was not seen in any of the treated animals. In another study by Haskell Laboratory (1970b), 24 or 49 mg/kg antimony trioxide (suspended in a similar mixture as above) was applied to the intact shaved skin of guinea pigs. One day after the treatment, mild erythema was observed in 2/10 and 5/10 animals treated with 24 mg antimony trioxide/kg and 49 mg antimony trioxide/kg, respectively. All of the responses had disappeared 2 d after the initial dosing. Sensitization As mentioned under the Irritation section, no skin reactions were observed in the controlled human study conducted by Industrial Bio-Test Laboratories, Inc. (1973), indicating that antimony trioxide is not a skin sensitizer. Haskell Laboratory (1970a, b) treated groups of five guinea pigs with nine dermal applications of 31 mg antimony trioxide/kg (25%) or 49 mg antimony trioxide/kg (50%) in a fat/acetone/dioxane mixture on shaved and abraded skin, or four intradermal injections of 1 mg antimony trioxide in either acetonedimethyl phthalate or propylene glycol solutions, over the course of 3 wk. After a 2-wk rest period, each group of animals received challenge applications of the suspensions on both intact and abraded skin. Sensitization was not observed in any of the test animals. 

ANTIMONY TRIOXIDE 234 Systemic Effects Death occurred in one out of four rabbits following a single dermal exposure to 8 g/kg antimony trioxide (Myers et al. 1978), and in one out of four rabbits exposed to 2 g/kg antimony trioxide (Ebbens 1972). Systemic toxicity and death occurred in three out of eight rabbits, but not in rats, following short-term exposure (20–21 d) to an unspecified dose of antimony trioxide (Fleming 1938). Gross pathologies were seen in the liver, lung, stomach, and kidney. Other Systemic Effects No studies were identified that investigated the immunological, neurological, reproductive, developmental, or carcinogenic effects of antimony trioxide following dermal exposure to antimony trioxide. Inhalation Exposure Systemic Effects In humans, the lungs are the primary targets following inhalation exposure to antimony trioxide. Several studies of antimony smelter workers show that workers developed pneumoconiosis, chronic cough, and upper airway inflammation following chronic exposure to antimony trioxide (McCallum 1963, 1967; Cooper et al. 1968; Potkonjak and Pavlovich 1983). In addition, one study reported systemic effects following inhalation exposure in smelter workers, including weight loss, nausea, vomiting, nerve tenderness, and tingling (Renes 1953). In those studies, however, a causal role for antimony trioxide in the observed human health effects could not be confirmed because of the lack of individual exposure data for the workers and exposure to other compounds, including arsenic, lead, and alkali, that could be confounders. The lungs are also the primary target tissues in animals following inhalation exposure (see Table 10–2). All experimental inhalation studies were conducted using whole-body exposure. Details of particle size and purity are provided in footnotes. Guinea pigs exposed to antimony trioxide2 (average concentration: 45.4 mg antimony trioxide/m3, 2–3 hr/d, 6 mo) developed pneumonitis, liver and spleen effects, and decreased white blood cell counts (Dernehl 1945). Similarly, pneumonia was seen following exposure of rats (100–125 mg antimony trioxide/m3, 100 hr/mo, 14.5 mo) and rabbits (89 mg antimony 2Assumed particle size <1 µm and 99.8% pure based on production technique. 

ANTIMONY TRIOXIDE 235 TABLE 10–2 Toxic Effects of Antimony Trioxide Following Inhalation Exposure Species, Exposurea Duration- Effects NOAEL/ RDDR HEC Reference Strain, (mg/m3); adjusted LOAEL NOAEL/ Number Duration Concentrations (mg/m3) LOAEL (mg/m3)b Rat, NS, 100–125; 13.9–17.4 Diffuse, LOAEL: ND ND Gross et al. 50 100 hr/mo interstitial 13.9 1955b for 14.5 mo fibrosis; 18% died of pneumonia Rabbit, 89; 100 hr/ 12.4 85% died of LOAEL: ND ND Gross et al. NS, 20 mo for 10 mo pneumonia 12.4 1955b Guinea 45.4 3.8–5.7 Pneumonitis, LOAEL: ND ND Dernehl pig, NS, (average); 2– liver and 3.8 1945 24 3 hr/d, 7 d/ spleen effects, wk for 6 mo decreased white blood cell counts Rat, 45.5; 7 hr/d, 9.4 Interstitial LOAEL: ND LOAEL: 5 Groth et al. Wistar, 5 d/wk for fibrosis, 9.4 1986 90/sex 52 wk; 20- alveolar-wall wk cell observation hypertrophy and hyperplasia, and cuboidal and columnar cell metaplasia of the lungs Rat, 0.25, 1.08, 0.04, 0.19, 0.88, 6% decrease NOAEL: 0.324 NOAEL: Newton et Fischer 4.92, or 4.19 in body weight 0.88 (M) 0.585 0.006 al. 1994 344, 55/ 23.46; 6 hr/ Increased lung LOAEL: ND LOAEL: Newton et sex d, 5 d/wk for weight 4.19 (M) 0.51 al. 1994 13 wk; 27- NOAEL: ND wk 0.19 observation LOAEL: 0.88 

ANTIMONY TRIOXIDE 236 Species, Exposurea Duration- Effects NOAEL/ RDDR HEC Reference Strain, (mg/m3); adjusted LOAEL NOAEL/ Number Duration Concentrations (mg/m3) LOAEL (mg/m3)b Chronic NOAEL: ND ND Newton et interstitial 0.19 (F) al. 1994 inflammation, 0.88 (M) granulomatous LOAEL: inflammation, 0.88 (F) increased 4.19 (M) alveolar macrophages Rat, 0.06, 0.51 or 0.01, 0.09, 0.80 Increased LOAEL: 0.435 LOAEL: Newton et Fischer 4.5; 6 hr/d, 5 alveolar 0.01 0.004 al. 1994 344, 65/ d/wk for 1 macrophage sex) yr; 1-yr observation Interstitial NOAEL: 0.435 NOAEL: Newton et inflammation 0.09 0.46 0.039 al. 1994 and LOAEL: LOAEL: Newton et granulomatous 0.80 0.348 al. 1994 inflammation BMC (F): BMC(F): 0.16 0.074 Rat, 1.9, 5.0; 6 hr/ 0.3, 0.9 Discoloration LOAEL: ND ND Watt 1983 Wistar, d, 5 d/wk for and increased 0.3 50F 1 yr alveolar macrophages 

ANTIMONY TRIOXIDE 237 Interstitial NOAEL: 0.9 ND ND Watt 1983 inflammation and LOAEL: 0.3 granulomatous inflammation Swine, 1.9, 5.0; 6 hr/d, 0.3, 0.9 Minimal lung fibrosis ND ND ND Watt 1983 Sinclair S-1, 5 d/wk for 1 yr 3F Rat, NS, 6–7 0.027, 0.082, 0.027, Pre-implantation NOAEL: 0.027 ND ND Grin et al. 0.27; 24 hr/d 0.082, 0.27 loss, fetal growth LOAEL: 0.082 1987 for 21 d of retardation, and pre- gestation and post- implantation embryo death BMC, benchmark concentration; F, female; HEC, human equivalent concentration; LOAEL, lowest-observed-adverse-effect level; M, male; ND, not determined; NOAEL, no-observed-adverse-effect level; RDDR, regional deposited dose ratio. aAll inhalation exposures were whole body. bConcentrations are adjusted for continuous exposure. 

ANTIMONY TRIOXIDE 238 trioxide/m3, 100 hr/mo, 10 mo) to antimony trioxide3 (Gross et al. 1955b). Interstitial flbrosis, hypertrophy, and hyperplasia were seen in male and female Wistar rats (90/sex-group) exposed to antimony trioxide4 (45.5 mg antimony trioxide/m3, 7 hr/d, 5 d/wk for 1 yr, followed by a 20-wk observation period) (Groth et al. 1986); those effects were more pronounced in the females. Watt (1983) investigated the effects of exposure to antimony trioxide5 (1.6 or 4.2 mg Sb/m3, equivalent to 1.9 or 5.0 mg antimony trioxide/m3, 6 hr/d, 5 d/wk for 1 yr) in female CDF Fischer rats (148 animals divided into three dose groups) and Sinclair S-1 miniature swine (eight animals divided into three dose groups). In rats, blood urea nitrogen (BUN) was consistently elevated at the high concentration, but was statistically significant only after 6 mo of exposure. No other changes in hematology, serum biochemistry or histology were reported. A concentration-related increase in lung weight was also observed in the rats. Swine were examined immediately after the treatment, at which time there was minimal fibrosis and no other statistically significant effects were observed. Newton et al. (1994) conducted a preliminary, subchronic study in which male and female F-344 rats (55/ sex/group) were exposed to antimony trioxide6 (concentrations of 0.25, 1.08, 4.92, and 23.46 mg antimony trioxide/m3) 6 hr/d, 5 d/wk for 13 wk followed by a 27-wk observation period. A decrease in body weight was seen in the males at the highest concentration tested and an increase in absolute lung weight was seen at the two highest exposure concentrations. Minimal-to-moderate microscopic pathologies were seen in the highest exposure group. The Newton et al. (1994) pilot study was followed by a 1-yr chronic study 3Average particle size by electron micrograph=0.6 µm, few particles up to 1 µm; calculated average particle size by weight=0.4 µm; purity not reported. 4It was noted by the authors that they had difficulty generating the target concentration of 50 mg antimony trioxide/m3; the mean daily time-weighted average (TWA)=45.5 mg/m3; the range was 0–191.1 mg antimony trioxide/m3; particle size: median circular area equivalent diameter=0.347 µm, mass median diameter (MMD) =1.23 µm, mass median aerodynamic diameter (MMAD) =2.80 µm, analyzed using a scanning electron microscope and image analyzer; purity: 80% antimony by proton-induced X-ray emission, 0.04 mg arsenic/g and 2.3 mg lead/g. 5Particle size averaged 0.44 µm (geometric standard deviation [SD] =2.23) and 0.40 µm (geometric SD=2.13) for low and high concentration, respectively; purity=99.4% antimony, 0.02% arsenic, 0.2% lead. MMAD, a critical parameter in inhalation studies, could not be directly derived from the data in this study. 6Test material not milled; particle size: count median diameter=0.485–0.62 µm, MMD=1.49–2.50 µm, MMAD=3.05–5.7 µm; purity=99.68±0.10%, contaminants not reported. 

ANTIMONY TRIOXIDE 239 (Bio/dynamics 1990, as cited in EPA 1999). In that study, F-344 rats (65/sex/ exposure level) were exposed to antimony trioxide7 (measured concentrations were 0, 0.06, 0.51, or 4.5 mg antimony trioxide/m3) 6 hr/d, 5 d/wk for 1 yr, followed by a 1-yr observation period. Five rats/sex/group were killed after 6 and 12 mo of exposure, and at 6 mo postexposure. All survivors were killed 12 mo after the end of the exposure period. Animal body weights were monitored. Complete gross and histopathological examinations were performed on all animals. The sections of the lungs examined included both the right lobes and the major bronchi. The only exposure-related changes occurred in the lungs and included chronic interstitial inflammation, granulomatous inflammation, and increased alveolar macrophages. Pinpoint black foci, thought to be aggregates of macrophages laden with antimony trioxide, were seen in the lungs of exposed animals, most frequently during the post-exposure observation period. In a subsequent analysis performed by the EPA (1999), it was noted that in the low- and mid- exposure groups, there was no indication that the particle-laden macrophages were anything but part of a normal, compensatory response. However, the clearance half-time of the high-exposure group was more than three times that of the mid-exposure group, indicating that clearance mechanisms were severely compromised. In some instances, clearance of particles is slowed by high lung burdens of inert particles, which leads to high lung particle burdens for extended periods (months to years), and pathologies that cannot be directly attributed to the toxicity of the chemical (Witschi and Last 1996). However, in this study (Newton et al. 1994), the decreased clearance appeared to be due to the inherent toxicity of antimony trioxide, rather than a particle overload phenomenon. Newton et al. (1994) reported a 50% increase in the clearance time of antimony trioxide at a dust volume of 270 nanoliter (nL), but benign dust particles have to be at about 1,000 nL to have that effect on clearance (Muhle et al. 1990, as cited in EPA 1999). However, some scientists believe that particle overload could account for the increased clearance time rather than inherent toxicity of antimony trioxide. Despite those observations, the subcommittee considered this study to be appropriate for calculation of an RfC. Based on additional statistical analysis (EPA 1999) of the male and female rats that died spontaneously or were killed at 18 and 24 mo, a LOAEL for interstitial inflammation and granulomatous inflammation of 4.5 mg antimony trioxide/ m3 and a NOAEL of 0.51 mg antimony trioxide/m3 (NOAEL[HEC] of 0.042 mg antimony trioxide/m3) were identified from this study. 7MMAD=3.7 µm, geometric S.D.=1.7; purity=99.68±0.10%, contaminants not reported. 

ANTIMONY TRIOXIDE 240 Reproductive and Developmental Effects Reproductive and developmental effects following inhalation exposure to antimony have been reported in one human study. Based on an English abstract of a study by Belyaeva (1967), women working in an antimony plant had a greater incidence of gynecological problems (not detailed), early interruption of pregnancy, and spontaneous late abortions compared to women working under similar conditions who were not exposed to antimony. Belyaeva (1967) also reported a reduction in the number of offspring and a disruption of ovulation in rats exposed to 250 mg/m3 antimony trioxide for 2 mo (particle size and purity not specified). In a study by Grin et al. (1987) that was translated for the subcommittee, pregnant rats (six to seven/group) were exposed to antimony trioxide (0.027, 0.082, and 0.27 mg antimony trioxide/m3, 24 hr/d; particle size and purity not reported) throughout gestation (21 d). Changes in clinical parameters at the highest exposure concentration tested included a very large increase in the amount of hemoglobin, blood leukocytes, serum lipids, and total protein in blood. The subcommittee noted that the effects on the hemoglobin and protein levels in the blood might indicate that the dams were sick, and therefore the maternal effects might have impacted the fetal effects. In the fetuses, gross macroscopic changes were seen at the two highest exposure concentrations tested, with increased bleeding in fetal brain membranes and liver, an increase in the size of the kidney cavity and the cerebral ventricles, and isolated cases of ossification at the highest exposure concentration tested. Some of the fetal effects in this study are listed in Table 10–3. Based on these data, 0.082 mg antimony trioxide/m3 can be considered a LOAEL and 0.027 mg antimony TABLE 10–3 Results of a Reproductive Toxicity Study on Antimony Trioxide (Grin et al. 1987) Concentration (mg/m3)a Number of dead and Preimplantation loss (%) Postimplantation loss (%) Total death rate resorbed/ female (%) 0 0.57±0.20 8.6±1.45 5.90±2.10 14.03±1.92 0.027 0.5±0.22 10.93±2.49 6.25±2.80 16.31±2.03 0.082 0.67±0.33 14.63±1.56 6.08±2.82 19.91±2.31 0.27b 1.20±0.48c 13.56±2.69c 12.22±4.90c 24.32±4.29c aExposure occurred throughout gestation. bSubstantial maternal toxicity was noted at this dose. cSignificantly different from controls by Student's t-test, p<0.05. 

ANTIMONY TRIOXIDE 241 trioxide/m3 a NOAEL in this study (Grin et al. 1987). However, the study is of limited use for quantitative toxicity assessment purposes because of the lack of information on the purity and particle size of the antimony trioxide used and the fact that maternal toxicity was seen. Therefore, the subcommittee decided not to use the study by Grin et al. (1987) for the determination of a critical level. As summarized in Reprotox (1999), studies with antimony compounds other than the trioxide have shown that, although antimony can enter the fetus (Gerber et al. 1982), antimony compounds are not teratogenic in chicks (Ridgway and Karnofsky 1952), rats (Rossi et al. 1987), or sheep (James et al. 1966). However, antimony trichloride (0.1 and 1 mg/dL in drinking water for 38 d) did decrease pup body weight and had some effects on cardiovascular responses to noradrenaline, isoprenaline, and acetylcholine (Rossi et al. 1987). Cancer Three epidemiological studies have evaluated the potential carcinogenicity of antimony following occupational exposure (Jones 1994; Potkonjak and Pavlovich 1983; Schnorr et al. 1995). Jones (1994) studied a cohort of 2,508 smelter workers and reported that antimony exposure was associated with an increased risk of lung cancer, with a standardized mortality ratio (SMR) of 2.18 (p < 0.001) in workers employed prior to 1961, but not in those employed after 1960. No data on cigarette smoking were provided, and many possible confounding exposures existed in the workplace, including exposure to arsenic, sulfur dioxide, and polycyclic aromatic hydrocarbons. Schnorr et al. (1995) conducted a retrospective cohort study of 1,014 smelter workers and reported a lung cancer SMR of 1.39; the 90% confidence interval was 1.01–1.88, indicating that even at the 90% confidence level, this SMR was only marginally statistically significant. No data on cigarette smoking were reported and workplace exposures levels were not measured. Potkonjak and Pavlovich (1983) evaluated 51 workers exposed to 5.5–64 mg antimony trioxide/m3 (particle size < 5 µm) for an average of 18 yr. No malignancies were observed in that study. Although the study by Jones (1994) suggests a correlation between antimony exposure and lung cancer risk, the use of this study is limited by the lack of an appropriate control population, and failure to control for bias and confounding factors. Results from animal studies are also conflicting. Two animal studies reported that antimony trioxide induced lung cancers in two strains of rats (Groth et al. 1986; Watt 1983). However, additional studies in rats (Newton et al. 1994) and a study in pigs (Watt 1983) did not confirm this effect. 

ANTIMONY TRIOXIDE 242 In a study by Groth et al. (1986), Wistar rats (90/sex/group) were exposed to antimony trioxide8 for 1 yr (target concentration=50 mg antimony trioxide/m3, 7 hr/d, 5 d/wk, killed 20 wk after end of exposure). The incidence of lung tumors was increased in female rats only, with tumors occurring in 19 of the 70 exposed females compared to 0 of the 70 control females. Of the lung tumors, nine were squamous-cell carcinomas, five were scirrhous carcinomas,9 and 11 were bronchioloalveolar adenomas and carcinomas. Some rats had more than one type of lung tumor. Rats were 8 mo of age at the beginning of exposure, and the first tumor was seen in a rat killed after wk 53. In an unpublished study, conducted by Watt (1983), groups of 48–50 female Charles River CDF rats were exposed to antimony trioxide10 (0, 1.9, or 5.0 mg antimony trioxide/m3, 6 hr/d, 5 d/wk) for 1 yr. Surviving rats were kept for up to 17 mo postexposure. Only 13–18 rats/group survived until 29 mo. Non-neoplastic lung lesions included focal fibrosis, pneumonocyte hyperplasia, cholesterol clefts,11 and multinucleated giant cells. Adenomatous hyperplasia of the lung was evident at the high concentration. The most common lung tumor was scirrhous carcinoma (incidences of 0/41, 0/44, and 15/45 in the control, low-, and high-concentration groups, respectively). Bronchioloalveolar adenomas were also increased at the high-exposure concentration group (incidences of 1/41, 1/44, and 4/45 in the control, low-, and high-concentration groups, respectively). Squamous cell carcinomas (2/45 at the high concentration) were observed in the high-exposure concentration group, but not in the low-exposure concentration or control groups. The study authors noted that the neoplasms appeared to arise from the alveolar epithelial lining cells. The tumor incidence was not significantly increased in any other tissue. This study was limited in that only 13–18 rats remained in each dose group at the end of 29 mo. In addi 8The authors noted difficulty in consistently generating 50 mg antimony trioxide/m3, the mean daily TWA=45.5 mg/m3, range=0–191.1 mg antimony trioxide/m3; particle size: median circular area equivalent diameter=0.347 µm, MMD=1.23 µm, MMAD =2.80 µm, analyzed using a scanning electron microscope and image analyzer; purity: 80% antimony by proton- induced X-ray emission, 0.04 mg arsenic/g, and 2.3 mg lead/g. 9An adenocarcinoma with a small number of tumor cells, in relation to an abundant amount of dense collagenous stroma, isolated and dispersed throughout the fibrous components (Becker et al. 1986). 10Generated by a modified hammer mill, average particle sizes were 0.44 µm (geometric SD=2.23) and 0.40 µm (geometric SD=2.13) for low and high concentrations, respectively; purity=99.4% antimony, 0.02% arsenic, 0.2% lead. 11Elongated defects that represent the site of a cholesterol crystal that has been dissolved during the preparative procedures (Becker et al. 1986). 

ANTIMONY TRIOXIDE 243 tion, continuing the study until 29 mo after study initiation increased the potential for age-related tumors, thus decreasing the study sensitivity. However, in light of the low background and clear increases seen in scirrhous carcinoma, these limitations do not affect the study conclusions. Watt (1983) also examined the carcinogenicity of antimony trioxide in female pigs and found no neoplasms at the end of the 1-yr exposure (1.9 or 5.0 mg antimony trioxide/m3, 6 hr/d, 5 d/wk). The negative response could be due to a low sensitivity of this species, or the lack of an appropriate observation period following exposure. No increase in cancers was observed in F-344 rats (65/sex/group) exposed to antimony trioxide12 (0, 0.06, 0.51, or 4.5 mg antimony trioxide/m3, 6 hr/d, 5 d/wk) for 1 yr, and observed up to 1 yr after exposure (Newton et al. 1994). Extensive gross necropsy and histopathology were conducted. The lung tissue examined included the major bronchi. Elevated total leukocyte counts and atypical lymphocytes indicated leukemia in all groups. However, the authors noted that leukemia is a common finding in aged F-344 rats. Two males (one from the control group and one from the 4.5-mg/m3 concentration group) and one female (0.51-mg/m3 concentration group) had pulmonary carcinomas; the carcinomas were not considered to be treatment related. According to the authors, other neoplastic findings occurred sporadically or with an incidence similar to that of the controls (Newton et al. 1994). Based on the pathological examination of the lungs from all three rat studies discussed above (Groth et al. 1986; Watt 1983; Newton et al. 1994), Newton et al. (1994) suggested that the differences in carcinogenesis are due to a different deposition pattern of antimony trioxide in the lungs. Newton et al. (1994) noted, however, that particle size could not explain these differences (see Table 10–4). Although the rats were reportedly exposed to similar concentrations in the Watt (1983) and Newton et al. (1994) studies, the rats in the Watt (1983) study had more damage and considerably more test material in the lung. It was suggested that the rats in the Watt (1983) study actually had a higher exposure than measured based on the extent of particle deposition (Newton et al. 1994). Newton et al. (1994) also concluded that the foreign body reaction cannot completely account for the tumors observed in the Groth et al. (1986) study, since females, but not males, were affected in that study. 12MMAD=3.7 µm, geometric SD=1.7; purity=99.68+0.10%, contaminants not reported. 

