Page 6: Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Selenium
Part II. Science and Technical Considerations (continued)
Canadians can be exposed to selenium through its presence in air, food, consumer products, soil and drinking water. The main source of selenium exposure is through diet. Selenium is an essential element and has minimum daily dose requirements set by international organizations (Institute of Medicine, 2000; WHO and FAO, 2004; Otten et al., 2006).
Although some exposure data are available, they are not sufficient to modify the default proportion (20%) of the daily intake allocated to drinking water (allocation factor) in the calculation of the maximum acceptable concentration.
Selenium levels in drinking water are generally low, but can vary depending on geological formations (CCME, 2009). The selenium concentrations in drinking water were measured in various locations across Canada as part of the National Survey of Disinfection By-Products and Selected Emerging Contaminants in Canadian Drinking Water (Health Canada, 2012a). Source water, treatment facilities and various points in distribution systems were sampled and analysed for dissolved and total selenium (after acid digestion). All samples (n = 65) collected in 2009/2010 were below the detection limit of 2 µg/L. However, it should be noted that any selenium leached from plumbing materials would not have been captured in this survey.
Provincial and territorial data were obtained from members of the Federal-Provincial-Territorial Committee on Drinking Water in 2012, and show the levels of selenium found in drinking water systems across Canada.
In Newfoundland and Labrador, data were provided for source and tap water since 2002 (Newfoundland and Labrador Department of Environment and Conservation, 2012). In tap water, 8246 out of 8689 (95%) samples were below the detection limit of 1 µg/L; 416 (5%) samples had selenium concentrations between 1 and 5 µg/L; and 27 (0.3%) samples had selenium concentrations between 5 and 11 µg/L. Similar measurements were shown for source water.
In Nova Scotia, selenium was above the detection limit of 2 μg/L in only 19 out of 1055 samples (1.8%) of raw or treated drinking water sourced from surface water, groundwater or distribution system collected between 2000 and 2012 (Nova Scotia Department of the Environment, 2012). The average selenium concentration was 2.1 µg/L, and the maximum concentration was 12 µg/L. Five of the 19 samples were from surface water, and the selenium concentrations in these samples were in the range 1-2 µg/L.
In New Brunswick, the majority of raw surface water and groundwater samples were below the detection limit of 2 µg/L for selenium in measurements done between 1994 and 2012 in Crown lands and between 2008 and 2012 in municipal lands (New Brunswick Department of Health and Wellness, 2012). There is no drinking water treatment system in the province (municipal or Crown) designed specifically to treat for selenium. In the Crown and municipality water samples, 127 out of 5159 (2.5%) had selenium concentrations above 2 µg/L, with a mean concentration of 2.5 µg/L and a maximum concentration of 10.4 µg/L.
In Quebec, 3698 drinking water distribution installations measured 14 083 samples of water for selenium from 2005 to 2009. Three percent (424/14 083) of the measurements were above the detection limit (0.1 µg/L), and 0.05% (7/14 083) were above 10 µg/L, with the highest value being 27 µg/L (Ministère du Développement durable, de l'Environnement et des Parcs du Québec, 2012). Of these, the majority were samples from groundwater supply without filtration.
In Ontario, 38 out of 3427 samples (1.1%) collected between 2007 and 2011 contained selenium at concentrations above 2 µg/L (Ontario Ministry of the Environment, 2012). Selenium was detected in all 37 samples of raw and treated groundwater collected in Walkerton, at concentrations ranging between 11 and 16 µg/L.
Selenium has been measured at levels of 0.011-0.043 µg/L in six locations of the west arm of Lake Erie (Adams and Johnson, 1977).
In First Nations of Manitoba, the selenium content of tap water was analysed in 8-23 households in each of nine communities from the six ecozones of the province (Chan et al., 2012). Selenium was present at concentrations below the detection limit of 0.2 µg/L in four ecozones, and the maximum concentration detected in all ecozones was 3.5 µg/L. Most of the water supply of the communities comes from surface water: four from lakes, three from rivers and two from groundwater.
