OSHA comments from the January 19, 1989 Final Rule on Air Contaminants Project extracted from 54FR2332 et. seq. This rule was remanded by the U.S. Circuit Court of Appeals and the limits are not currently in force.

CARBON TETRACHLORIDE

CAS: 56-23-5; Chemical Formula: CCl4

      The current OSHA PELs for carbon tetrachloride are 10 ppm as an 8-hour TWA, 25 ppm as a STEL not to be exceeded for more than five minutes every four hours, and 200 ppm as a ceiling. OSHA proposed to revise these limits to a single limit of 2 ppm measured over 60 minutes, based on the NIOSH (1975a/Ex. 1-186) REL. The ACGIH has established a 5-ppm 8-hour TWA limit, with a skin notation, for this substance. Carbon tetrachloride is classified as a probable human carcinogen by EPA (Group B2) and IARC (Group 2B), and as a suspected human carcinogen by the ACGIH (Category A2), based on positive carcinogenicity studies in rats, mice, and hamsters. In the final rule, OSHA is establishing a 2-ppm 8-hour TWA limit for carbon tetrachloride. Carbon tetrachloride is a heavy, mobile liquid with a sweet odor.

       In humans, there have been three case reports of liver tumors developing after carbon tetrachloride exposure (Tracy and Sherlock 1968/Ex. 1-152; Johnstone 1948/Ex. 1-817; Simler, Maurer, and Mandard 1964/Ex. 1-225). In each case, the patient had been acutely overexposed to carbon tetrachloride, leading to nausea, stomach pains, and signs of severe liver damage.

       Blair, Decoufle, and Grauman (1979/Ex. 1-150) studied causes of death in 330 laundry and dry cleaning workers potentially exposed to carbon tetrachloride, as well as to trichloroethylene and tetrachloroethylene. Causes of death based on death certificates were compared to the age, sex, race, and cause-specific distribution of U.S. deaths from the same time period. The proportionate mortality ratio (PMR) for all malignant neoplasms was 128, which was statistically significant, indicating that the study group had a 28-percent higher proportion of total deaths due to cancer compared with the U.S. general population. The excess cancer deaths were due to liver, lung, and cervical cancer and leukemia. Although the excess lung and cervical cancer may reflect socioeconomic differences among these workers, the excess liver cancer seen in this study is consistent with findings in animal studies on carbon tetrachloride.

       In animals, carbon tetrachloride has produced hepatocellular carcinomas in all species evaluated (rats, mice, and hamsters). Male rats were given 47 or 94 mg/kg carbon tetrachloride and females were given 80 or 159 mg/kg by gavage for 78 weeks (NCI 1976a/Ex. 1-119; NCI 1976b/Ex. 1-168; NCI 1977b/Ex. 1-169). The incidence of hepatocellular carcinomas was increased in animals exposed to carbon tetrachloride as compared with pooled colony controls but was statistically significant only for low-dose females. The lower incidence of carcinomas in female rats at the high dose (1/49) compared to the low dose (4/49) was attributed by the authors to the increased lethality that occurred among these rats before tumors could be expressed.

       In this same study, mice of both sexes received 1250 or 2500 mg/kg carbon tetrachloride by gavage. Hepatocellular carcinomas were found in 49/49 low-dose and 47/48 high-dose males (compared with 5/77 in the control males) and in 40/40 low-dose and 43/45 high-dose females (compared with 1/80 in the control females) (NCI 1976a/Ex. 1-119; NCI 1976b/Ex. 1-168; NCI 1977b/Ex. 1-169).

       Edwards, Heston, and Dalton (1942/Ex. 1-68) administered carbon tetrachloride by gavage (64 mg/mouse administered 46 times over four months) to a mouse strain known to have a low incidence of spontaneous hepatomas. The incidence of hepatomas was 52 percent (28/54) for males and 32 percent (6/19) for females. Previous hepatoma incidence data for untreated mice of this strain were 2/71 for males and 0/81 for females. Carbon tetrachloride administered by gavage has also been shown to produce neoplastic changes in the livers of four additional strains of mice (Andervont 1958/Ex. 1-81; Edwards 1941/ Ex. 1-86; Eschenbrenner and Miller 1943/Ex. 1-113).

       Della Porta, Terracini, and Shubik (1961/Ex. 1-136) gave weekly gavage treatments of 10 to 20 ug to hamsters for 30 weeks, and the animals were observed for an additional 25 weeks. All 10 hamsters dying or killed between weeks 43 and 55 had liver cell carcinomas, in comparison with 0/254 in historical controls.

       Risk estimate for carbon tetrachloride. Three data sets have sufficient dose-response information to allow quantitative risk estimation: the rat and mouse bioassay data (NCI 1976a/ Ex. 1-119; NCI 1976b/Ex. 1-168; NCI 1977b/Ex. 1-169) and the Edwards, Heston, and Dalton (1942/Ex. 1-68) mouse data. To increase sample sizes, the data were pooled for male and female animals in each of the three studies. (In the NPRM, OSHA erroneously indicated that four data sets were pooled; see Ex. 110.) The estimated risk presented in Table C15-5 is the geometric mean of the risk calculated from each of the three data sets.

