Cancer mortality in a Chinese population exposed to hexavalent chromium in drinking water.

BACKGROUND: In 1987, investigators in Liaoning Province, China, reported that mortality rates for all cancer, stomach cancer, and lung cancer in 1970-1978 were higher in villages with hexavalent chromium (Cr+6)-contaminated drinking water than in the general population. The investigators reported rates, but did not report statistical measures of association or precision. METHODS: Using reports and other communications from investigators at the local Jinzhou Health and Anti-Epidemic Station, we obtained data on Cr+6 contamination of groundwater and cancer mortality in 9 study regions near a ferrochromium factory. We estimated: (1) person-years at risk in the study regions, based on census and population growth rate data, (2) mortality counts, based on estimated person-years at risk and previously reported mortality rates, and (3) rate ratios and 95% confidence intervals. RESULTS: The all-cancer mortality rate in the combined 5 study regions with Cr+6-contaminated water was negligibly elevated in comparison with the rate in the 4 combined study regions without contaminated water (rate ratio = 1.13; 95% confidence interval = 0.86-1.46), but was somewhat more elevated in comparison with the whole province (1.23; 0.97-1.53). Stomach cancer mortality in the regions with contaminated water was more substantially elevated in comparison with the regions without contaminated water (1.82; 1.11-2.91) and the whole province (1.69; 1.12-2.44). Lung cancer mortality was slightly elevated in comparison with the unexposed study regions (1.15; 0.62-2.07), and more strongly elevated in comparison with the whole province (1.78; 1.03-2.87). Mortality from other cancers combined was not elevated in comparison with either the unexposed study regions (0.86; 0.53-1.36) or the whole province (0.92; 0.58-1.38). CONCLUSIONS: While these data are limited, they are consistent with increased stomach cancer risk in a population exposed to Crz=6 in drinking water.

Review of the evidence regarding the carcinogenicity of hexavalent chromium in drinking water.

Recent analyses have revealed that 38% of municipal sources of drinking water in California have detectable levels of hexavalent chromium. This observation provided new impetus to characterize the carcinogenic risk associated with oral exposure to hexavalent chromium in drinking water. Notwithstanding the well-characterized increases in cancer associated with inhalation exposure to this chemical, the marked reduction of hexavalent chromium to trivalent chromium in the stomach suggests that exposure to hexavalent chromium in drinking water may not pose a carcinogenic risk. A reevaluation of studies that investigated the toxicokinetics, the genotoxicity, and the mechanism of carcinogenicity of hexavalent chromium, as well as the available human and animal cancer studies, was undertaken to determine if there is evidence that exposure to this chemical in drinking water may pose a carcinogenic risk. Mechanistic studies suggest the potential for a carcinogenic response if hexavalent chromium enters cells. Both toxicokinetic and genotoxicity studies indicate that a portion of an orally administered dose of hexavalent chromium is absorbed and gets into cells of several tissues, causing DNA damage. The only lifetime oral study of hexavalent chromium in animals conducted thus far yielded a statistically significant increase in stomach tumors compared to controls. Also, in a limited-term cancer study, co-exposure to hexavalent chromium in drinking water and ultraviolet light produced skin tumors in mice. The only available cancer study of humans exposed to hexavalent chromium in drinking water revealed a statistically significant increase in stomach tumors. Moreover, a meta-analysis of occupational studies also revealed a statistically significant increase in stomach cancers. The increases in stomach tumors in both human and animal studies, along with the toxicokinetic, genotoxic, and mechanistic data, suggest that oral exposure to this agent appears to pose a carcinogenic risk.