4
TOXICITY AND RELATED DATA ON SELECTED CADMIUM COMPOUNDS

FACED WITH THE TASK of evaluating the potential toxicity of zinc cadmium sulfide (ZnCdS), a compound with largely unknown toxic potential but reasonably well-known physical and chemical properties, the subcommittee considered it to be prudent to examine toxicity and related data on the most-toxic element in ZnCdS, cadmium. The toxicity of zinc, its interaction with cadmium, and the toxicity of copper and silver are discussed in Appendix D. This chapter reviews the physical and chemical properties, toxicokinetics, and toxicity of cadmium and cadmium compounds.

It should be noted that a substance can be insoluble in water or acids in vitro but soluble in vivo; cadmium oxide (CdO) is an example. (In vivo solubility is the ability of a material to leave the site of administration and be distributed systemically to other parts of the body.) The toxic potency of cadmium compounds depends on their in vivo solubility and bioavailability. The greater the solubility of the cadmium compound, the greater its systemic toxic potency. However, it is less clear whether water



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests 4 TOXICITY AND RELATED DATA ON SELECTED CADMIUM COMPOUNDS FACED WITH THE TASK of evaluating the potential toxicity of zinc cadmium sulfide (ZnCdS), a compound with largely unknown toxic potential but reasonably well-known physical and chemical properties, the subcommittee considered it to be prudent to examine toxicity and related data on the most-toxic element in ZnCdS, cadmium. The toxicity of zinc, its interaction with cadmium, and the toxicity of copper and silver are discussed in Appendix D. This chapter reviews the physical and chemical properties, toxicokinetics, and toxicity of cadmium and cadmium compounds. It should be noted that a substance can be insoluble in water or acids in vitro but soluble in vivo; cadmium oxide (CdO) is an example. (In vivo solubility is the ability of a material to leave the site of administration and be distributed systemically to other parts of the body.) The toxic potency of cadmium compounds depends on their in vivo solubility and bioavailability. The greater the solubility of the cadmium compound, the greater its systemic toxic potency. However, it is less clear whether water

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests solubility is a major factor in determining the carcinogenicity of cadmium compounds. The bioavailability of a compound relies heavily, if not exclusively, on its physical and chemical properties, particularly its solubility. The bioavailability of a metal ion of a metal compound is a key determinant of the toxicity of the compound. This must be taken into account whenever an attempt is made to extrapolate the quantitative and qualitative effects of a metal compound with, say, high solubility and known toxicity to a compound with unknown toxicity, such as ZnCdS that contains the same metal but is poorly or not at all soluble in water or acid. Because ZnCdS is neither water-soluble nor apparently bioavailable, the subcommittee believes that the use of toxicity data on cadmium compounds to estimate the toxicity of ZnCdS constitutes a worst-case-scenario; this approach will lead to an overestimate of the risk associated with ZnCdS. The Army, in its risk-assessment reports on ZnCdS exposures, considered cadmium as a surrogate. Appendix G contains, in more detail, the subcommittee's evaluation of this approach. The Environmental Protection Agency, Agency for Toxic Substances and Disease Registry (ATSDR), and the Center for Disease Control reviewed the Army's risk-assessment reports; and Appendix H contains the subcommittee's review of those evaluations. PHYSICAL AND CHEMICAL PROPERTIES OF CADMIUM COMPOUNDS The metal cadmium is insoluble in water (Weast 1985). Cadmium sulfide, CdS, has low solubility; its reported solubility limit is 1.3 mg/L (9.0 mol/ m3) at 20°C (Weast 1985). The sulfate and chloride salts of cadmium have very high solubilities in water (Weast 1985): cadmium chloride, CdCl2, has a reported solubility limit of 1,680,000 mg/L at 20°C and cadmium sulfate, CdSO4, a reported solubility limit of 608,000 mg/L at 20°C. Table 4-1 shows the water-solubility ranges of selected cadmium compounds. It is generally assumed that the toxicity of metal compounds usually involves an interaction between the free metal ion and a target tissue (Goyer 1995). Inorganic metal compounds that are soluble in water or

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests can be made soluble in biologic fluids or in cells (such as alveolar macrophages) are typically more toxic than are metal compounds with poor solubility. Cadmium compounds are no exception. They have a wide range of water solubility, from 1.3 mg/L for CdS to 1,680,000 mg/L for CdCl2 (see Table 4-1). The ATSDR toxicological profile on cadmium shows that cadmium compounds highly soluble in vivo (such as CdCl2, CdSO, and cadmium oxide, CdO) are considerably more toxic and can cause death at substantially lower concentrations than cadmium compounds poorly soluble in vivo (such as CdS and cadmium carbonate, CdCO3). The only oxidation state of importance for cadmium under environmental conditions is +2. In the presence of sulfide ions and under reducing conditions, CdS is formed over a wide pH range. In aqueous systems, water hardness and pH determine the speciation of cadmium. In basic systems, cadmium hydroxide, Cd(OH)2, can precipitate. In fresh water at typical environmental pH of 6-8, Cd2+ is the predominant species (Bodek and others 1988). The resulting precipitation of CdS in reducing environments can control the effective solubility of cadmium in natural waters. TABLE 4-1 Water Solubilities of Selected Cd Compounds Name Formula Water Solubility at 20°C Cadmium metal Cd Insoluble Cadmium chloride CdCl2 1,680,000 mg/L Cadmium oxide CdO 5 mg/L, but soluble in lung Cadmium sulfide CdS 1.3 mg/L Cadmium carbonate CdCO3 Insoluble Cadmium red 70% Cd, 16% Se, 13% S Insoluble Cadmium yellow 77% Cd, 22% S, 0.3 % Zn Insoluble Cadmium selenide CdSe Insoluble Cadmium sulfate CdSO4 608,000 mg/L Zinc Cadmium Sulfide ZnCdS Insoluble   Source: Adapted from Weast 1985.

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests TOXICOKINETICS OF CADMIUM COMPOUNDS INSOLUBLE CADMIUM COMPOUNDS Klimisch (1993) studied the lung deposition, lung clearance, and renal accumulation of inhaled CdS in rats. Rats were exposed to CdS at 0.2, 1.0, and 8.0 mg/m3 6 h/day for 10 d. About 40% of the lung burden was cleared rapidly (t1/2 = 1.4 d) and 40% slowly (t1/2 = 42 d), leaving a residue of 20% of the initial lung burden 90 d after the end of the exposure. Only 1% of the CdS cleared from the lungs accumulated in the kidneys. Klimisch then used the data of Loeser (1974) on the amount of cadmium accumulated in the kidneys after oral exposures to CdS to calculate an intestinal-absorption factor of 0.02% for CdS. The absorption factor for inhaled CdS based on the data of Rusch and others (1986) was 0.070.1%. That contrasts with results of similar studies with CdCl2 (which is water-soluble) in which 35% of the material cleared the lungs and accumulated in the kidneys. The author concluded that the bioavailability of cadmium from CdS is much lower than that from the more-soluble forms of cadmium, such as CdCl2. A study on the toxicokinetics of inhaled CdS (Oberdörster 1990; Oberdörster and Cox 1990) provides evidence that inhaled CdS is not absorbed from the lungs into systemic circulation. Rats were given CdS by inhalation. Failure to detect cadmium in their kidneys, the primary target organs for cadmium distribution and toxicity, suggested that the cadmium in the inhaled particles did not leave the lungs, enter the bloodstream, and affect the kidneys. No inflammatory response to CdS was detected in the lung after exposure to the same amount of substance as had caused an inflammatory response to CdCl2 and CdO. Of the inhaled CdS, 75% was exhaled immediately. Oberdörster (1990) estimated that 90% of the remaining inhaled CdS was removed slowly from the lung into the gastrointestinal tract by normal mechanical clearance processes. Rusch and others (1986) exposed rats for 2 h to various cadmium compounds at 100 mg/m3 (based on cadmium content) and found that insoluble forms of cadmium, such as cadmium red and cadmium yellow (furnace-treated dyes that contain cadmium in the form of CdS and cadmium selenide, CdSe), were not transported from lung to kidney. For example, cadmium yellow consists

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests of crystals containing 77% cadmium, 22% sulfur, and less than 1% zinc and selenium; after inhalation exposure at high concentrations, kidney cadmium content was indistinguishable from that in controls. SOLUBLE CADMIUM COMPOUNDS Soluble cadmium compounds can be absorbed from the skin, intestinal tract, or respiratory tract into the bloodstream and transported throughout the body with the potential for causing systemic injury. When such compounds are absorbed, the most-sensitive sites for injury are the lungs, kidneys, and bones; studies have also been conducted on the effects of soluble cadmium compounds on the immune system and the reproductive and developmental system. TOXICITY OF CADMIUM COMPOUNDS NONCANCER EFFECTS The subcommittee reviewed the toxicity of in vivo soluble and insoluble cadmium compounds. Several comprehensive reviews on the toxicity of cadmium compounds are available (ATSDR 1993; IARC 1993; OSHA 1992; IPCS 1992). The main focus of the subcommittee's review was to evaluate the toxicity associated with acute inhalation exposures to cadmium, because the exposures from the use of ZnCdS as a tracer particle were brief and sporadic. These data are summarized below. IN VIVO INSOLUBLE CADMIUM COMPOUNDS Many animal studies have identified toxic responses to the inhalation or ingestion of soluble cadmium compounds, but relatively few studies have examined the health effects of cadmium compounds that are poorly soluble in vivo, such as CdS, CdSe, CdCO3, or cadmium dyes. Nevertheless, enough experimental information is available to show that the toxicity of

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests CdS is quite different from the toxicity of soluble cadmium compounds. The main differences are that it takes considerably higher CdS concentrations to produce lung lesions and that substantially less cadmium becomes bioavailable from inhaled or ingested CdS than from the more-soluble cadmium compounds. Studies that compared the potential inflammatory response to intra-tracheal instillation of 30 µg of different cadmium compounds in rats (CdCl2, CdO, and CdS) reported that CdS failed to elicit any substantial influx of neutrophils, lavagable protein and enzyme activities in bronchoalveolar lavage fluid 24 h after exposure, whereas the other cadmium compounds tested resulted in substantial changes in these characteristics (Oberdörster and others 1985). When rats inhaled CdS at 1,000 µg/m3 22 h/d, 7 d/wk for 30 d, a mild inflammatory response was observed in lungs, but recovery had occurred 2 mo after exposure (Glaser and others 1986). IN VIVO SOLUBLE CADMIUM COMPOUNDS LUNG—In humans, high concentrations of CdO fumes or dusts (milligrams per cubic meter) are irritating to the respiratory tract; irritation does not occur following lower inhalation exposures. (CdO is insoluble in water; however, it is soluble in vivo.) Long-term chronic occupational exposure can cause emphysema in humans. In experimental animals, inhalation of soluble cadmium compounds has been shown to produce acute lung injury (chemical pneumonitis) and chronic lung injury (fibrosis and emphysema). In an inhalation study by Buckley and Bassett (1987), the lowest-observed-adverse-effect level (LOAEL) resulting in mild, reversible pulmonary inflammation in rats exposed to CdO was 500 µg/m3 (the equivalent of cadmium at about 440 µg/m3) following a 3-h exposure. In a similar study by Grose and others (1987), the highest air concentration of CdCl2 or CdO that did not result in detectable lung lesions in rats and rabbits exposed for 2 h was the equivalent of cadmium at 450 µg/m3; this concentration was considered a no-observed-adverse-effect level (NOAEL) (ATSDR 1993). Thus, for 2 h, an exposure at 450 µg/m3 (900

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests µg-hr/m3) was a NOAEL in the studies by Grose and others (1987), and exposure for 3 h at 440 µg/m3 (1,320 µg-hr/m3) was a LOAEL in the Buckley and Bassett study (1987). The two studies clearly define that for inhalation exposure to soluble cadmium compounds, the NOAEL for the respiratory tract in the rat is cadmium at 900-1,300 µg-hr/m 3. KIDNEY—The kidney is a major target organ for cadmium toxicity after chronic, low-level exposure to forms of cadmium that can be absorbed by the body and transported to the kidney. Workers occupationally exposed to cadmium and cadmium compounds (for example, CdO dust and cadmium fumes in factories producing nickel-cadmium batteries) suffer from a high incidence of abnormal renal function, indicated by proteinuria and a decrease in glomerular filtration rate (Falck and others 1983; Friberg 1950; Thun and others 1989). (For an extensive review of the literature on effects of cadmium on the kidney, see ATSDR 1993.) An exposure to cadmium that produces a kidney concentration of 200 µg/g wet weight is considered a threshold for renal dysfunction in an adult population chronically exposed to cadmium (Friberg and others 1974; Kjellström and others 1977, 1984; Roels and others 1983). Cadmium concentrations in the kidneys of North American adults are about 20 µg/g wet weight in nonsmokers and 40 µg/g in smokers (Chung and others 1986). ATSDR (1993) has compiled a list of animal studies whose results yield NOAELs for renal effects of cadmium exposure. The NOAELs for acute exposure, which is the type of exposure of concern to the subcommittee, were considered for inhalation and oral routes. No acute-inhalation studies of cadmium toxicity to the kidney were listed; acute oral exposures to CdCl2 as high as 150 mg/kg for 1 d in rats were reported to have no renal effect (Kotsonis and Klaassen 1977). Chronic exposure of humans at 0.0021 mg/kg per day for a lifetime has been considered to have no renal effect (Nogawa and others 1989). ATSDR has set a chronic inhalation minimal-risk level (defined as an estimate of the highest daily human exposure to a chemical at a dose that is likely to be without appreciable risk of adverse noncancer effects over a specified duration of exposure) for cadmium-induced renal toxicity of 0.2 µg/m3 (ATSDR 1993). OSHA's

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests estimates of risk posed by exposure to a time weighted average (TWA) of 5 µg/m3 range between 14 and 23 excess cases of kidney dysfunction per 1,000 workers (OSHA 1992). BONE—Cadmium affects calcium metabolism by decreasing the rate of bone formation and increasing calcium loss from bone. The relevant studies have been reviewed by Bhattacharyya and others (1995). The effects of low-dose exposures on bone metabolism are subtle, but chronic exposures, particularly in combination with other factors, lead to calcium loss from bone. Factors contributing to the severity of cadmium toxicity to bone include renal disease, dietary deficiencies (particularly low dietary calcium), vitamin D deficiency, and low estrogen concentrations. As in renal toxicity, cadmium toxicity to bone results from exposure to forms of cadmium that are soluble enough to allow it to reach the bone. No published studies have shown bone loss from short-term high-level exposures to cadmium. IMMUNESYSTEM—The effects of cadmium on the immune system of animals or humans are inconsistent (Cifone and others 1990; Exon and Koller 1986; Funkhouser and others 1994; Horiguchi and others 1993; Kastelan and others 1981; Koller 1973), and the in vitro effects do not correlate well with the in vivo effects. Cadmium stimulated the immune response in some studies and suppressed it in others. A NOAEL cannot be determined until a definite immune pattern is characterized for this chemical. DEVELOPMENTALANDREPRODUCTIVESYSTEM—ATSDR has published a toxicological profile on cadmium (ATSDR 1993) that includes an excellent review of the reproductive and developmental toxicity of cadmium. We will not reiterate all of the reported studies but highlight here the ones most relevant for this report. It should be noted that most studies have focused on occupational exposure of male workers to cadmium. Noticeably absent are environmental studies of cadmium exposure that include potentially sensitive populations, such as pregnant or nursing women, children, and the elderly. Of most interest to the subcommittee

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests were reports of the effect of acute exposures to cadmium on the developmental and reproductive systems, but such studies were not available. DEVELOPMENTALTOXICITY—Two epidemiologic studies looked at pregnancy outcomes in occupationally exposed populations (ATSDR 1993); the primary route of exposure was probably inhalation. A Russian study observed lower birthweights of children in cadmium-exposed women (exposures up to 35 mg/m3) than in unexposed controls; however, the decrease was not correlated with duration of maternal exposure to cadmium (Tsvetkova 1970). Dose-response information was not available for this study. A study of French women occupationally exposed to heavy metals showed doubled concentrations of cadmium in the hair of mothers and newborns relative to matched controls (1.45 versus 0.59 ppm in exposed mother's hair and 1.27 versus 0.53 ppm in transplacentally exposed newborn hair) but did not show any adverse effects in the newborns (Huel and others 1984). Loiacono and others (1992) conducted a prospective study of women living around a lead smelter plant in Yugoslavia. Women were believed to have been exposed to lead and cadmium dust from the smelter. The study's hypothesis was that placental cadmium accumulation was associated with decreased birthweight. Multiple regression was used to control for potential confounders of birthweight and gestation, such as smoking; no association was found between placental cadmium and birthweight or gestational age. Cadmium has been shown to cause developmental toxicity in rodents after inhalation. Such effects as delay in ossification, decrease in locomotor activity, and impairment of reflexes were noted after exposures to CdO as low as 0.02 mg/m3, 5 h/d 5 d/wk for 5 mo before mating and from the 1st to the 20th day of pregnancy (Baranski 1985). No adverse effects on maternal homeostasis were reported for this exposure, but maternal toxicity was significant at the two higher exposures tested (CdO at 0.16 and 1.0 mg/m3). Therefore, one can infer from this study that the ambient concentrations of CdO might cause developmental toxicity in humans. However, CdO is soluble in vivo whereas ZnCdS is insoluble in vivo; therefore, the results of Baranski (1985) cannot be extrapolated to inhala-

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests tion exposures involving ZnCdS. After exposures of female pregnant rats to CdO at 0.16 mg/m3, decreases in weight gain, osteogenesis, and viability were also observed in offspring. Both maternal and fetal weight changes were noted after exposure to CdCl2 for 21 d at 0.6 mg/m3 (Prigge 1978). Numerous rodent studies have shown that oral exposure to soluble cadmium can reduce fetal or pup weights and cause skeletal malformations (ATSDR 1993). Oral exposure to cadmium compounds soluble in water has been shown to cause reduced fetal or pup weights (ATSDR 1993; Ali and others 1986; Baranski 1987; Kelman and others 1978; Petering and others 1979; Pond and Walker 1975; Sorell and Graziano 1990; Sutou and others 1980; Webster 1978; Whelton and others 1988). A few of those studies have also shown malformations, especially of the skeleton (Baranski 1985; Machemer and Lorke 1981; Schroeder and Mitchener 1971). Neurodevelopmental effects in rats have been identified as the most sensitive end point in these oral-exposure studies with cadmium compounds soluble in water. Baranski and others (1983) reported reduced locomotor activity and impaired balance in pups from dams exposed to 0.04 mg/kg per day before and throughout gestation, and Ali and others (1986) reported hyperactivity in offspring from dams exposed to cadmium at 0.7 mg/kg per day during gestation. Chisolm and Handorf (1985, 1987) have also postulated that cadmium exposure can play a role in pregnancy-induced hypertension and suggested that animals late in gestation might be uniquely sensitive to this effect. Those studies suggest that the ambient concentrations of CdO, a compound soluble in vivo, might cause developmental toxicity in humans. However, ZnCdS is an insoluble cadmium compound, and that suggests that it is unlikely to produce developmental toxicity at the low ambient concentrations. (See discussion earlier in this chapter about the solubility of cadmium compounds and implications for noncancer end points.) REPRODUCTIVETOXICITY—Early studies of the effect of cadmium exposure on human male fertility did not find adverse effects on fertility (Kazantzis and others 1963) or on urinary androgen excretion (Favino and others 1968). However, those early studies included only 12 and 10 occupationally exposed men, respectively, so the findings need to be viewed as preliminary.

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests Limited negative evidence from two occupational epidemiologic studies suggests that inhalation exposure to low doses of cadmium does not cause serious reproductive effects in men or women (Mason 1990; Tsvetkova 1970). Only a few end points were evaluated in those studies including menstrual cycling and serum hormone concentrations. Such important end points as fertility and reproductive function were not studied. Cadmium has been shown to concentrate in testicular tissue after occupational exposure (Smith and others 1960). Saaranen and others (1989) compared 24 men attending fertility clinics with 38 fertile men whose wives conceived within 6 mo to determine whether cadmium concentration in seminal fluid and sperm affected fertility. Concentrations of lead, cadmium, and zinc in the testes of 41 autopsied men were evaluated (Oldeneid and others 1993); no associations between seminal-plasma cadmium content and semen quality or fertility changes were noted, despite the ability of cadmium to reach human semen. Favino and others (1968) did not detect changes in androgen function in 10 workers in a cadmium-alkaline battery plant even though measurable amounts of urinary cadmium were detected and workers had experienced ulceration of the nasal mucosa and anosmia. DERMALTOXICITY—A search of the available reports on the toxicity of cadmium and its compounds failed to identify any studies in which skin exposures have produced serious adverse health effects in humans (ATSDR 1993). In a few studies, patients were exposed to dermally applied solutions of 2% CdCl2 in distilled water. Of 1,502 patients tested, only 25 showed a positive reaction that was interpreted as reflecting acute skin irritation rather than contact allergy to cadmium (Wahlberg 1977). One patient showed a reaction also when tested with a 1% CdCl2 solution. However, in another study involving close to 1,500 patients, a 1% CdCl2 solution caused no skin toxicity (Rudzki and others 1988). OTHERNONCANCERTOXICITY—Cardiovascular, gastrointestinal, hematologic, musculoskeletal, or hepatic effects in humans or experimental animals following exposure to cadmium have not been reported (ATSDR 1993).

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests GENOTOXICITY OF CADMIUM COMPOUNDS This section reviews the genotoxicity of soluble and insoluble cadmium compounds. A wide range of tests have been conducted to determine the genotoxicity of cadmium. It has been tested in bacteria, plants, insects, and mammalian cells, including human cells, in vitro and in vivo. Comprehensive reviews of the various investigations have been provided by Occupational Safety and Health Administration (OSHA 1992) and the Agency of Toxic Substances and Disease Registry (ATSDR). Both positive and negative results were reported. Positive mutagenicity results were generally observed with soluble cadmium compounds (such as CdCl2), whereas insoluble compounds generally yielded negative results. Positive mutagenicity were observed in some studies that used bacterial cells (Bruce and Heddle 1979; Kanematsu and others 1980; Mandel and Ryser 1984; Wong 1988) and in most studies that used yeast or mammalian cell cultures (Denizeau and Marion 1989; Oberly and others 1982). Chromosomal aberrations have been found in most studies that involved cadmium treatment of mammalian cells (Deaven and Campbell 1980; Rohr and Bauchinger 1976) and in some studies that used human lymphocytes in culture (Gasiorek and Bauchinger 1981; Shiraishi and others 1972) and bone marrow cells after intraperitoneal (Mukherjee and others 1988a) and oral (Mukherjee and others 1988b) exposure of mice. Evidence of chromosomal aberrations in humans after inhalation (Deknudt and Leonard 1975; O'Riordan and others 1978) or oral (Bui and others 1975; Tang and others 1990) exposure to cadmium compounds is conflicting. Cadmium does not appear to cause germ-cell mutations or chromosomal damage after oral (Sutou and others 1980; Zenic and others 1982) or intraperitoneal (Epstein and others 1972; Mailhes and others 1988) exposure in animals, but does so after subcutaneous exposure (Watanabe and Endo 1982; Watanabe and others 1979). Overall, cadmium appears to have the capability of altering genetic material, particularly chromosomes in mammalian cells, but germ cells appear to be protected except at high acute parenteral doses (ATSDR 1993). The subcommittee concludes that until more conclusive mutagenicity

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests studies are conducted and reported, cadmium may be considered to be a potential mutagenic agent. CARCINOGENICITY OF CADMIUM COMPOUNDS Studies on the carcinogenicity of cadmium and cadmium compounds were reviewed most recently by the International Agency for Research on Cancer (IARC) in 1993. IARC concluded that cadmium and cadmium compounds are carcinogenic in humans on the basis of epidemiologic studies of occupationally exposed workers, experimental studies in animals, and genotoxic effects in a variety of types of eukaryotic cells, including human cells (IARC 1993). A less-complete review by ATSDR in the same year characterized the evidence of carcinogenicity in animals as ''strong'' but of that in humans as "limited" (ATSDR 1993). Most of the evidence of carcinogenicity involves lung cancer after long-term inhalation. LUNG CANCER Epidemiologic studies show some increase in lung cancer among workers in 5 of 6 occupationally exposed populations in Europe and the United States (IARC 1993). Workers were exposed predominantly to cadmium as CdO (fume or dust) or CdS. The human evidence linking cadmium to lung cancer is strongest in the U.S. plant where exposures were well characterized, and the persistent dose-response relationship with cadmium cannot be readily explained by cigarette-smoking or other known occupational lung carcinogens (IARC 1993). Epidemiologic studies of cancer in workers exposed to cadmium have been conducted in five plants in the United States, England, and Sweden (Tables 4-2 and 4-3). Two of the plants (groups 1 and 2) are metallurgic facilities, two (groups 4 and 5) are nickel-cadmium battery manufacturers, and one (group 3) is a compendium of 17 plants in the U.K., including a large zinc smelter. The mortality experience of these populations has been studied repeatedly (Elinder and others, 1985); Tables 4-2 and 4-3 present

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests the most recent results. A statistically significant increase in mortality from lung cancer has been reported among U.S. cadmium-recovery workers with 2 or more years of employment (Thun and others 1985, Thun 1990), in nickel-cadmium battery workers in the U.K. (Sorahan 1987) and Sweden (Elinder and others 1985; Jarup and others 1990), and in the British 17-plant study (Kazantzis and Armstrong 1982; Kazantzis and others 1988). TABLE 4-2 Lung-Cancer Mortality Among Cadmium Workers Group Type of Industry Deaths Observed/ Expected Standardized Mortality Ratio 95% Confidence Interval 1 Cadmium recovery plant, U.S. (Thun and others 1990) —Overall cohort —2+yr 24/17.50 24/10.76 137 223 (88-204) (143-332) 2 Cadmium-copper alloy plant, U.K. (Kazantzis and Armstrong 1982) 47/46.4 101 (72-130) 3 17 plants, U.K. (Kazantzis and others 1988) 277/240.9 115 (101-129) 4 Nickel-cadmium battery, U.K. (Sorahan 1987) 110/84.5 130 (107-157) 5 Nickel-cadmium battery, Sweden (Jarup and others 1990) 14/5.8 241 (132-405)a a 20 years of employment, compared with regional rates. Table 4-3 indicates those cohorts for which a dose-response relationship is present between cadmium exposure and mortality from lung cancer. In three cohorts (Thun and others 1990; Sorahan 1987; Kazantzis and others 1988), the standardized mortality ratio (SMR) for lung cancer increases with either length of employment or cumulative exposure to cadmium.

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests TABLE 4-3 Dose-Response Relationships for Lung Cancer in Studies of Cadmium-Exposed Workers Group Type of Industry Dose-Response Relationship Evident Range in Standardized Mortality Ratio 1 Cadmium recovery plant, U.S. (Thun and others 1990) Yes 46, 263,373a 2 Cadmium-copper alloy plant, U.K. (Kazantzis and Armstrong 1982) No 87, 114, 72 3 17 plants, U.K. (Kazantzis and others 1988) Yes 112,121,194b 4 Nickel-cadmium battery, U.K. (Sorahan 1987) Possibly (Marginal trend before 1947)c 5 Nickel-cadmium battery, Sweden (Jarup and others 1990) No 234, 232d a Death rates in cadmium workers compared with men in Colorado, followup to 1984. b Author does not attribute to cadmium. c Latency not considered. d Regional rates, 20-yr latency. Despite the value of the occupational studies in providing information about the potential carcinogenicity of cadmium, there are major differences in the intensity and duration of exposure between the communities exposed to ZnCdS by the Army and the occupational conditions. For example, at the U.S. cadmium-recovery plant, air concentrations of cadmium averaged 1,000 µg/m3 before 1950 and 500 µg/m3 from 1950 to 1960. Those measures are average concentrations breathed over an 8-h workday, often for decades. Because of the high cumulative exposures to cadmium, workers at the U.S. plant often developed cadmium-induced kidney toxicity (low-molecular-weight proteinuria) and obstructive lung disease even without smoking (Thun and others 1985). CdS can produce lung tumors in rats. Rats exposed to 10 weekly intratracheal instillations of 250 µg of CdS had a statistically significant in-

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests crease in the incidence of lung tumors (OSHA 1992; Pott and others 1987). In inhalation studies, a statistically significant increase in number of lung tumors (bronchioalveolar, adenocarcinoma, and squamous cell tumors) was seen in rats exposed 22 h/day, 7 d/wk for 18 mo at concentrations of 90 µg/m3 or higher (Oldiges and others 1989). Later studies by König and others (1992) indicated that 50-63% of CdS used in the inhalation studies of Oldiges and others (1989) might have been converted to the more-soluble CdSO4, which is known to be a pulmonary carcinogen in rats. Heinrich and others (1989), who exposed both mice and hamsters by inhalation to CdS at concentrations of 90-1,000 µg/m3 for up to 64 wk, failed to identify any statistically significant increase in incidence of lung tumors but did report increases in lipoproteinosis, fibrosis, and hyperplasia. A better understanding of the molecular and cellular mechanisms of cadmium carcinogenicity is required to determine how different the carcinogenic potential of CdS is from that of more-soluble cadmium compounds. Human data from studies of occupationally exposed workers have been used to estimate cancer risk associated with exposure to cadmium (OSHA 1992). On the basis of simple linear regression, lung-cancer risk is estimated to increase with cumulative cadmium exposure by 14.6 x 10-6 per µg/m3. That estimate is used in risk calculation in Chapter 6. PROSTATIC CANCER Epidemiologic studies of cadmium and prostatic cancer are inconclusive. Early studies of occupational exposure suggested increased deaths from prostatic cancer, but the association weakened with continued followup and has not shown a clear dose-response trend (IARC 1993). Japanese men who ingested cadmium-contaminated rice during World War II experienced higher mortality from prostatic cancer (standardized mortality ratio, 1.66) and a higher incidence of prostatic hyperplasia in one polluted area but not in another (Shigematsu and others 1982). The high prevalence of cadmium-induced kidney toxicity (30-80%) in both areas indicated systemic circulation of cadmium (IPCS 1992). Animal experiments do not show that inhaled or ingested cadmium induced prostatic cancer.

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests OTHER CANCERS Experimental studies of animals have not consistently shown other types of cancer after cadmium inhalation or ingestion (IARC 1993). Leukemia can be produced by cadmium ingestion in zinc-deficient rats; it can be prevented by simultaneous zinc supplementation (IARC 1993). Epidemiologic studies in the Toyama prefecture of Japan, where residents ate highly contaminated rice, found no evidence of increased death from stomach or liver cancer or all cancers combined (Shigematsu and others 1982). No increase in other cancers about which concern was expressed in the public meetings on ZnCdS has been observed in the highly exposed communities in Japan. (IPCS 1992). CONCLUSIONS Faced with the task of evaluating the potential toxicity of ZnCdS, a substance with largely unknown toxic potential but reasonably well-known physical and chemical properties, the subcommittee considered it to be prudent to examine the toxicity and related data on the most-toxic element in ZnCdS, cadmium. On the basis of a review of the physical and chemical properties, toxicokinetics, and the toxicity of cadmium compounds, the subcommittee reached the following conclusions: Bioavailability of cadmium is a key determinant of the toxicity of a cadmium compound. Cadmium compounds differ in their toxicity: compounds that are insoluble in vivo are less toxic than soluble ones. Inhaled particles of ZnCdS would have direct contact with lung tissue but minimal systemic absorption due to poor solubility. Its lack of solubility also suggests that it is highly unlikely that free cadmium ions would become bioavailable to target organs as a result of inhalation of ZnCdS. Toxicokinetic studies with CdS showed that it is much less bioavailable than cadmium compounds that are soluble in vivo. Cadmium compounds that are soluble in vivo are toxic primarily to the lungs, kidneys, and bone. There are very few reported studies on the toxicity of cadmium compounds that are poorly soluble in vivo, such as

OCR for page 39
Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests CdS. In several studies, CdS failed to elicit lung or other target-organ toxicity in experimental animals, presumably because it is not bioavailable. Because ZnCdS is neither water-soluble nor apparently bioavailable, the subcommittee believes that exposure to ZnCdS is not likely to produce adverse health effects and that the use of toxicity data on soluble cadmium compounds to estimate the potential for noncancer toxic effects of ZnCdS would constitute a worst-case scenario. Thus, this approach will overestimate the risk. Noncancer-toxicity data are available on soluble and insoluble cadmium compounds, but the only cancer-potency data available on cadmium are based on relatively high occupational exposures to cadmium compounds, mostly of CdO or mixtures of CdO and CdS. Inhaled cadmium has been shown in occupational studies and laboratory studies of animals to cause lung cancer but not cancer at other body sites. Cadmium exposures associated with increased lung-cancer risk in human and animal studies were to much higher concentrations for longer periods and involved more biologically soluble compounds than the exposures to cadmium from ZnCdS in the Army's testing program. A quantitative risk assessment for lung cancer based on occupational exposure of humans to cadmium compounds is likely to overestimate the lung-cancer risk for ZnCdS exposures from the Army's dispersion tests.