ANTIMONY TRIOXIDE 244 TABLE 10–4 Antimony Trioxide Particle Size (Micrometers) Particle Sample Count Geometric Mass Mass Median N Reference Sizing Median Standard Median Aerodynamic Technique Diameter Deviation Diameter Diameter SEM: Feret High 0.40 2.13 2.22a 5.06b NA Watt 1983 diameter chamber 0.44 2.23 3.03 6.9 NA Low chamber SEM: Chamber 0.347 1.23c 2.8b 1,948 Groth 1986 equivalent sample area diameter TSI Particle Groups II- 1.79 3.76±0.84 20 Newton et Sizer IV al. 1994 Cascade Group IV 1.80 4.55 2 Impactor CMD, count median diameter; GSD, geometric standard deviation; MMAD, mass median aerodynamic diameter; MMD, mass median diameter; NA, not available; SEM, scanning electron microscopy; sqrt, square root; TEM, transmission electron microscopy; TSI, TSI Aerodynamic Particle Sizer AP340. aCalculated using MMD=CMD exp (31n2GSD). bCalculated using MMAD=MMD×sqrt (density). cCalculated using the equivalent area diameters and an assumed spherical particle with density of 5.2 g/cm2. Source: Modified from Newton et al. 1994. 

ANTIMONY TRIOXIDE 245 The International Agency for Research on Cancer (IARC) classifies antimony trioxide as a possible carcinogen to humans, group 2B, based on sufficient evidence for the carcinogenicity in experimental animals (by inhalation), but inadequate evidence for the carcinogenicity in humans (IARC 1989). That assessment was completed before the publication of the negative study by Newton et al. (1994). In summary, based on the weight of evidence, the subcommittee concluded that there is suggestive evidence that antimony trioxide is carcinogenic and a quantitative cancer risk assessment was performed based on the study by Watt (1983) (see Cancer section under Quantitative Toxicity Assessment). Other Systemic Effects No studies were identified that investigated the immunological or neurological effects of antimony trioxide following inhalation exposure. Oral Exposure Systemic Effects There are no data on the health effects of antimony trioxide in humans following oral exposure. Oral exposure studies conducted in animals are summarized in Table 10–5. An oral LD50 of >20 g/kg body weight has been reported in rats for antimony trioxide (Smyth and Carpenter 1948, as cited in ATSDR 1992; Ebbens 1972). Diarrhea has been reported in rats administered 16.7 g/kg body weight antimony trioxide in oil by gavage (Myers et al. 1978). The same dose in water given by gavage (Gross et al. 1955a), or provided in food (Smyth and Thompson 1945) did not produce any observable toxicity. Rats gavaged with 8.6–29 g/kg body weight antimony trioxide exhibited hypoactivity and ruffed fur within 1 hr after dosing, but returned to normal after 2 d (Ebbens 1972). No gross pathologic alterations were observed upon necropsy in that study. No significant treatment-related effects were seen in rats following gavage with 134–501 mg/kg-bw/d antimony trioxide when administered in either 0.4% hydrochloric acid or 4% citric acid/0.4% hydrochloric acid for 20 d. Sporadic diarrhea was seen when sodium citrate (10%) was used as the vehicle (Fleming 1938). In a 21- d study, two dogs were gavaged daily with 1,000 mg antimony trioxide (79 mg/kg-bw/d) in water (Fleming 1938). The animals developed 

ANTIMONY TRIOXIDE 246 TABLE 10–5 Toxic Effects of Antimony Trioxide Following Oral Exposure Species, Strain, Dose Duration, Route Effects NOAEL/ Reference Sex, Number LOAEL (mg/kg- d) Rat, Sherman, NS Single dose, oral LD50>20 g/kg ND Smyth and NS, 6/dose Carpenter 1948, as cited in ATSDR 1992 Rat, Charles 10.25, 15.38, Single dose, LD50>34.6 g/kg; ND Ebbens 1972 River, M/F, 4/ 23.07, 34.6 g/kg gavage hypoactivity and dose in water ruffed fur at all doses Rat 16.7 g/kg in oil Single dose, Diarrhea ND Meyers et al. 1978 gavage Rat 16.7 g/kg in Single dose, No effects observed ND Gross et al. 1955a water gavage Decreased weight gain ND 1.3 g/kg-d 240 d, feed Rat, Albino, M, 16.7 g/kg 60, Single dose, Decreased food ND Smyth and ND 10/dose 270, 1,070 mg/ feed 30 d, feed consumption, Thompson 1945 kg-d decreased weight gain, increased red blood cell count in high-dose group Dog, NS, NS, 2 79 mg/kg-d 7.7 21 d, gavage 11 Diarrhea, reversible ND Fleming 1938 mg/kg-d in 21% d, gavage Diarrhea, weight loss, ND citric acid, gastrointestinal and second liver lesions treatment in same dogs 

ANTIMONY TRIOXIDE 247 Rat 670 mg/kg-d 12 wk, feed Decreased weight gain, ND Hiraoka 1986, as spleen weight, heart cited in ATSDR weight; increased lung 1992 weight Rat, Wistar, M, 500, 1,000 mg/ 24 wk, feed Decreased red blood cell LOAEL: 500 Sunagawa 1981 5/dose kg-d count (500, 1,000), increased serum glutamic oxaloacetic transaminase; no macroscopic changes Rat, Wistar, M/ M: 84, 421, 90 d, feed Increased red blood cell Based on enzyme Hext et al. 1999 F, 12/sex/dose 1,686 mg/kg-d count (high-dose M/F), changes in F: F: 97, 494, increased urine volume NOAEL: 494 1,879 mg/kg-d (high-dose M/F), LOAEL: 1,879 decreased urine specific gravity (high-dose F), increased serum cholesterol (high-dose F), increased triglycerides (high-dose M), decreased alkaline phosphatase activity (high-dose M, mid- and high-dose F), increased aspartate and alanine aminotransferase (high-dose F) F, female; LOAEL, lowest-observed-adverse-effect level; M, male; ND, not determined; NOAEL, no-observed-adverse-effect level; NS, not specified. 

ANTIMONY TRIOXIDE 248 severe diarrhea, which lasted 6 or 7 d, but resolved prior to completion of the treatment, suggesting either that the diarrhea was not a severe response or that the animals adapted to the treatment. Following the 21-d treatment with antimony trioxide dissolved in water, the same dogs were administered 7.7 mg/kg-bw/d antimony trioxide dissolved in 5% citric acid for 11 d. Diarrhea, weight loss, and gastrointestinal and liver lesions were observed. Although the usefulness of this study is limited by the small number of animals used and the lack of control group, the results suggest that solubility plays a role in the toxicity of orally administered antimony trioxide (Fleming 1938). In a short-term exposure toxicity study of antimony trioxide in which groups of 10 male albino rats received antimony trioxide in their diet (0%, 0.1%, 0.45%, 1.8%; corresponding to 0, 60, 270, 1,070 mg antimony trioxide/kg-d) for 30 d, rats in the high-dose group (1,070 mg/kg-d) had significantly decreased food consumption (41%) and decreased body weight gain (43%) compared with controls (Smyth and Thompson 1945). Hematological examination indicated that rats in the high-dose group had an increased red blood cell count but no change in hemoglobin concentration compared to controls. The high dose of 1,070 mg antimony trioxide/kg-d was considered a NOAEL for the derivation of the oral reference dose (RfD) because the subcommittee did not consider an increase in red blood cell count to be an adverse effect and because the other effects are probably related to decreased food consumption. Rats fed with 670 mg antimony trioxide/kg bw-d in the diet for 12 wk had a decrease in overall weight gain, spleen and heart weight, and an increase in lung weight (Hiraoka 1986; as cited in ATSDR 1992). Reduced weight gain was also seen in rats given approximately 1.3 g antimony trioxide/kg bw-d in food for 240 d. No gross or microscopic changes were seen in those animals (Gross et al. 1955a). Sunagawa (1981) fed male Wistar rats (five animals/group) 0%, 1.0%, or 2.0% antimony trioxide (calculated to be 0, 500, or 1,000 mg antimony trioxide/kg bw-d) in the diet for 24 wk. Exposure to antimony trioxide had no effect on gross appearance or behavior, and did not affect body weight, food and water intake, or organ weights in the rats. Red blood cell count was significantly decreased (not dose-dependent) in both treated groups compared with controls. No changes were observed in white blood cell count, hematocrit, or hemoglobin concentration. Serum glutamic oxalacetic transaminase (SGOT) was significantly increased (p < 0.05) in both dose groups. Histopathological evaluation of the liver indicated some (not statistically significant) disorders of hepatic laminae, cloudy swelling in hepatic cords, and vacuolar degeneration in hepatic cells. Based on the suggestion of liver toxicity and the decreased red blood cell count, a LOAEL of 500 mg antimony trioxide/kg bw- d was identified from this study. However, this study is of limited usefulness for the deriva 

ANTIMONY TRIOXIDE 249 tion of an RfD for antimony trioxide because of the small number of animals used and the fact that only the abstract and data tables were available in English. Hext et al. (1999) fed male and female Wistar rats (12/sex/dose) diets containing 0, 1,000, 5,000, or 20,000 ppm antimony trioxide for 90 d (0, 84, 421, and 1,686 mg antimony trioxide/kg-d for males, and 0, 97, 494, and 1,879 mg antimony trioxide/kg-d for females; based on measured food consumption and body weights). Urine volume was significantly increased, and specific gravity was significantly decreased in high-dose females. Serum cholesterol and urine volume in high-dose females (dose-related trend), and triglycerides and red blood cell count in high-dose males were increased. Alkaline phosphatase (AlkP) activity was significantly decreased in high-dose males and mid- and high-dose females (dose-related trend). SGOT and serum glutamic aminotransferase (SGPT) were significantly increased in high-dose females. Absolute and relative liver weights were increased by approximately 10% in high-dose males and females. No other treatment-related effects were seen. The subcommittee concluded that the effects seen in this study are adverse when considered together with the data from Sunagawa (1981) and Smyth and Thompson (1945). Based on the increase in serum enzymes (statistically significant only in females), and the liver weight, 1,879 mg/kg-d is identified as a LOAEL for this study; the NOAEL is 494 mg/kg-d. Other Systemic Effects No studies were identified that investigated immunological, neurological, reproductive, developmental, or carcinogenic effects of antimony trioxide following oral exposure. Genotoxicity Although a single oral gavage of antimony trioxide (400, 666.67, and 1,000 mg/kg) did not cause chromosome aberrations in mouse bone marrow cells, aberrations were observed following repeated administration of those doses (Gurnani et al. 1992). Repeated oral doses of antimony trioxide, however, did not cause unscheduled DNA synthesis in the liver cells of rats, or an increase in the micronucleated polychromatic erythrocytes in the mouse bone marrow micronucleus assay (Elliott et al. 1998). Antimony trioxide was not mutagenic in Salmonella typhimurium or E. coli strains (Kanematsu et al. 1980; Kuroda et al. 1991), but it did cause sister 

ANTIMONY TRIOXIDE 250 chromotid exchange (SCE) in V79 Chinese hamster cells (Kuroda et al. 1991). DNA damage occurred following antimony trioxide treatment in Bacillus subtilis in Rec assays (Kanematsu et al. 1980; Kuroda et al. 1991). QUANTITATIVE TOXICITY ASSESSMENT Noncancer Dermal Assessment There are inadequate dermal toxicity data on antimony trioxide to derive a reference dose for dermal exposure. Inhalation RfC In 1995, the EPA derived a reference concentration (RfC) value for antimony trioxide (EPA 1999) based on the study by Newton et al. (1994). The subcommittee agrees that the Newton et al. (1994) study is the critical study for the derivation of an inhalation RfC, and that the critical end points chosen by the EPA are appropriate. The subcommittee, therefore, used EPA's benchmark concentration (BMC) analysis to determine their recommended level for antimony trioxide. The BMC was calculated for chronic pulmonary inflammation,13 granulomatous inflammation, and fibrosis in males, females, and both sexes combined. The lower 95% confidence level on the concentration corresponding to a 10% extra risk of pulmonary inflammation (i.e., a 10% increase in the incidence of pulmonary inflammation) (the BMCL10) was determined. The most sensitive end point was chronic inflammation in female rats, for which the BMCL10 was 0.87 mg antimony trioxide/m3. Adjusted for intermittent exposure of 6 hr/d, 5 d/wk, the BMC10(ADJ) was 0.16 mg antimony trioxide/m3. The human equivalent concentration, BMC10 (ADJ, HEC), of that exposure was calculated to be 0.074 mg/m3 (using a regional deposited dose ratio [RDDR] for the thoracic region of 0.46). That value is similar to the HEC of 0.042 mg/m3 calculated from the NOAEL of 0.51 mg antimony trioxide/m3. The derivation of the RfC is shown in Table 10–6. To derive the RfC from the BMC10 (ADJ, HEC) of 0.16 mg antimony trioxide/m3, a composite uncertainty factor of 300 was used, which included a factor of 3 for interspecies extrapolation, a factor 13The lung tissue examined included the right lobes and the major bronchi. 

ANTIMONY TRIOXIDE 251 of 10 for intraspecies variation, a factor of 3 for database inadequacies, and a factor of 3 for a less-than-lifetime exposure that was longer than the standard subchronic study. Division of BMC10 (ADJ, HEC) by the composite uncertainty factor resulted in an RfC of 0.2 µg antimony trioxide/m3. TABLE 10–6 Inhalation Reference Concentration for Antimony Trioxide Critical Effect Species Effect Level (mg/m3) Uncertainty Factors RfC (µg/m3) Reference Pulmonary toxicity Rats BMC10: (ADJ, HEC) 0.074 UFA: 3 0.2 Newton et al. 1994 UFH: 10 UFS: 3 UFD: 3 Total: 300 BMC10, concentration corresponding to an extra risk of 10% (adjusted for intermittent exposure and calculated as the human equivalent concentration); RfC, reference concentration; UFA, extrapolation from animals to humans; UFH, intraspecies variability; UFS, extrapolation from a study of less-than-lifetime duration; UFD, inadequate or deficient toxicity database. The key study used for the derivation of the inhalation RfC was assigned a medium confidence level. Although it was well conducted and well documented, it is not a lifetime exposure study. Confidence in the database is medium because of the absence of adequate developmental or reproductive toxicity studies. Therefore, confidence in the RfC is medium. Oral RfD The database for developing an oral reference dose (RfD) for antimony trioxide is limited to one high- quality subchronic feeding study in rats (Hext et al. 1999). That study is supported by data from a subchronic study in rats (Sunagawa 1981), and short-term studies in rats (Smyth and Thompson 1945) and dogs (Fleming 1938). Overall, those data indicate that the hematological system (increased serum enzymes), the liver (increased liver weights), and the gastrointestinal tract are the target organs for antimony trioxide. Based on the weight of evidence, the subcommittee considered the increases in serum enzymes in females and the increase in liver weight in males and females at 1,879 mg Sb2O3/kg bw-d to be adverse effects (Hext et al. 1999). Therefore, the LOAEL for that study is 1,879 mg antimony trioxide/kg-d and the NOAEL is 494 mg antimony trioxide/kg-d. A composite uncertainty factor of 3,000 is 

ANTIMONY TRIOXIDE 252 applied to that NOAEL to yield an RfD of approximately 0.2 mg antimony trioxide/kg-d. The composite uncertainty factor comprises a factor of 10 for interspecies extrapolation; a factor of 10 to for intraspecies variability; a factor of 10 for extrapolation from a subchronic to a chronic study; and a factor of 3 for data base deficiencies (i.e., lower than the default of 10 because there is some data that indicate there is no progression in severity of effects). A summary of the derivation of that oral RfD is provided in Table 10–7. The key study used for the derivation of the RfD was conducted according to current testing guidelines and is well documented; therefore, confidence in the key study is high. However, confidence in the overall database is low, because there are no adequate data on developmental or reproductive effects. Several additional studies are needed to complete the database, including a bioassay in a second species, a multigeneration reproduction study, and developmental toxicity studies in two species. Longer-term assays would also be informative. As a result, the confidence for the derived RfD is low to medium. Cancer Dermal The carcinogenicity of antimony trioxide by the dermal route of exposure cannot be determined because of lack of data. Inhalation Based on the weight of evidence (from animal studies), the subcommittee concludes that the data are suggestive of carcinogenicity following inhalation TABLE 10–7 Oral Reference Dose for Antimony Trioxide Critical Effect Species Effect Level (mg/kg-d) Uncertainty Factors RfD (mg/kg-d) Reference Increases in serum Female rats NOAEL=494 UFA: 10 0.2 Hext et al. 1999 enzymes; increased UFH: 10 liver weight UFS: 10 UFD: 3 Total: 3,000 NOAEL, no-observed-adverse-effect level; RfD, reference dose; UFA, extrapolation from animals to humans; UFH, intraspecies variability; UFS, extrapolation from a study of less-than-lifetime duration; UFD, inadequate or deficient toxicity database. 

ANTIMONY TRIOXIDE 253 exposure to antimony trioxide. The cancer risk from antimony trioxide following inhalation exposure was estimated based on the study by Watt (1983). It should be noted, however, that the study by Watt (1983) is not published in the peer review literature and the results are controversial. A linear extrapolation from the observable region to the low-dose region is appropriate because there are insufficient data to suggest a nonlinear mode of action. The data by Watt (1983) was modeled using the linear multistage model. The three tumor end points from the Watt (1983) study that were modeled were bronchioalveolar adenomas, scirrhous carcinomas, and squamous- cell carcinomas.14 Concentrations were adjusted for discontinuous exposure (multiplied by 6 hr/24 hr×5 d/7 d), converted to a HEC using the regional deposited dose ratio (RDDR=1.8342) of particles for the thoracic region (MMAD=0.4 microns, sigma g=2.2), and adjusted for the less-than-lifetime exposure. The modeling results are listed in Table 10–8. Because all the tumors occurred in the bronchioalveolar region and appeared to be arising from the alveolar epithelial lining cells, the three tumor types were combined, and total bronchioalveolar tumors were also modeled for the LED10. Using the combined bronchioalveolar tumor incidence yields the most conservative (health-protective) estimate of the risk, with an LED10 of 0.14 mg antimony trioxide/m3. Based on a linear extrapolation, the unit risk (cancer potency factor) of lung cancer is 7.1×10−4/µg antimony trioxide/m3. Oral The carcinogenicity of antimony trioxide by the oral route of exposure cannot be determined because of lack of data. EXPOSURE ASSESSMENT AND RISK CHARACTERIZATION Noncancer Dermal The assessment of noncancer risk by the dermal route of exposure is based on the scenario described in Chapter 3. This exposure scenario assumes that an 14The incidence of tumors in all animals was used in the hazard identification. However, since the concentration-response modeling is based on tumors following a lifetime exposure, and there were several interim kills in this study, only tumors in the animals sacrificed at study termination were used in the modeling. 

ANTIMONY TRIOXIDE 254 adult spends 1/4th of his or her time sitting on furniture upholstery treated with antimony trioxide, that 1/4th of the upper torso is in contact with the upholstery, and that clothing presents no barrier. Antimony trioxide is considered to be ionic, and is essentially not absorbed through the skin. However, to be conservative, the subcommittee assumed that ionized antimony trioxide permeates the skin at the same rate as water, with a permeability rate of 10−3 cm/hr (EPA 1992). Using that permeability rate, the highest expected application rate for antimony trioxide (2.5 mg/cm2), and Equation 1 in Chapter 3, the subcommittee calculated a dermal exposure level of 2.0×10−2 mg/kg-d. The oral RfD for antimony trioxide (0.2 mg/kg-d; see Oral RfD in Quantitative Toxicity section) was used as the best estimate of the internal dose for dermal exposure. Dividing the exposure level by the oral RfD yields a hazard index of 0.1. Thus it was concluded that antimony trioxide used as a flame retardant in upholstery fabric is not likely to pose a noncancer risk by the dermal route. TABLE 10–8 Results of Modeling for the Watt (1983) Study on Antimony Trioxide Multistage model (mg/m3) Risk by EPA methods (mg/m3)−1 End point ED10 LED10 0.1/ED10 0.1/LED10 Adenomas 0.51 0.24 0.185 0.417 Scirrhous carcinomas 0.35 0.21 0.286 0.476 Squamous-cell carcinomas 0.83 0.43 0.120 0.233 Bronchioalveolar tumor (combined) 0.24 0.14 0.417 0.714 ED10, effective dose corresponding to a 10% tumor response in test animals; LED10, lower 95% bound on the effective dose corresponding to a 10% tumor response in test animals. Inhalation Particles The assessment of the noncancer risk by the inhalation route of exposure is based on the scenario described Chapter 3. This scenario corresponds to a person spending 1/4th of his or her life in a room with low air-change rate 

ANTIMONY TRIOXIDE 255 (0.25/hr) and with a relatively large amount of fabric upholstery treated with antimony trioxide (30 m2 in a 30- m3 room), with this treatment gradually being worn away over 25% of its surface to 50% of its initial quantity over the 15-yr lifetime of the fabric. A small fraction, 1%, of the worn-off antimony trioxide is released into the indoor air as inhalable particles and is breathed by the occupant. Equations 4 through 6 in Chapter 3 were used to estimate the average concentration of antimony trioxide in the air. The highest expected application rate for antimony trioxide is 2.5 mg/cm2. The estimated release rate is 2.3× 10−7/d. Using those values, the estimated time-averaged exposure concentration for antimony trioxide is 0.24 µg/m3. Division of that exposure concentration (0.24 µg/m3) by the inhalation RfC (2×10−4 mg/m3; see Quantitative Toxicity Assessment section) results in a hazard index of 1.2, indicating that under the worst-case exposure scenario, antimony trioxide might possibly pose a noncancer risk via inhalation of particles. Vapors In addition to the possibility of release of antimony trioxide in particles worn from upholstery fabric, the subcommittee considered the possibility of its release by evaporation. However, because of antimony trioxide's negligible vapor pressure at ambient temperatures, the subcommittee considered antimony trioxide not likely to pose a noncancer risk by exposure to vapors. Oral Exposure The assessment of the noncancer risk by the oral exposure route is based on the scenario described in Chapter 3. That exposure assumes that a child sucks on 50 cm2 of fabric backcoated with antimony trioxide daily for two yr, one hr/d. The highest expected application rate (per unit time) for antimony trioxide is about 2.5 mg/ cm2. The fractional release rate of antimory trioxide is estimated as 0.001/d, based on the leaching of antimony from polyvinyl chloride cot mattresses (Jenkins et al. 1998). Using those assumptions and Equation 15 in Chapter 3, the average oral dose rate is estimated to be 0.00052 mg/kg-d. Division of that exposure estimate (0.00052 mg/kg-d) by the oral RfD (0.2 mg/kg-d; see Quantitative Toxicity Assessment section) results in a hazard index of 2.6×10−3. Therefore, under the worst-case exposure assumptions, antimony trioxide, used as a flame retardant in upholstery fabric, is not likely to pose a noncancer risk by the oral route of exposure. 

ANTIMONY TRIOXIDE 256 Cancer There are inadequate data to assess the carcinogenicity of antimony trioxide from dermal or oral exposures. Inhalation (Particles) The average room-air concentration and average exposure concentration for antimony trioxide were obtained as described for the noncancer risk assessment of particles. The estimated time-averaged exposure concentration is 0.24 µg/m3. Using the inhalation unit cancer risk (cancer potency factor) of 7.1×10−4/µg antimony trioxide/m3, the lifetime excess cancer risk estimate from exposure to antimony trioxide as particles is 1.7×10−4. However, the inhalation unit risk (cancer potency factor) of antimony trioxide is itself suspect (see Hazard Identification Section). Furthermore, even if the reservations concerning the study by Watt (1983) are discounted and the inhalation unit risk is considered to be accurate, better exposure assessment is required before any definitive conclusions can be drawn about the carcinogenic risk from antimony trioxide via inhalation in the particulate phase. Inhalation (Vapors) Antimony trioxide has negligible vapor pressure at ambient temperatures, so antimony trioxide used as a flame retardant in upholstery fabric is not likely to pose a cancer risk for exposure to vapors. RECOMMENDATIONS FROM OTHER ORGANIZATIONS The American Conference of Governmental Industrial Hygienists (ACGIH) has established a Threshold Limit Value (TLV) for antimony trioxide of 0.5 mg antimony/m3 (AGCIH 1999). The Occupational Safety and Health Administration (OSHA) and the National Institute for Occupational Safety and Health (NIOSH) do not have standards for exposure to antimony trioxide. EPA's inhalation RfC of 0.2 µg antimony trioxide/mg3 is the same as that of the subcommittee. 

ANTIMONY TRIOXIDE 257 DATA GAPS AND RESEARCH NEEDS There are little data on the toxicity of antimony trioxide following dermal exposure. The hazard index of 0.1 indicates that antimony trioxide is not likely to pose a non-cancer risk from dermal exposure. Therefore, the subcommittee does not recommend further research on the effects of antimony trioxide from dermal exposure for the purposes of flame-retarding upholstery furniture. The subcommittee's risk characterization indicates that antimony trioxide might possibly pose a risk for noncancer and cancer end points via inhalation in the particulate phase. Therefore, better exposure information is essential to accurately assess the risks of antimony trioxide use as a flame retardant in upholstery fabric. If that research shows that actual exposures are substantially lower than the subcommittee's estimated levels, there will be a reduced need to perform toxicity studies. One study indicated that there are reproductive effects following inhalation of antimony trioxide. However, the purity of the antimony trioxide in that study is not known, and studies of other antimony compounds show no reproductive effects (Reprotox 1999). The study on which the quantitative toxicity assessment for cancer is based is suspect, and further studies would clarify if antimony indeed poses a cancer risk following inhalation exposure. There are no studies that evaluated the chronic toxicity of antimony trioxide from the oral route of exposure. There are no studies that have measured exposure from the oral route. The hazard index of 2.6×10−3 indicates that antimony trioxide is not likely to pose a noncancer risk from oral exposure. Therefore, the subcommittee does not recommend further studies of antimony trioxide following oral exposure for the purposes of its use as a flame retardant in upholstery furniture fabric. With respect to cancer, the effects following inhalation exposure are portalof-entry specific (i.e., only occur in the lung), and, therefore, the subcommittee does not recommend carcinogenic studies following other routes of exposure. Based on an inhalation hazard index greater than one and a potential cancer risk following inhalation exposure, the subcommittee recommends that the potential for particle release from treated fabric be investigated. REFERENCES ACGIH (American Conference of Government Industrial Hygienists). 1999. Threshold Limit Values and Biological Exposure Indices. Cincinnati, OH: American Conference of Government Industrial Hygienists, Inc. 

ANTIMONY TRIOXIDE 258 ATSDR (Agency for Toxic Substances and Disease Control). 1992. Toxicological Profile for Antimony. Agency for Toxic Substances and Disease Registry, U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA. Bailly, R., R.Lauwerys, J.P.Buchet, P.Mahieu, and J.Konings. 1991. Experimental and human studies on antimony metabolism: Their relevance for the biological monitoring of workers exposed to inorganic antimony. Br. J. Ind. Med. 48(2):93–97. Becker, E.L., Landau, S.I., A.Manuila, W.J.H.Butterfeild, A.M.Harvey, R.H. Heptinstall, and L.Thomas, eds. 1986. International Dictionary of Medicine and Biology. New York: Wiley. Belyaeva, A.P. 1967. The effect produced by antimony on the generative function. [Abstract]. Gig. Truda. Prof. Zabol. 11(1):32–37. Bio/dynamics (Bio/dynamics, Inc.). 1990. A One-year Inhalation Toxicity Study of Antimony Trioxide in the Rat (With a One-year Recovery Period). Project No. 83–7647. Bui/dynamics, Inc., East Millstone, N.J. Budavari, S., M.J.O'Neil, A.Smith, and P.E.Heckelman. 1989. The Merck Index, 11th Ed. S.Budavari, M.J.O'Neil, A.Smith, and P.E.Heckelman, eds. Rahway, N.J.: Merck & Co., Inc. Cooper, D.A., E.P.Pendergrass, A.J.Vorwald, R.L.Mayock, and H.Brieger. 1968. Pneumoconiosis among workers in an antimony industry. Am. J. Roentgenol. Radium Ther. Nucl. Med. 103(3):495–508. Dernehl, C.U., C.A.Nau, and H.H.Sweets. 1945. Animal studies on the toxicity of inhaled antimony trioxide. J. Ind. Hyg. Toxicol. 27(9):256– 262. Ebbens, K. 1972. Acute Toxicity Studies with Antimony Oxide 2996–30. Industrial Bio-Test Labs, Inc., Northbrook, IL. EPA/OTS Doc #88– 920007957. Elliott, B.M., J.M.Mackay, P.Clay, and J.Ashby. 1998. An assessment of the genetic toxicology of antimony trioxide. Mutat. Res. 415(1–2): 109–117. EPA (U.S. Environmental Protection Agency). 1992. Dermal Exposure Assessment: Principles and Applications. EPA/600/8–91–011B. Office of Health and Environmental Assessment, U.S. Environmental Protection Agency, Washington, D.C. EPA (U.S. Environmental Protection Agency). 1999. Integrated Risk Information System (IRIS). [Online]. Available: http://www.epa.gov/ iris/subst/0676.htm Office of Research and Development. National Center for Environmental Assessment, Cincinnati, OH. Downloaded June, 1999; Antimony file dated 1991, antimony trioxide file dated 1995. Fleming, A.J. 1938. The Toxicity of Antimony Trioxide. E.I. du Pont de Nemours and Co, Wilmington, Del. EPA/OTS Document #87220297. Gerber, G.B., J.Maes, and B.Eykens. 1982. Transfer of antimony and arsenic to the developing organism. Arch. Toxicol. 49(2): 159–168. Gerhardsson, L., D.Brune, G.F.Nordberg, and P.O.Wester. 1982. Antimony in lung, liver and kidney tissue from deceased smelter workers. Scand. J. Work Environ. Health 8(3):201–208. Grin, N.V., N.N.Govorunova, A.N.Bessemrnyi, and L.V.Pavlovich. 1987. Embryotoxic action of antimony oxide in an experiment. [Article in Russian]. Gig. Sanit. 10:85–86. 

ANTIMONY TRIOXIDE 259 Gross, P., J.H.U.Brown, M.L.Westrick, R.P.Srsic, N.L.Butler, and T.F.Hatch. 1955a. Toxicologic study of calcium halophosphate phosphors and antimony trioxide. I. Acute and chronic toxicity and some pharmacologic aspects. Arch. Ind. Health 11:473–478. Gross, P., M.L.Westrick, J.H.U.Brown, R.P.Srsic, H.H.Schrenk, and T.F.Hatch. 1955b. Toxicologic study of calcium halophosphate phosphors and antimony trioxide. II. Pulmonary studies. Arch. Ind. Health 11:479–486. Groth, D.H., L.E.Stettler, J.R.Burg, W.M.Busey, G.C.Grant, and L.Wong. 1986. Carcinogenic effects of antimony trioxide and antimony ore concentrate in rats. J. Toxicol. Environ. Health 18(4):607–26. Gurnani, N., A.Sharma, and G.Talukder. 1992. Comparison of the clastogenic effects of antimony trioxide on mice in vivo following acute and chronic exposure. Bio-Metals 5(1):47–50. Haskell Laboratory (Haskell Laboratory for Toxicology and Industrial Medicine). 1970a. Primary Skin Irritation and Sensitization Tests in Guinea Pigs. U.S. EPA OTS document ID# 878220294. Haskell Laboratory (Haskell Laboratory for Toxicology and Industrial Medicine). 1970b. Primary Skin Irritation and Sensitization Tests. U.S. EPA OTS document ID# 878220307. Hatlelid, K. 1999. Toxicity Review for Calcium Molybdate and Zinc Molybdate. Memorandum, dated March 2, 1999, from Kristina Hatlelid, Toxicologist, Division of Health Sciences, to Ronald Medford, Assistant Executive Director for Hazard Identification and Reduction, U.S. Consumer Product Safety Commission, Washington, DC. Hext, P.M., P.J.Pinto, and B.A.Rimmel. 1999. Subchronic feeding study of antimony trioxide in rats. J. Appl. Toxicol. 19(3):205–209. Hiraoka, N. 1986. The toxicity and organ distribution of antimony after chronic administration to rats. J. Kyoto Prefect Univ. Med. 95:997– 1017. IARC (International Agency for Research on Cancer). 1989. Antimony trioxide and antimony trisulfide. Pp. 291–305 in IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 47, Some Organic Solvents, Resin Monomers and Related Compounds, Pigments and Occupational Exposures in Paint Manufacture and Painting. Lyon, France: World Health Organization, International Agency for Research on Cancer. ICRP (International Commission on Radiological Protection). 1981. Metabolic data for antimony. Pp. 46–49 in Radiation Protection, ICRP Publication 30, Part 3, Including Addendum to Parts 1 and 2, Limits for Intakes of Radionuclides by Workers. International Commission on Radiological Protection. Elmsford, N.Y.: Pergamon Press. Industrial Bio-Test Laboratories, Inc. 1973. Human Repeated Insult Patch Test with TS-162. U.S. EPA OTS document I.D.# 878211206. James, L.F., V.A.Lazar, and W.Binns. 1966. Effects of sublethal doses of certain minerals on pregnant ewes and fetal development. Am. J. Vet. Res. 27(116): 132–135. 

ANTIMONY TRIOXIDE 260 Jenkins, R.O., P.J.Craig, W.Goessler, and K.J.Irgolic. 1998. Antimony leaching from cot mattresses and sudden infant death syndrome (SIDS). Hum. Exp. Toxicol. 17(3):138–139. Jones, R.D. 1994. Survey of antimony workers: Mortality 1961–1992. Occup. Environ. Med. 51(11):772–776. Kanematsu, N., M.Hara, and T.Kada. 1980. Rec assay and mutagenicity studies on metal compounds. Mutat. Res. 77(2): 109–116. Kim, K.-W., B.-S.Choi, S.-K.Kang, H.-Y.Kim, S.S.Park, Y.-S.Cho, M.-G.Song, and Y.-H.Moon. 1997. Assessment of workers' exposure to antimony trioxide in Korea. J. Occup. Health 39(4):345–348. Kuroda, K., G.Endo, A.Okamoto, Y.S.Yoo, and S.Horiguchi. 1991. Genotoxicity of beryllium, gallium, and antimony in short-term assays. Mutat. Res. 264(4): 163–170. Leffler, P, L.Gerhardsson, D. Brune, and G.F. Nordberg. 1984. Lung retention of antimony and arsenic in hamsters after the intratracheal instillation of industrial dust. Scand. J. Work Environ. Health 10(4):245–251. Lide, D.R. 1991–1992. Handbook of Chemistry and Physics, 72nd Ed. Boca Raton, FL: CRC Press. Lüdersdorf, R., A.Fuchs, P.Mayer, G.Skulsuksai, and G.Schäcke. 1987. Biological assessment of exposure to antimony and lead in the glass- producing industry. Int. Arch. Occup. Environ. Health 59(5):469–474. McCallum, R.I. 1963. The work of an occupational hygiene service in environmental control. Ann. Occup. Hyg. 6(2):55–64. McCallum, R.I. 1967. Detection of antimony in process workers' lungs by x-radiation. Trans. Soc. Occup. Med. 17(4): 134–138. Muhle H., B.Bellmann, O.Creutzenberg, U.Henrich, M.Ketkar, and R.Mermelstein. 1990. Dust overloading of lungs after exposure of rats to particles of low solubility: Comparative studies. J. Aerosol. Sci. 21(3):374–377. Myers, R.C., E.R.Homan, C.S.Weil, and G.A.Webb. 1978. Antimony Trioxide Range-Finding Toxicity Studies. Carnegie-Mellon Institute of Research, Carnegie-Mellon University, Pittsburgh, PA. Sponsored by Union Carbide Corp., Danbury, CT. EPA/OTS Document #878210813. Newton, P.E., H.F.Bolte, I.W.Daly, B.D.Pillsbury, J.B.Terrill, R.T.Drew, R.Ben-Dyke, A.W.Sheldon, and L.F.Rubin. 1994. Subchronic and chronic inhalation toxicity of antimony trioxide in the rat. Fundam. Appl. Toxicol. 22(4):561–576. Potkonjak, V., and M.Pavlovich. 1983. Antimoniosis: A particular form of pneumoconiosis. I. Etiology, clinical and X-ray findings. Int. Arch. Occup. Environ. Health 51(3): 199–207. Renes, L.E. 1953. Antimony poisoning in industry. Arch. Ind. Hyg. Occup. Med. 7:99–108. Reprotox. 1999. Subscription information service. [Online]. Available: dhttp://reprotox.org/ Ridgway, L.P., and D.A.Karnofsky. 1952. The effects of metals on the chick embryo: Toxicity and production of abnormalities in development. Ann. N.Y. Acad. Sci. 55:203–215. 

ANTIMONY TRIOXIDE 261 Rossi, F., R.Acampora, C.Vacca, S.Maione, M.G.Matera, R.Servodio, and E. Marmo. 1987. Prenatal and postnatal antimony exposure in rats: Effect on vasomotor reactivity development of pups. Teratog. Carcinog. Mutagen. 7(5):491–496. Schnorr, T.M., K.Steenland, M.J.Thun, and R.A.Rinsky. 1995. Mortality in a cohort of antimony smelter workers. Am. J. Ind. Med. 27 (5):759–770. Smyth, H.F., and C.P.Carpenter. 1948. Further experience with range finding test in the industrial toxicology laboratory. J. Ind. Hyg. Toxicol. 30:63–68. Smyth, H.F., Jr., and W.L.Thompson. 1945. The Single Dose and Subacute Toxiciry of Antimony Trioxide (Sb2O3). Mellon Institute of Industrial Research, University of Pittsburgh, Pittsburgh, PA. Sponsored by Union Carbide Corp. EPA/OTS Doc #878210812. Sunagawa, S. 1981. Experimental studies on antimony poisoning. [Article in Japanese]. Igaku Kenkyu 51(3): 129–142. Watt, W.D. 1983. Chronic Inhalation Toxicity of Antimony Trioxide: Validation of the Threshold Limit Value, Dissertation, Wayne State University, Detroit, MI. White, G.P., Jr., C.G.T.Mathias, and J.S.Davin. 1993. Dermatitis in workers exposed to antimony in melting process. J. Occup. Med. 35 (4):392–395. Witschi, H.R., and J.A.Last. 1996. Toxic responses of the respiratory system. Pp. 443–462 in Casarett & Doull's Toxicology: The Basic Science of Poisons, 5th Ed. C.D.Klaassen, M.O.Amdur, and J.Doull, eds. New York: McGraw-Hill.