       Inhalation risk was calculated assuming an air intake of 20 m3 per 24-hour day and a 40-percent absorption rate for humans (EPA 1984a/Ex. 1-1130). All four studies suggest that a common biological mechanism, cell death and regeneration, occurs and leads to the development of the same tumor type.

       Table C15-5 presents the estimates of lifetime human risk from carbon tetrachloride exposure, calculated by the linearized multistage model (GLOBAL83), at the final rule's 2-ppm limit, the ACGIH limit of 5 ppm, and the former 10-ppm OSHA PEL. Both the maximum likelihood estimates (MLE) and the 95-percent upper-confidence limits of human risk are given, as well as the corresponding expected number of excess cancer deaths per 1,000 workers exposed over a working lifetime.

       Based on this risk estimate, the MLE at the former OSHA limit of 10 ppm is 17.9 excess deaths per 1,000 exposed workers, clearly indicating that a significant cancer risk exists at the former PEL.

       Risk at the current ACGIH limit of 5 ppm is estimated to be 9.2 excess deaths per 1,000 workers exposed over their working lifetimes. At the final rule's limit of 2 ppm, residual risk continues to be significant, according to the Supreme Court's guidance in the Benzene decision and the analysis presented in the introduction to this section; the risk predicted at 2 ppm is 3.7 excess deaths per 1,000 workers exposed over their working lifetimes. However, risk at the 2-ppm limit is substantially reduced compared with risk at the former OSHA PEL of 10 ppm. The estimate shows that approximately 14 cancer deaths per 1,000 workers would potentially be avoided over a lifetime by reducing the limit to 2 ppm.

       Both NIOSH (Ex. 193) and the AFL-CIO (Ex. 194) supported OSHA's proposed 2-ppm 60-minute ceiling PEL and believed that carbon tetrachloride should be designated as a potential carcinogen. The Dow Chemical Company (Ex. 3-741), however, stated that, for a number of reasons, it believes that the cancer risk from exposure to carbon tetrachloride has been overestimated. First, Dow argues that carbon tetrachloride enhances the occurrence of naturally forming liver tumors by causing increased cell death and turnover. Becauseclear threshold effect levels have been demonstrated for liver toxicity, Dow believes that a threshold-type response would be expected for carcinogenic effects "since liver toxicity appears to be a precursor to carcinogenic activity" (Ex. 3-741, p. 34). The Halogenated Solvents Industry Alliance (Ex. 8-89) also expressed the opinion that OSHA overstated the potential cancer risk byusing a linear, threshold model. Dow (Ex. 3-741) concludes that a level of 50 ppm (the threshold for liver toxicity observed in six-month inhalation studies in monkeys) represents the threshold concentration for human toxicity and carcinogenicity from exposure to carbon tetrachloride.

       In support of its position, Dow included a review of toxicity data by J.M. Norris of Dow Chemical (Ex. 3-741, Appendix A). Mr. Norris cites studies that suggest that species sensitivity to liver toxicity is related to cytochrome P-450 content in liver and that rodents have greater unit P-450 activity and are more sensitive to carbon tetrachloride-induced liver toxicity than are Rhesus monkeys. Since the unit P-450 activity of Rhesus monkeys is comparable to that of humans, Mr. Norris concludes that "the monkey may be the appropriate animal for extrapolation to man" (Ex. 3-741, Appendix A, p. 10).

       After reviewing the evidence presented by Mr. Norris, OSHA is unpersuaded that the 50-ppm no-effect level observed in monkeys should be used to establish a PEL to protect workers from the significant cancer risk associated with exposure to carbon tetrachloride. The monkey data cited by Dow and Mr. Norris are results from a study of only six months' duration, and only one or two monkeys were tested at dose levels near the no-observed-effect level. Mr. Norris acknowledged that these limitations warrant the use of a safety factor to derive an adequate exposure limit; applying an appropriate safety factor to the 50-ppm NOEL would yield a PEL no higher, and perhaps well below, the final rule's 2-ppm limit, given the seriousness of the toxicologic endpoint (carcinogenicity). OSHA concludes that the approach it has used to assess cancer risk (i.e., combining data from several animal studies to estimate risk with a widely used dose-response model) provides better information on which to base a revised PEL than do the results of a single short-term, small-sample animal study.

       Based on the evidence presented above and the quantitative estimates of carbon tetrachloride-related cancer risk, OSHA concludes that occupational exposure to carbon tetrachloride at the former 10-ppm PEL presents a significant risk of cancer to workers (13.9 cancer deaths per 1,000 workers). OSHA's risk assessment shows that reducing this limit to 2 ppm will substantially reduce this risk (3.7 deaths per 1,000 workers). The Agency concludes that cancer represents a material impairment of health and functional capacity. Accordingly, OSHA is revising its limits for carbon tetrachloride to a single limit of 2 ppm; however, in the final rule, OSHA is establishing this limit as an 8-hour TWA. OSHA has determined that a TWA limit is more appropriate for carbon tetrachloride since low-level exposure to carbon tetrachloride presents a chronic, rather than an acute, health hazard. OSHA also believes that establishing a TWA limit will simplify the development of compliance and exposure monitoring strategies for employers, since an 8-hour TWA limit is more conventional than a 60-minute limit. Therefore, in the final rule, OSHA is establishing a 2-ppm 8-hour TWA PEL for carbon tetrachloride.