The critical health effect of interest in relation to occupational exposure to oxidic nickel is, again, respiratory cancer. Unlike metallic nickel, which does not appear to be carcinogenic, and soluble nickel, whose carcinogenic potential is still open for debate, the evidence for the carcinogenicity of certain oxidic nickel compounds is more compelling. That said, there is still some uncertainty regarding the forms of oxidic nickel that induce tumorigenic effects. Although oxidic nickel is present in most major industry sectors, it is of interest to note that epidemiological studies have not consistently implicated all sectors as being associated with respiratory cancer. Indeed, excess respiratory cancers have been observed only in refining operations in which nickel oxides were produced during the refining of sulfidic ores and where exposures to oxidic nickel were relatively high (>5 mg Ni/m3) (ICNCM, 1990; Grimsrud et al., 2000). At various stages in this process, nickel-copper oxides may have been formed. In contrast, no excess respiratory cancer risks have been observed in workers exposed to lower levels (<2 Ni/m3) of oxidic nickel free of copper during the refining of lateritic ores or in the nickel-using industry.
Specific operations where oxidic nickel was present and showed evidence of excess respiratory cancer risk include refineries in Kristiansand, Norway, Clydach, Wales, and Copper Cliff and Port Colborne, Ontario, Canada. In all instances, workers were exposed to various combinations of sulfidic, oxidic, and soluble nickel compounds. Nevertheless, conclusions regarding the carcinogenic potential of oxidic nickel compounds have been gleaned by examining those workers predominantly exposed to oxidic nickel.
In the case of Kristiansand, this has been done by examining workers in the roasting, smelting and calcining department (ICNCM, 1990) and by examining all workers by cumulative exposure to oxidic nickel (ICNCM, 1990; Andersen et al., 1996). In the overall cohort, there was evidence to suggest that long-term exposure (=15 years) to oxidic nickel (mainly nickel-copper oxides at concentrations of 5 mg Ni/m3 or higher) was related to an excess of lung cancer. There was also some evidence that exposure to soluble nickel played a role in increasing cancer risks in these workers (see Section 5.3). The effect of cigarette smoking has also been examined in these workers (Andersen et al., 1996; Grimsrud, 2001), with Andersen et al., 1996 showing a multiplicative effect (i.e., interaction) between cigarette smoking and exposure to nickel. Evidence of excess nasal cancers in this group of workers has been confined to those employed prior to 1955. This evidence suggests that oxidic nickel has been a stronger hazard for nasal cancer than soluble nickel, as 12 cases (0.27 expected) out of 32 occurred among workers exposed mostly to nickel oxides.
In the Welsh and Canadian refineries, workers exposed to some of the highest levels (10 mg Ni/ m3 or higher) of oxidic nickel included those working in the linear calciners and copper and nickel plants (Wales) and those involved in sintering operations in Canada. In Wales, oxidic nickel exposures were mainly to nickel-copper oxides or impure nickel oxide; in Canada, exposures were mainly to high-temperature nickel oxide with lesser exposure to nickel-copper oxides. Unfortunately, in the latter case, oxidic exposures were completely confounded by sulfidic nickel exposures, making it difficult to distinguish be- tween the effects caused by these two species of nickel. Both excess lung and nasal cancer risks were seen in the Welsh and Canadian workers (Peto et al., 1984; Roberts et al., 1989a; ICNCM, 1990).
In contrast to the above refinery studies, studies of workers mining and smelting lateritic ores (where oxidic nickel exposures would have been primarily to silicate oxides and complex nickel oxides free of copper) have shown no evidence of nickel-related respiratory cancer risks. Studies by Goldberg et al. (1987; 1992) of smelter workers in New Caledonia showed no evidence of increased risk of lung or nasal cancer at estimated exposures of 2 mg Ni/m3 or less. Likewise, in another study of smelter workers in Oregon there was no evidence of excess nasal cancers (Cooper and Wong, 1981; ICNCM, 1990). While there were excess lung cancers, these occurred only in short-term workers, not long-term workers. Hence, there was no evidence to suggest that the lung cancers observed were related to the low concentrations (=1 mg Ni/ m3) of oxidic nickel to which the men were exposed (ICNCM, 1990).
In nickel-using industries, the evidence for respiratory cancers has also largely been negative. As noted in previous sections (Sections 5.1 and 5.2), most studies on stainless steel and nickel alloy workers that would have experienced some level of exposure to oxidic nickel have shown no significant nickel-related excess risks of respiratory cancer (Polednak, 1981; Cox et al., 1981; Cornell, 1984; Moulin et al., 1993, 2000; Svensson et al., 1989; Simonato et al., 1991; Gerin et al., 1993; Hansen et al., 1996; Jakobsson et al., 1997; Arena et al., 1998). In Swedish nickel-cadmium battery workers, there is some evidence of an increased incidence of nasal cancers, but it is not clear whether this is due to exposure to nickel hydroxide, cadmium oxide, or a combination of both (Jarup et al, 1998). In addition, little is known about the previous employment history of these workers. It is, therefore, not clear whether past exposures to other potential nasal carcinogens may have contributed to the nasal cancers observed in these workers. In contrast, no nickel-related increased risk for lung cancer has been found in these or other nickel- cadmium battery workers (Kjellström et al, 1979; Sorahan and Waterhouse, 1983; Andersson et al., 1984; Sorahan, 1987; Jarup et al., 1998).
From the overall epidemiological evidence, it is possible to speculate that the composition of oxidic nickel associated with an increase of lung or nasal cancer may primarily be nickel-copper oxides produced during the roasting and electrorefining of sulfidic nickel-copper mattes. However, careful scrutiny of the human data also reveals that high respiratory cancer risks occurred in sintering operations – where exposures to nickel-copper oxides would have been relatively low – and, possibly, in nickel-cadmium battery workers, where oxidic exposures would predominantly have been to nickel hydroxide. In addition to the type of oxidic nickel, the level to which nickel workers were exposed must also be taken into consideration. Concentrations of oxidic nickel in the high-risk cohorts (those in Wales, Norway, and Port Colborne and Copper Cliff, Canada) were considerably higher than those found in New Caledonia, Oregon, and most nickel-using industries. In the case of the nickel-cadmium battery workers, the early exposures that would have been critical to the induction of nasal cancers of long latency were believed to have been relatively high (>2 mg Ni/ m3). Hence, it may be that there are two variable – the physicochemical nature of the oxide and the exposure level – that contribute to the differences seen among the various cohorts studied.
Animal data shed some light on the matter. In the previously mentioned NTP studies, nickel oxide was administered to rats and mice in a two-year carcinogenicity bioassay (NTP, 1996c). The nickel oxide used was a green, high-temperature nickel oxide calcined at 1,350°C; it was administered to both rats and mice for 6 hours/ day, 5 days/week for 2 years. Rats were exposed to concentrations of 0, 0.5, 1.0, or 2.0 mg Ni/ m3. These concentrations are equivalent to over 5.0 to 20 mg Ni/m3 workplace aerosol after adjusting for particle size differences and animal to human extrapolation (Hsieh et al., 1999; Yu et al., 2001). After two years, no increased incidence of tumors was observed at the lowest exposure level in rats. At the intermediate and high concentrations, 12 out of 106 rats and 9 out of 106 rats, respectively, were diagnosed with either adenomas or carcinomas. On the basis of these results, the NTP concluded that there was some evidence of carcinogenic activity in rats. In contrast, there was no evidence of treatment-related tumors in male mice at any of the doses administered (1.0, 2.0 and 4.0 mg Ni/m3) and only equivocal evidence in female mice exposed to 1.0 but not 2.0 or 4.0 mg Ni/m3. Carcinogenic evidence for other oxidic nickel compounds comes from animal studies using routes of exposure that are not necessarily relevant to man (i.e.,intratracheal instillation, injection). In these studies, nickel-copper oxides appear to be as potent as nickel subsulfide in inducing tumors at injection sites (Sunderman et al., 1990). There is, however, no strong evidence to indicate that black (low temperature) and green (high temperature) nickel oxides differ substantially with regard to tumor-producing potency. Some forms of both green and black nickel oxide produce carcinogenic responses, while other forms have tested negative in injection and intratracheal studies (Kasprzak et al., 1983; Sunderman, 1984; Sunderman et al., 1984; Berry et al., 1985; Pott et al., 1987, 1992; Judde et al., 1987; Sunderman et al., 1990).
On the whole, comparisons between human and animal data suggest that certain oxidic nickel compounds at high concentrations may increase respiratory cancer risks and that these risks are not necessarily confined to nickel-copper oxides. However, there is no single unifying physical characteristic that differentiates oxidic nickel compounds with respect to biological reactivity or carcinogenic potential. Some general physical characteristics which may be related to carcinogenicity include: particle size =5 µm, a relatively large particle surface area, presence of metallic or other impurities and/or amount of Ni (III). Phagocytosis appears to be a necessary, but not sufficient condition for carcinogenesis. Solubility in biological fluids will also affect how much nickel ion is delivered to target sites (i.e., cell nucleus) (Oller et al., 1997). The ability of particles to generate oxygen radicals may also contribute to their carcinogenic potential (Kawanishi et al., 2001).
With respect to non-malignant respiratory effects, oxidic nickel compounds do not appear to be respiratory sensitizers. Based upon numerous epidemiological studies of nickel-producing workers, nickel alloy workers, and stainless steel workers, there is little indication that exposure to oxidic nickel results in excess mortality from chronic respiratory disease (Polednak, 1981; Cox et al., 1981; Enterline and Marsh, 1982; Roberts et al., 1989b; Simonato et al., 1991; Moulin et al., 1993, 2000; Arena et al., 1998). In the few instances where excess risks of non-malignant respiratory disease did appear-for example, in refining workers in Wales-the excesses were seen only in workers with high nickel exposures (>10 mg Ni/m3), in areas that were reported to be very dusty. With the elimination of these dusty conditions, the risk that existed in these areas seems largely to have disappeared by the 1930s (Peto et al., 1984). In a study using radiographs of nickel sinter plant workers exposed to very high levels of oxidic and sulfidic nickel compounds (up to 100 mg Ni/ m3), no evidence that oxidic or sulfidic nickel dusts caused a significant fibrotic response in workers was reported (Muir et al., 1993). In a recent study of Norwegian nickel refinery workers, an increased risk of pulmonary fibrosis was found in workers with cumulative exposure to sulfidic and soluble, but not oxidic nickel (Berge and Skyberg, 2001). The previously mentioned Kilburn et al. (1990) and Sobaszek et al. (2000) studies (see Section 5.1.1) showed mixed evidence of chronic effects on pulmonary function in stainless steel welders. Broder et al. (1989) showed no differences in pulmonary function of nickel smelter workers versus controls in workers examined for short periods of time (1 week); however, there were some indicators of a healthy worker effect in this cohort which may have resulted in the negative findings. Anosmia (loss of smell) has been reported in nickel-cadmium battery workers, but most researchers attribute this to cadmium toxicity (Sunderman, 2001).
Animal studies have shown various effects on the lung following relatively short periods of exposure to high levels of nickel oxide aerosols (Bingham et al., 1972; Murthy et al., 1983; Dunnick et al., 1988; Benson et al., 1989; Dunnick et al., 1989). Effects have included increases in lung weights, increases in alveolar macrophages, fibrosis, and enzymatic changes in alveolar macrophages and lavage fluid. Studies of repeated inhalation exposures to nickel oxide (ranging from two to six months) have shown that exposure to nickel oxide may impair particle lung clearance (Benson et al., 1995; Oberdörster et al., 1995). Chronic exposures to a high-temperature nickel oxide resulted in statistically significant inflammatory changes in lungs of rats and mice at 0.5 mg Ni/m3 and 1.0 mg Ni/m3, respectively (NTP, 1996c). These values correspond to workplace exposures above 5-10 mg Ni/ m3. At present, the significance of impaired clearance seen in nickel oxide-exposed rats and its relationship to carcinogenicity is unclear (Oller et al., 1997).