In Saskatchewan, the vast majority of selenium concentrations measured in groundwater, surface water and treated drinking water samples collected between 2001 and 2011 were below the detection limit of about 1 µg/L (Saskatchewan Department of Environment and Resource Management, 2012). In general, Saskatchewan was found to have higher levels of selenium than other provinces and territories. Concentrations above 2 µg/L were detected in 7.2 % (217 out of 2997) of the samples measured in groundwater, surface water and treated drinking water and above 10 µg/L in 1.3% of the samples.
In west-central Saskatchewan and east-central Alberta, selenium was below the detection limit of 0.1 µg/L in six samples taken from the Battle River (Anderson, 1994).
In British Columbia, the Ministry of Environment reported that selenium was monitored in various rivers, and concentrations ranged from 2 to 9 µg/L. The concentration of selenium in water in a coal mining area was 2.5 µg/L (Nagpal and Howell, 2001).
In First Nations of British Columbia, the selenium content of tap water was analysed in 21 communities distributed among the eight ecozones of the province. Selenium was present at concentrations below the detection limit of 0.2 µg/L in three ecozones, and the maximum concentration detected in all ecozones was 1.4 µg/L. Source water supply varied between communities: 14 come from wells, 6 from creeks/streams, 2 from a river and 4 from lakes (Chan et al., 2011).
The Northwest Territories tested for selenium in raw or treated drinking water in 24 communities in 2009. Selenium was below the detection limit of 0.2-1 µg/L in 21 communities and detected in 3 communities at 0.2-0.9 µg/L (Government of the Northwest Territories, 2011).
Historically, leaded brasses used in potable water systems have been found to leach lead. Alternative non-leaded brass alloys have been developed--i.e., brass with no lead intentionally added and the lead replaced with other metals, such as bismuth, selenium and phosphorus. These metals improve the mechanical characteristics of the brass. However, there are limited data on the potential of non-leaded brass to leach metals. The Water Research Foundation has funded two projects on leaching of non-leaded brasses, which are currently in progress.
Brass is generally used in applications for distribution system and premise plumbing components (i.e., fittings), including brass components used with plastic piping. NSF International (NSF)/American National Standards Institute (ANSI) Standard 61 (NSF/ANSI, 2011a) is a health-based leaching standard which limits the leaching of selenium into drinking water to 0.005 mg/L. Under the National Plumbing Code of Canada, fittings must meet the requirements of plumbing standards for plumbing supply fittings, components and plastic pipes (NRCC, 2010). The Canadian Standards Association (CSA) standard for plastic pipes and the harmonized American Society for Mechanical Engineers/CSA standard for plumbing supply fittings as well as plumbing fittings (CSA, 2011a,b) require that components used for drinking water applications comply with the requirements of NSF/ANSI Standard 61. Materials in contact with drinking water meeting NSF/ANSI Standard 61 would be expected to leach very low amounts of selenium into drinking water.
Food is the main source of exposure to selenium. The major selenium species present in food are organic. Organic selenium is contained in the amino acid derivatives selenomethionine and selenocysteine, which are highly bioavailable (IPCS, 2006; Norton and Hoffmann, 2011). Although present in lower quantities, inorganic selenite and selenate can also be found in vegetables and mushrooms (Whanger, 2002; Thiry et al., 2012).
Concentrations of selenium in food items vary between countries and regions and depend on the food item and soil conditions. Concentrations can even vary within a single plant (Valdiglesias et al., 2009; Lemire et al., 2010). For example, the selenium content of Brazil nuts was found to range from 0.03 to 512 µg/g in the high-selenium area covered by the Tapajos River basin in Brazil. In Canada, selenium concentrations found in common food items vary; entrails and internal organs, beef meats, fish, eggs and Brazil nuts have a content generally between 30 and 310 µg/kg, vegetables generally between 1 and 100 µg/kg, and whole cow milk and cereals around 10-1350 µg/kg (Valdiglesias et al., 2009). Baked goods and bread are the main sources of selenium in food in Canada and contribute to 51% of the intake (Dabeka, 1994).
The Canadian Total Diet Study (TDS) is a Health Canada initiative that measures the dietary intakes of different chemicals for different age-sex groups of the Canadian population (Health Canada, 2011). For adults, the average dietary intakes of selenium were estimated at 1.9 µg/kg body weight (bw) per day in 2005 (Toronto) and 2006 (Halifax) and 2.7 µg/kg bw per day in 2007 (Vancouver). This study attested that food is the main source of selenium intake and found that dietary exposure to selenium ranged from 113 to 220 µg/day in Canadian adults, based on four different diets in Winnipeg, Halifax and Toronto (see Table 1 in Section 5.8 below for a summary). These estimates are within the range of those provided in a report published in 1975, indicating that selenium exposure from food has not changed substantially since then (Thompson et al., 1975).
According to the data from the TDS, the dietary intakes of selenium were 4.5 µg/kg bw per day in 2005 (Toronto), 4.4 µg/kg bw per day in 2006 (Halifax) and 7.7 µg/kg bw per day in 2007 (Vancouver) for infants 0-6 months of age. Selenium concentrations in powdered infant formula ranged from 25.1 to 49 ng/g in Canada, based on TDS data from 2005 (Toronto), 2006 (Halifax) and 2007 (Vancouver) (Health Canada, 2011). In a study conducted in the Ottawa area, infant formulas (prepared with demineralized, deionized water) were found to contain selenium at concentrations of 3-21 µg/L (unsupplemented formula) and 16-35 µg/L (supplemented formula) (L'Abbé et al., 1996). Assuming an average consumption rate of infant formula of 0.75 L/day by infants 0-6 months of age that are exclusively formula fed (Health Canada, 1994), this provides 2.2-15.7 and 12.0-26.2 µg of selenium per day for the unsupplemented and supplemented formulas, respectively (Health Canada, 1994). Breast milk samples from women in eastern Ontario contained 13-25 µg of selenium per litre, providing infants with an estimated exposure of 11-20 µg/day (L'Abbé et al., 1996).
The mean concentrations of selenium in particulate matter less than or equal to 10 µm in diameter in the air of 22 Canadian cities in 2009 were between 2 and 5 ng/m3 (Environment Canada, 2012). Based on these low concentrations of selenium in ambient air, intake from air would be negligible compared with that from other media.
Selenium supplements in the form of natural health products are available from organic and inorganic sources at doses between 3.5 and 400 µg/day in Canada (Health Canada, 2007). Sources of selenium in nutritional supplements include selenium salts, such as selenium citrate, sodium selenate and sodium selenite, and selenium chelates from hydrolysed vegetable and animal protein, which are also known as selenium proteinates. The selenium monograph of Health Canada's Natural Health Products Directorate (Health Canada, 2007) suggests a safe product dose of up to 400 µg daily for selenite, selenate and organoselenium, based on the Institute of Medicine's recommendations (Otten et al., 2006).
Selenium sulphide is used in pharmaceuticals and cosmetics. For example, antidandruff shampoo containing selenium sulphide is classified as a natural health product in Canada. Absorption is unlikely from this cosmetic usage, as there is no substantial dermal penetration of this form of selenium through intact skin (IARC, 1975).
The presence of selenium in soil varies widely and is a reflection of the mineralogy of the parent material (Whanger, 1989). In Earth's crust, concentrations of selenium are relatively low in general (0.05-0.09 mg/kg), and the world mean concentration is 0.4 mg/kg (range of 0.01-2 mg/kg) (Fordyce et al., 2000; Johnson et al., 2010). In seleniferous areas, concentrations of 1200 mg/kg have been measured (Fordyce et al., 2000). Sedimentary rocks tend to have higher concentrations, but concentrations rarely exceed 0.1-0.3 mg/kg (Johnson et al., 2010). Also, very high concentrations of selenium have been observed in phosphate rocks (> 300 mg/kg) and coal and black shales (20 to > 600 mg/kg).
The content of selenium in the soil was analysed for five Canadian regions (Appalachian, Canadian Shield, St. Lawrence Lowlands, Interior Prairies and Cordilleran). For all regions combined, the selenium concentrations ranged from 0.02 to 3.7 µg/g of soil, and a mean concentration of 0.30 µg/g of soil was reported (CCME, 2009).
Biomarkers of exposure have been used extensively in epidemiological studies on selenium (Mayne, 2003). Selenium levels in blood compartments (plasma, serum, erythrocytes) and urine are the most commonly used biomarkers and represent recent exposure (Rajpathak et al., 2005). In addition, selenium levels in blood are also good biomarkers of long-term exposure to dietary selenium (Longnecker et al., 1991). In contrast, selenium concentrations in nails and hair are measurements of long-term exposure (Rajpathak et al., 2005).
Statistics Canada, Health Canada and the Public Health Agency of Canada launched Cycle 1 of the cross-sectional Canadian Health Measures Survey to collect health data and biological specimens in approximately 5600 Canadians aged 6-79 years distributed among five age groups (6-11, 12-19, 20-39, 40-59 and 60-79 years) at 15 sites between 2007 and 2009. The geometric mean whole blood selenium concentration was 201 µg/L (95% confidence interval [CI] = 197-206 µg/L), and the geometric mean urinary selenium concentration was 49 µg/L (95% CI = 45-53 µg/L; n = 5492), for the total Canadian population aged 6-79 years (Health Canada, 2010a). Whole blood selenium concentrations ranged from the 10th percentile of 169 µg/L to the 95th percentile of 253 µg/L. No data were provided for children under 6 years of age.
The estimated total daily intakes of selenium from drinking water, air, soil and food for the 0- to 6-month, 7-month to 4-year and 20+-year age groups in the Canadian population are shown in Table 1. Daily selenium intakes from dietary supplements and other consumer products were not estimated, as there are no available data on the proportion of the general population using these products. Individual variability of selenium intakes is possible for each source.
|Age group||Daily intake of selenium from various sources (µg/kg bw per day)|
|Drinking waterTable 1 Footnote 1||AirTable 1 Footnote 2||SoilTable 1 Footnote 3||FoodTable 1 Footnote 4||Total|
|0-6 months non-breastfed infants||0.21||0.000 98||0.001 5||5.6||5.8|
|0-6 months breastfed infants||0||0.000 98||0.001 5||1.9||1.9|
|7 months to 4 years||0.12||0.001 3||0.001 1||5.6||5.7|
|20+ years||0.043||0.001 1||0.000 086||2.2||2.2|
The Institute of Medicine (2000) derived tolerable upper intake level (ULs) for selenium of 45 µg/day for infants aged 0-6 months, 60 µg/day for infants aged 7-12 months, 90 µg/day for children 1-3 years of age, 150 µg/day for children 4-8 years of age and 280 µg/day for children 9-13 years of age. The UL for infants aged 0-6 months was based on a human milk selenium concentration (n = 241 U.S. women from 17 states) of 60 µg/L that was without adverse effects in a study done by Shearer and Hadjimarkos (1975). The UL for adults of 400 µg/day was derived based on the studies of Yang and colleagues (Yang et al., 1989a,b; Yang and Zhou, 1994). The infant UL and the adult UL are similar on a body weight basis. Also, there is no evidence indicating increased sensitivity to selenium toxicity for any age group. Thus, the UL of 7 µg/kg bw/day was adjusted for older infants, children and adolescents on the basis of relative body weight.
- Date modified: