Page 11: Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Selenium
Part II. Science and Technical Considerations (continued)
Selenium has been classified by the International Agency for Research on Cancer in Group 3: not classifiable as to its carcinogenicity to humans (IARC, 1975). The vast majority of epidemiological and animal studies do not demonstrate an increase in cancer incidence following a wide range of chronic exposures to selenium via food or supplements; a protective effect has even been suggested (Harr et al., 1967; Klein et al., 2003; Longtin, 2003; WHO, 2011; Ferguson et al., 2012).
Selenium is an essential nutrient and a component of several proteins and enzymes in the body that are known to play important roles, including regulation of thyroid hormones and antioxidant defences (Institute of Medicine, 2000; Otten et al., 2006). A deficiency in selenium may lead to chronic diseases such as Keshan disease (characterized by cardiomyopathy) and Kashin-Beck disease (characterized by rheumatism) (Yang, 1984) and may also be associated with a form of cretinism related to hypothyroidism (Spallholz, 2001; WHO and FAO, 2004; Xia et al., 2005). In order to protect the Canadian population from the aforementioned diseases, Health Canada adopted the RDA for selenium established by the Institute of Medicine (2000). These recommended daily intakes vary between 15 and 55 µg of selenium per day, depending on the age group. Selenium deficiency is not likely to be a concern in Canada, as the estimates from the Canadian TDS (2005-2011) show that the Canadian population meets the Institute of Medicine's recommended daily intake from food, which represents the main source of selenium.
Selenium toxicity generally occurs when exposure levels are much higher than the recommended daily intakes. Selenosis symptoms resulting from chronic exposure to high levels of selenium are characterized by hair loss, nail anomalies or loss, skin anomalies, garlic odour of the breath, tooth decay and, more severely, disturbances of the nervous system.
The studies of Yang and colleagues focus on the symptoms of selenosis in a Chinese population (Enshi County) exposed to high levels of selenium (Yang et al., 1989a,b) and the follow-up of five recovered sensitive individuals from the same population (Yang and Zhou, 1994). The source of selenium was mainly food (plant based) (Yang et al., 1983). The level of intake was classified as low, medium or high and estimated through questionnaire distribution and measurements of selenium in food items. Selenosis symptoms were classified according to their severity (Yang et al., 1989b). Symptoms were not present in individuals with a blood selenium concentration of 1000 µg/L or below. Blood selenium levels in the range of 1000-2000 µg/L induced symptoms in up to 35% of individuals, whereas blood selenium concentrations in the range of 2000-3300 µg/L or higher induced symptoms in 45% of individuals. Symptoms were mainly (97% of the time) present in adults. Persistent selenosis symptoms were observed in five Chinese individuals with blood selenium concentrations ranging between 1054 and 1854 µg/L. The authors calculated that a blood selenium concentration of 1054 µg/L corresponded to an intake of 910 µg/day and identified these as the minimum blood selenium concentration and selenium intake causing toxicity, respectively. The authors also indicated that the maximum daily safe intake of selenium was 750-850 µg/day.
After their diet was improved, those same five patients participated in a follow-up study (Yang and Zhou, 1994). By 1992, their symptoms had disappeared, and their blood selenium concentrations had dropped from an average of 1346 µg/L to 968 µg/L, the latter representing a mean intake of 819 µg/day, which was identified as the NOAEL by the authors. After taking into account interindividual variations, the authors identified a safe maximum daily intake of 400 µg/day (corresponding to a blood selenium concentration of 0.559 mg/L).
Although the key studies of Yang and colleagues (Yang et al., 1989a,b; Yang and Zhou, 1994) have some limitations, they provide qualitative and quantitative information useful for the dose-response assessment and risk characterization of selenium associated with high levels of intake. The absence of symptoms in other populations with high dietary levels of selenium exposure is supportive of the use of the NOAEL identified by Yang and colleagues and the Institute of Medicine (Otten et al., 2006) as a basis for the risk assessment. Findings from the studies by Longnecker et al. (1991) and Lemire et al. (2012), also performed in areas with high levels of selenium, confirm the findings from the studies by Yang and colleagues, as selenosis symptoms were not observed at exposure levels up to 724 µg/day and blood selenium concentrations up to 1500 µg/L, respectively. The UL derived by the Institute of Medicine is based on the adult subpopulation, as selenosis is a chronic health effect and no selenosis symptoms were observed in children in the Chinese, Venezuelan or Amazonian studies (Institute of Medicine, 2000). The Institute of Medicine (2000) stated that there is no known seleniferous area in Canada or the United States with recognized cases of selenosis.
Some epidemiological studies and clinical trials have reported associations between high selenium exposure and potential adverse health effects. The results of the NPC trial (Stranges et al., 2007) and the SELECT trial (Klein, 2009; Lippman et al., 2009) suggest a potential association between selenium intake and diabetes risk in a selenium-replete population, such as the one in the United States. A few epidemiological studies have also found an association between selenium exposure and other diseases, such as ALS (Vinceti et al., 2010) and glaucoma (Bruhn et al., 2009). Although these endpoints are important and might be linked to selenium intake, the generalizability of their results to the Canadian population is questionable and limited. These studies would also need to control for many biases and confounding factors involved in the development of these diseases (Health Canada, 2012b; Thayer et al., 2012). Trials designed to measure the effects of selenium exposure on specific diseases have to be conducted before conclusions can be drawn.
The UL, which is the highest level of nutrient intake that is likely to pose no risk of adverse health effects for almost all individuals in the general population, was calculated by the Institute of Medicine (IOM, 2000) as follows:
Equation 1 - Text Description
Health Canada (2010b) has adopted IOM's UL of 400 µg of selenium per day. According to IOM (2000), there is no evidence indicating an increased sensitivity to selenium toxicity for any age group.
Using this UL, the health-based value (HBV) for selenium in drinking water is derived as follows:
Equation 2 - Text Description
The U.S. EPA's Integrated Risk Information System has classified selenium as not classifiable as to its carcinogenicity in humans (class D) based on inadequate human data and inadequate evidence of carcinogenicity in animals (U.S. EPA, 1991a). As an exception for selenium compounds, selenium sulphide and selenium disulphide are classified by the U.S. EPA as class B2: probable human carcinogens based on sufficient animal data and inadequate human data, based on a comprehensive gavage study in mice and rats (NCI and NTP, 1980; U.S. EPA, 1991b). As these compounds are not soluble in water (ATSDR, 2003) and as the primary routes of exposure to these compounds are dermal and inhalation, they are not relevant to the risk assessment for selenium in drinking water.
Other organizations have set guidelines or regulations pertaining to the concentration of selenium in drinking water based on either the Chinese (Yang and Zhou, 1994) or the American (Longnecker et al., 1991) population studies or directly on the Institute of Medicine's UL of 0.4 mg/day (which is based on the Chinese population studies).
The current U.S. EPA maximum contaminant level (MCL) for selenium is 50 µg/L, which, for this particular compound, equals the maximum contaminant level goal (MCLG), "because analytical methods or treatment technology do not pose any limitation" (U.S. EPA, 2012). The MCLG (U.S. EPA, 1991a) was based on the Chinese data from Yang et al. (1989a,b). As part of its 6-year review, the U.S. EPA (2012) determined that the "MCL and MCLG for selenium are still protective of human health."
The California Office of Environmental Health Hazard Assessment (OEHHA, 2010) established a public health goal (PHG) of 30 µg/L for selenium in drinking water. The calculation of this PHG considered the NOAEL (0.015 mg/kg bw per day) for toxic non-cancer effects (hair loss and nail damage) observed in Chinese population studies. California's current standard (MCL) for selenium is 50 µg/L, which was adopted by the California Department of Health Services in 1994. It is based on the 1991 U.S. EPA rule.
The World Health Organization (WHO, 2011) established a provisional drinking water guideline value of 40 µg/L based on the Institute of Medicine's UL of 0.4 mg/day, an allocation factor of 20% and a consumption of 2 L of drinking water per day. The provisional designation was based on uncertainties inherent in the scientific database. It was noted that a drinking-water guideline for selenium would be unnecessary for most Member States and that achieving a proper balance between recommended intakes and undesirable intakes was essential to consider in establishing the guideline value.
The Australian drinking water guideline for selenium is 10 µg/L (NHMRC and NRMMC, 2011) based on the absence of effects associated with an average selenium intake of 0.24 mg/day by individuals living in a high-selenium region of South Dakota and eastern Wyoming over a 2-year period (Longnecker et al., 1991). The guideline value was based on a 10% allocation factor and assuming a 70 kg adult drinking 2 L of water per day.
- Date modified: