Transparency data

Discussion paper on the effects of excess Selenium on maternal health

Published 4 September 2026

This is a paper for discussion. This does not represent the views of the Committee and should not be cited.

Introduction

1. The Scientific Advisory Committee on Nutrition (SACN) last considered maternal diet and nutrition in relation to offspring health in its reports on ‘The influence of maternal, foetal and child nutrition on the development of chronic disease in later life’ (SACN, 2011) and on ‘Feeding in the first year of life’ (SACN, 2018). In the latter report, the impact of breastfeeding on maternal health was also considered.

2. In 2019, SACN agreed to conduct a risk assessment on nutrition and maternal health focusing on maternal outcomes during pregnancy, childbirth and up to 24 months after delivery; this would include the effects of chemical contaminants and excess nutrients in the diet. The assessment would also consider infant outcomes, but only where they relate to the neonatal period. The neonatal period was normally considered to be up to 28 days after birth, but the relevant time window for neonatal effects would be determined on a case-by-case basis.

3. SACN agreed that, where appropriate, other expert Committees would be consulted and asked to complete relevant risk assessments e.g., in the area of food safety advice. This subject was initially discussed by the COT during the horizon scanning item at the January 2020 meeting with a scoping paper being presented to the Committee in July 2020. This included background information on a provisional list of chemicals proposed by SACN, which was subject to change following discussion by COT who would be guiding the toxicological risk assessment process: candidate chemicals or chemical classes can be added or removed as the COT considered appropriate. The list was brought back to the COT with additional information in September 2020 where it was agreed that papers on a number of components, including selenium, should be prioritised for review.

4. Following discussion of the first prioritisation paper on substances to be considered for risk assessment, the Committee decided that selenium should be considered in a separate paper. Whilst SACN previously evaluated the implications for health of dietary intakes of selenium in the UK population in 2013, the following statement discusses the risks posed specifically to maternal health by selenium in the diet and the environment.

Background

5. Selenium is an essential trace element found in nature, belonging to the chalcogen group (Perrone, Monteiro, and Nunes, 2015), that is required for the synthesis of selenoproteins, including enzymes involved in antioxidant defence, thyroid hormone metabolism and maintenance of cellular redox homeostasis (EVM, 2003). This element can have both metallic and non-metallic properties (EVM, 2003; Perrone et al, 2015). Selenium exists in several oxidation states, however it is most commonly reduced to the −2 (selenide, Se−2) oxidation state, or oxidised to the +4 (selenite, SeO32−) or +6 (selenate, SeO42−) oxidation states (Perrone et al, 2015).

6. Selenium is found in a variety of foods, particularly nuts, offal, eggs, poultry, and mushrooms. Fruits and vegetables generally contain lower levels, except for Brassica vegetables (such as cabbage and cauliflower), which can contain comparatively higher amounts of selenium (COT, 2019). Brazil nuts contain 800 to 8300 µg 100 g−1 selenium making them the richest source of selenium in the diet (Perrone et al, 2015). Selenium content in soil directly impacts levels of selenium in plants, along with soil pH and redox potential (Perrone et al, 2015).

7. Selenium may also be present in enriched and fortified foods. Enriched foods are produced using selenium-enriched fertilisers, whereas fortified foods contain added selenium compounds. Selenium can also be consumed as supplements. Authorised selenium sources for use in UK food supplements include sodium selenate, sodium hydrogen selenite, sodium selenite and selenium-enriched yeast, with the latter being subject to specific legal specifications (Directive 2002/46/EC; EFSA NDA Panel, 2023 (hereafter referred to as EFSA, 2023)). Doses of selenium within supplements typically range from about 2.5 to 400 μg (EFSA, 2023; NIH ODS, 2025).

8. There are currently no UK-specific recommendations for selenium supplementation during pregnancy, and NHS advice does not suggest selenium supplementation for pregnant women. Most pregnant women are considered able to achieve adequate selenium intakes through a varied diet, with supplementation generally not recommended due to the relatively narrow margin between adequate and excessive intake (Hubalewska-Dydejczyk et al, 2020; Picciano and McGuire, 2018)

9. EFSA (2014) established an Adequate Intake (AI) of 70 µg/day for adult women, which also applies during pregnancy. This was based on the levelling off of Selenoprotein P (SelP) at intakes ranging from 60 to 100 µg/day in in population groups from Finland, the UK, New Zealand and the USA. For lactating women an additional 15 µg/day (based on an average amount of selenium secreted in breast milk of 12 μg/day and a 70% absorption efficiency from usual diets) was estimated, resulting in an AI of 85 µg/day (EFSA, 2014; 2023). However, selenium intakes across Europe are generally low. The suboptimal selenium status observed in the UK has been partly attributed to the low selenium content of UK soils, which results in lower selenium concentrations in produced foods, particularly wheat (Broadley et al., 2006). In line with this, dietary surveys have reported that a substantial proportion of the UK population have selenium intakes below recommended levels (Stoffaneller and Morse, 2015; Rayman, 2000; The Food Foundation, 2026).

10. Selenium’s bioavailability, function, toxicity to humans, and adverse health effects depend on its chemical form, as organic or inorganic compounds (Perrone et al, 2015). In the diet, organic compounds of selenium are more prevalent. In plant-based foods, the amino acid analogue L‑selenomethionine (SeMet) predominates, and in animal-derived foods, L‑selenocysteine (SeCys) is the principal form of selenium present (COT, 2019; EFSA, 2023). Inorganic compounds such as selenate and selenite are present in lower amounts in food (COT, 2019).

11. SeCys is the biologically active selenium containing amino acid, which is incorporated into selenoproteins, of which over thirty have been identified (EVM, 2003). Selenoproteins, namely glutathione peroxidases (GPx), iodothyronine deiodinases, SelP, and thioredoxin reductases, carry out a range of essential metabolic and antioxidant functions. GPx’s protect the cell from oxidative stress by degrading toxic H2O2, and in placental tissue, may additionally support placental differentiation (Kyriakou et al., 2026). Iodothyronine deiodinases are responsible for the production of thyroid hormones, converting thyroxine (T4) into the biologically active triiodothyronine (T3) (EVM, 2003; Labunskyy, Hatfield and Gladyshev, 2014). SelP functions as the major selenium transport protein in plasma (EFSA , 2023).

12. Selenium toxicity depends on the form of the compound, with soluble forms like selenite, selenate and SeMet having higher potential toxicity. Acute toxicity can cause hypersalivation, vomiting, garlic‑like breath (due to the excretion of volatile selenium metabolites), severe gastrointestinal symptoms, hair loss, neurological disturbances and fatigue (SCF,2000; EVM,2003; CRN, 2025). Chronic toxicity, known as chronic selenosis, occurs following chronic ingestion of excess selenium from the diet and/or other sources, such as dietary supplements (CRN, 2025). Chronic selenosis is characterised by changes in hair (alopecia; hair loss, brittleness) and nails (dry, thickened, brittle, discoloured), garlic‑like breath, skin lesions and neurological effects that can progress to numbness, convulsions and paralysis (EVM, 2003; CRN, 2025; IOM, 2000). Studies from seleniferous regions in the USA and China suggest selenosis occurs at intakes above 910 μg /day (15 μg/kg bw for a 60‑kg adult) (EVM, 2003; CRN, 2025).

Previous Evaluations

13. The following section summarises the Health Based Guidance Values (HBGV’s) for selenium set by scientific committees and regulatory bodies. Please note that further details on the toxicity studies used to set HBGV’s are detailed in the toxicity section of this paper.

Scientific Committee on Food (SCF), Opinion of the Scientific Committee on Food on the Tolerable Upper Intake Level of Selenium (2000)

14. In 2000, the SCF derived a Tolerable Upper Intake Level (UL) for selenium from all sources of food and supplements, at 300 µg/day for adults  ≥ 18 years, including pregnant and lactating women. The SCF concluded that the UL also applies to pregnant and lactating women, as there was no evidence of increased susceptibility in these groups, no adverse effects in infants of mothers with high selenium intakes, and no adverse effects reported in lactating women consuming dietary selenium below the adult UL.

15. The UL of 300 µg/day for adults  ≥ 18 years was derived using a No Observed Adverse Effect Level (NOAEL) of 850 µg/day for clinical selenosis (i.e., hair or nail loss, nail abnormalities, mottled teeth, skin lesions and changes in peripheral nerves) in a study of 349 subjects (Yang et al., 1989b) and applying an uncertainty factor of 3 to account for uncertainties among studies used in deriving the UL. The NOAEL of 850 µg/day was derived based on the absence of clinical signs in individuals with blood selenium levels below 1000 μg/L.

16. The SCF stated that the study by Yang and Zhou (1994) supported this NOAEL and UL, finding that symptoms of selenosis disappeared after 5 subjects from the Yang et al 1989 study reduced their dietary intake to a mean of 819 µg Se/day.

Institute of Medicine (IOM), Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium and Carotenoids (2000)

17. The Food and Nutrition Board of the IOM, in cooperation with scientists from Canada, published a report in 2000 in which they establish updated Dietary Reference Intakes (DRIs) for selenium, vitamin C and vitamin E for the United States and Canada. As part of the work, the Board also established ULs for these three compounds.  

18. The IOM based their UL for selenium on the critical endpoints of hair and nail brittleness and loss, stating that these signs and symptoms were the most frequently reported signs and symptoms of chronic selenosis.

19. Similarly to the SCF (2000), the IOM identified Yang and Zhou (1994) as a useful data set in determining the dose-response of selenium toxicity from food sources. The IOM agreed that the mean NOAEL proposed by Yang and Zhou of 800 µg Se/day was reasonable (rounded down from 819 µg Se/day).

20. Using the NOAEL of 800 µg Se/day and an uncertainty factor of 2 to protect sensitive individuals, the IOM derived an UL of 400 µg Se/day from all sources of food and supplements for adults  ≥ 19 years, including pregnant and lactating women. The IOM stated that the UL for pregnant and lactating women was the same as the general population as they were unable to identify any reports of selenosis or teratogenicity in infants resulting from high,  but not toxic, intakes of selenium in mothers.

Expert Group on Vitamins and Minerals (EVM), Safe Upper Levels for Vitamins and Minerals (2003)

21. The EVM in 2003 established a Safe Upper Level (SUL) of 450 µg total Se/day for adults aged ≥ 18 years (assuming a maximum intake of 100 µg/day from food, and up to 350 µg/day selenium for supplementation). This was based on the key studies of Yang et al (1989a and b) from which the EVM used the Lowest Observed Adverse Effect Level (LOAEL) of 910 µg Se/day, and an uncertainty factor of 2 for LOAEL to NOAEL extrapolation, to calculate the SUL. No vulnerable groups were identified. It is not directly stated whether this SUL applies to pregnant and lactating women.

22. In their 1989b study, Yang et al correlated high selenium intakes with blood levels and determined that the LOAEL of 910 µg Se/day was the point at which marginal selenium toxicity occurred. Similarly to both the SCF and IOM, changes in the nails and hair were identified by the EVM as the most sensitive indicators of selenium toxicity.

23. The EVM highlighted discrepancies in the NOAELs described in the Yang et al studies. In their 1983 study, Yang et al investigated the average daily intake of selenium in areas of China where intakes were classified as either deficient, adequate, or high (with or without endemic selenosis). Results showed a range of selenium intakes from 240 to 1,510 µg Se/day, average 750 µg Se/day, where there was no selenosis. However, in a later study by Yang et al 1989b, the LOAEL of 910 µg Se/day was determined. The EVM proposed that the way dietary intakes were calculated across the two studies may have caused this difference, and it was also considered that the higher intakes in the 1983 study could be due to sensitisation to selenium in some of the population because of earlier outbreaks of selenosis.

24. Additionally, the SUL is described by the EVM as being consistent with the findings of both Clark et al (1996) and Longnecker et al (1991) who found that reported intakes of 0.2 mg (200 µg) supplemental Se/day and 0.24 mg (240 µg) total Se/day, respectively, were not associated with adverse effects.

World Health Organization (WHO) and Food and Agriculture Organization of the United Nations (FAO), Vitamin and Mineral Requirements in Human Nutrition (2004)

25. The WHO/FAO convened an expert consultation in 2004 to re-evaluate the role of micronutrients in human health and nutrition. The main goals were to review vitamin and mineral requirements, recommend nutrient intakes that could be used by Member States,  identify data gaps, and make recommendations for future research. This document follows on from the 1987 joint publication from the WHO, United Nations Environment Programme and the International Labour Organisation, a comprehensive review of the chronic and acute toxicity of selenium from food, drinking water, and the environment.

26. In 2004, the WHO/FAO suggested a provisional UL (from all sources) of 400 µg Se/day for adults ≥18 years, aligning with that set by the IOM for adults ≥ 19 years. However, the WHO/FAO note that an UL for pregnant and lactating women had yet to be determined.

27. The provisional UL of 400 µg Se/day for adults ≥ 18 years was not determined from a specific LOAEL or NOAEL, however the authors describe the conclusions from Yang et al (1983) where foods grown in seleniferous soils in China, that supplied more than 900 µg Se/day, were associated with an increased incidence of nail dystrophy. The WHO/FAO do not explicitly state the study used for determining their point of departure.

28. In addition, the WHO/FAO noted from Levander (1987) that human overexposure to selenium was difficult to  identify due to ill-defined signs and symptoms, and a lack of sensitive biochemical markers available for detecting impending selenium intoxication. It is unclear what uncertainty factor was applied to determine this provisional UL; however it is described as providing a fully adequate margin of safety.

National Health and Medical Research Council of Australia and New Zealand (NHMRC), Nutrient Reference Values for Australia and New Zealand (2006)

29. In 2006, the National Health and Medical Research Council (NHMRC) of Australia and New Zealand set an UL for adults ≥ 19 years, including pregnant and lactating women, of 400 µg Se/day, aligning with the IOM and WHO/FAO. This UL related to total selenium intake from the diet and supplements and was based on a NOAEL of 800 µg Se/day. The NHMRC derived this NOAEL from the previously described Chinese studies (Yang et al (1983, 1989b; Yang and Zhou.,1994) which they describe as being consistent with the US study by Longnecker et al (1991). The endpoints identified from these studies were brittleness and loss of hair and nails, gastrointestinal disturbance, skin rash, fatigue, irritability and nervous system abnormalities. It was concluded by the NHMRC that there was no evidence suggesting pregnant and lactating women had an increased susceptibility to toxicity.

30. The NHMRC applied an uncertainty factor of 2 to the NOAEL of 800 µg Se/day to protect sensitive populations due to data gaps, highlighting that although the toxic effect of selenium is not severe, the toxicity may be irreversible.

31. As part of their rationale for setting the UL, the authors considered The Nutritional Prevention of Cancer Trial (Duffield-Lillico et al, 2003) which showed at 200 µg supplemental Se/day, individuals with a high risk of non-melanoma skin cancer had an increased risk of squamous cell carcinoma and total non-melanoma skin cancer. It was concluded that there was uncertainty around how this risk would be applied to the general population.

32. In 2026, a public consultation was launched by the NHMRC requesting feedback on updated Nutrient Reference Values (NRVs) for selenium. This update is the result of emerging scientific evidence for selenium toxicity since the original evaluation in 2006 and includes new recommendations for the UL for selenium.

33. In their 2026 draft recommendations, the UL is lowered from 400 µg Se/day to 330 µg Se/day for adults ≥ 18 years, including pregnant and lactating women, based on a different rationale to the initial 2006 assessment. The NHMRC used a weight of evidence approach, utilising the 2023 European Food Safety Authority (EFSA) recommendations along with published primary studies, Australian and New Zealand population data, and advice published by key international bodies. The proposed UL of 330 µg Se/day was based on a study from Lippman et al (2009) where alopecia was selected as the critical effect. This study is described in more detail when discussing EFSA’s 2023 assessment below and in the toxicity section of this assessment.

Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT), Overarching Statement on the Potential Risks from Contaminants in the Diet of Infants Aged 0 to 12 Months and Children Aged 1 to 5 Years (2019)

34. The Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) last reviewed selenium in 2019, as part of a review of the risk of toxicity of chemicals in the diets of infants and young children aged 0 to 5 years, in support of a review by the SACN of Government recommendations on complementary and young child feeding.

35. The COT noted the existing UL’s from both the SCF (2000), and the EVM (2003). The SCF extrapolated the ULs from adults to children based on reference bodyweights, proposing UL values for children and adolescents of 60, 90, 130, 200 and 250 μg/day for children aged 1-3, 4-6, 7-10, 11-14 and 15-17 years respectively. The COT used these extrapolated values as the Health Based Guidance Values (HBGVs) in their risk assessment.

36. In their review, the COT concluded that ‘estimated dietary exposures for children aged 0 to < 12 months and 1 to < 5 years were below the UL, either from breastmilk or other foods and are therefore unlikely to be of toxicological concern’.

European Food Safety Authority (EFSA) Panel on Nutrition, Novel Foods, and Food Allergens (NDA Panel), Scientific Opinion on the Tolerable Upper Intake Level for Selenium (2023)

37. The European Commision (EC) requested that the EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA) re-evaluate the ULs for selenium for all populations, following on from an opinion adopted in 2019 ‘Safety of selenium-enriched biomass of Yarrowia lipolytica as a novel food pursuant to Regulation (EU) 2015/2283’. In this opinion, it was identified that in recent years, large population-based randomised controlled trials had emerged along with other sources of new data, thus warranting a reassessment of the UL that was established in 2000 by the SCF.

38. The Panel reviewed the previous NOAEL of 850 μg Se/day (Yang et al., 1989b), as used by the SCF, stating that available studies have evidenced selenium toxicity at exposures below this level, therefore the NOAEL should be revisited. However, they concluded that studies published after Yang et al (1989a) were not suitable for deriving a NOAEL.

39. EFSA therefore set an UL of 255 μg Se/day for adults ≥18 years, including pregnant and lactating women. This UL covers selenium intake from all dietary sources, including forms authorised for addition to food and use in food supplements (i.e. sodium selenate, sodium hydrogen selenite, sodium selenite, L-selenomethionine and selenium-enriched yeast).

 40. The UL was derived from a LOAEL of 330 μg Se/day (130 μg/day from background diet plus supplemental intake of 200 μg/day) from the SELECT trial (Lippman et al, 2009). The critical endpoint selected was alopecia, an early observable feature of selenosis and a well-established adverse effect resulting from altered protein metabolism.

41. EFSA noted that the outcomes of this study showed an increased risk of developing alopecia at selenium intakes of around 330 μg/day (baseline plus supplemental intake), compared with mean baseline serum selenium concentrations of 130 μg/day.

42. An uncertainty factor of 1.3 was applied to the LOAEL of 330 μg Se/day, to account for the mild and likely reversibility of alopecia, and two uncertainties: the use of a LOAEL as a reference point, and a lack of data in women. However, despite the lack of sex specific data for women, EFSA note the literature suggests there are no sex differences in susceptibility to selenium toxicity, and therefore EFSA consider that this UL also applies to pregnant and lactating women.

Table 1. Basis for Tolerable Upper Intake Levels (ULs) and Safe Upper Levels (SULs) for Selenium in Adults, Established by International Regulatory Bodies.

Regulatory Body (Year)	Upper Level(µg/day)	Point of Departure (µg/day)	Uncertainty Factor	Key study/ Studies (Year)	Population / Age Group
SCF (2000)	300	NOAEL = 850	3	Yang et al (1989b)	Adults ≥ 18 years (a)
IOM (2000)	400	NOAEL = 800	2	Yang and Zhou (1994)	Adults ≥19 years (a)
EVM (2003)	450 (SUL)	LOAEL = 910	2	Yang et al (1989a,b)	Adults ≥ 18 years
WHO/FAO (2004)	400	Not Specified	Not Specified	Not Explicitly Stated	Adults ≥ 18 years
NHMRC (2006)	400	NOAEL = 800	2	Yang et al (1983,b), Yang and Zhou (1994), Longnecker et al (1991)	Adults ≥19 years (a)
EFSA (2023)	255	LOAEL = 330	1.3	Lippman et al (2009)	Adults ≥ 18 years (a)

Including pregnant and lactating women.

Data on Effects on Maternal Health

43. EFSA conducted a comprehensive reassessment of selenium in 2023, as described in the Previous Evaluations section of this report. As the most recent assessment, the 2023 EFSA opinion forms the basis of this current assessment. EFSA also reviewed selenium in 2014, however this has not been described in detail due to the availability of the updated 2023 assessment, although information from the 2014 opinion has been included where relevant. In addition, a literature search was carried out for articles published from 1st January 2022 to 1st January 2026 to identify recent publications linking selenium to adverse effects in the maternal diet. Further details on the terms used for this search can be found in Annex A.

ADME 

Absorption

44. In the diet, organic SeMet and SeCys are the most abundant forms of selenium. Inorganic selenium compounds, selenite and selenate, represent a smaller proportion of overall dietary intake (EFSA, 2023; EFSA 2014).

45. SeMet is synthesised by plants, but not animals. SeMet can be incorporated into the non-specific protein pool by both plants and animals, where it may non-specifically replace methionine in proteins forming ‘selenium-containing’ proteins (EFSA, 2023; Combs, 2015).

46. The form of selenium used for selenoprotein synthesis is SeCys. SeCys is cotranslationally synthesised from selenide (formed from the reduction of selenite), and the amino acid serine, and is incorporated into peptides in selenoprotein mRNA at the site of UGA (stop) codons (EFSA, 2023; Labunskyy et al, 2014). Selenoproteins can only be synthesised by animals, some microorganisms and algae, but not plants (EFSA, 2023).

47. All forms of selenium are readily absorbed, as confirmed by human studies (EFSA, 2023; Jäger et al., 2016a and b). Selenium absorption can be influenced by other dietary constituents and has been reported to range from 50-90% in humans (EFSA, 2023; Combs, 2015).

48. Animal studies showed selenium containing amino acids are absorbed through active transport processes following digestion of the respective proteins (EFSA, 2023; Combs, 2015). SeMet is absorbed by intestinal methionine transporters after which it enters the methionine pool within the body (EFSA, 2023). The absorption of SeCys is less well characterised than that of SeMet, but it may be absorbed in a similar way to cysteine (EFSA, 2023; Ha et al., 2019).

49. Absorption of inorganic forms of selenium do not require digestion. Both are efficiently absorbed, by either passive diffusion (selenite) or carrier-mediated diffusion (selenate) (EFSA, 2023). During absorption, selenite is quickly reduced to selenide by thioredoxin reductase or glutathione in the intestinal mucosal cells (EFSA, 2023). Selenate must be reduced to selenite before further metabolism can take place (EFSA, 2023).

Distribution

50. After absorption, selenium forms are transported via the bloodstream , predominantly to the liver, where they are metabolised (Ha et al. 2019). In addition to the liver, metabolites are distributed throughout the body to multiple organs and tissues including the nervous system, lungs, kidneys, pancreas, skin, hair, bone, testes, and skeletal and cardiac muscle (ATSDR, 2013). Excess selenium can therefore result in organ specific pathological and biochemical effects (EFSA, 2023).

51. SelP and GPx3 are the most abundant selenoproteins in the plasma. SelP typically carries 30-60% of circulating selenium, and GPx3 10-30% (EFSA NDA Panel, 2014; hereafter referred to as EFSA, 2014). The remaining plasma selenium is mainly present as SeMet in albumin and other proteins, and a minor amount (<3%) present as small molecular compounds such as selenosugars (EFSA, 2014). The proportions of these selenium forms depends on both the amount and the chemical form of selenium in the diet. Selenium is also present inside platelets and red blood cells as glutathione peroxidase GPx1 (EFSA, 2014).

52. From the liver, selenium is transported typically as SeCys via SelP, but also in other selenoproteins (EFSA, 2023). It is not distributed evenly, with animal studies showing that some organs and tissues require more selenium than others (EFSA, 2023; Ha et al, 2019).  Observations from in vitro studies showed that for a cell to utilise SelP, it is incorporated into the cell then degraded in the lysosome, enabling released SeCys to become available for selenoprotein synthesis (EFSA ,2023).

53. Rodent studies have evidenced a selenoprotein hierarchy, with prioritised incorporation of selenium into particular selenoproteins, and the prioritisation of selenium supply to organs such as the brain, endocrine and reproductive organs (EFSA, 2014).

54. Geographic location leads to significant variation in selenium concentrations in human tissues, with low soil selenium concentrations resulting in lower selenium concentrations in subjects (EFSA, 2014).

55. Regarding pregnancy, EFSA (2014) summarised evidence showing that selenium accumulates in the organs of the growing foetus during pregnancy, with hepatic concentrations of selenium in the foetus remaining constant (details on concentrations were not described by EFSA). Following birth, hepatic selenium levels fall, suggesting the liver stores selenium during pregnancy and redistributes selenium stores after birth to other organs.

56. EFSA (2014) describe how a study in mice demonstrated a selenoprotein dependent maternal-foetal selenium transfer mechanism: apolipoprotein E receptor-2 (apoER2) mediated uptake of maternal SelP from maternal blood, as well as an additional mechanism that remained unidentified from the study (EFSA, 2014; Burk et al, 2013). Maternal transfer of SeMet or SeMet-containing proteins also provide selenium to the foetus (EFSA, 2014; Burk et al, 2013).

Metabolism

57. The liver is the key site of metabolism of selenium (EFSA, 2023). Animal and in vitro studies have evidenced the fate of selenium in the liver, and its transport throughout the body to other cells for metabolism (EFSA, 2023). All selenium compounds except SeMet, MeSeCys and γ-Glu-MeSeCys, are metabolised to hydrogen selenide in the liver (EFSA, 2023).

58. Following absorption, SeCys, selenate and selenite are metabolised through pathways that support selenoprotein synthesis. This process involves free SeCys degradation by SeCys lyase producing hydrogen selenide and alanine within the cell. SeCys has not been found freely in tissues as it is highly reactive, with SeCys lyase maintaining low intracellular concentrations through its conversion to hydrogen selenide (EFSA, 2023). Hydrogen selenide is then converted into selenophosphate and attached to selenocysteine-tRNA, enabling insertion of SeCys into selenoproteins during translation (EFSA, 2014). For selenate or selenite to be utilised for selenoprotein synthesis, they must be reduced to hydrogen selenide by a series of redox reactions utilising reduced glutathione and thioredoxin reductases to reduce selenate to selenite, and finally to hydrogen selenide (EFSA, 2014, EFSA, 2023).

59. EFSA (2023) describe a study on the human metabolic profile of urinary selenium by Lajin et al (2016), which demonstrates that once hydrogen selenide is generated it can be incorporated into selenoproteins, used to form selenosugars, or methylated for excretion.

60. Dietary SeMet can be metabolised to SeCys via the transsulfuration pathway however there is no physiological pathway for the synthesis of SeMet from SeCys in humans, making this an irreversible conversion (EFSA, 2023). Selenium deposited in the non-specific protein pool (as SeMet) can be released by degradation of proteins, where it can be used to form selenoproteins thereafter as SeCys (EFSA, 2023). SeMet may also be metabolised by γ-lyase to methylselenol for excretion as evidenced in mouse liver, although it’s unclear if this happens in humans (EFSA, 2023).

Excretion

61. Urinary excretion is the primary elimination route for selenium, with balance studies showing at selenium intakes between 10 and 80 µg/day, around 40-60% is excreted (EFSA, 2014; 2023). Selenium can also be excreted in the faeces from selenoproteins in intestinal mucosal cells and when simply unabsorbed (EFSA, 2014).

62. The rate of excretion is highest within 24 hours of ingestion, however the fraction of selenium excreted in urine depends on the chemical nature of selenium ingested (EFSA, 2014; Burk et al, 2006). In a randomised, placebo-controlled intervention study, Burk et al. (2006) supplemented 88 selenium replete adults with 200-600 μg/day selenium as SeMet, sodium selenite or selenium-enriched yeast for 16 weeks. Urinary excretion differed between selenium forms, with average excretion of 60 ± 26% for SeMet, 41 ± 15% for sodium selenite and 52 ± 23% for selenium-enriched yeast (EFSA, 2014; Burk et al., 2006). For groups receiving SeMet and sodium selenite, the difference reached significance (EFSA, 2014).

63. In humans, urinary selenium metabolites include Se-methyl-N-acetylgalactosamine, a methylated selenosugar, which is a major selenium compound excreted in urine after typical dietary selenium intake (EFSA, 2014). The main urinary selenium metabolites in humans are selenosugars 1 and 3, trimethylselenonium (TMSe), and selenate, while selenium methylselenoneine may also be excreted following the methylation of selenoneine derived from fish consumption (EFSA, 2014; EFSA, 2023). In the urine, there are also a number of unidentified selenium metabolites which are influenced by the selenium compound ingested, the dose, and an individual’s genetics (EFSA, 2023).

64. TMSe is produced through the sequential methylation of hydrogen selenide. Hydrogen selenide is first methylated to methylselenol and then to dimethylselenide (DMSe), primarily in the liver. It has been suggested that DMSe is subsequently converted to TMSe in the lungs. When selenium intake is high and the capacity to convert DMSe to TMSe is exceeded the excess DMSe is exhaled, resulting in the characteristic garlic-like odour on the breath (EFSA, 2023).

65. Some individuals may be unable to excrete TMSe due to a genetic polymorphism in the human indolethylamine N-methyltransferase (INMT) gene, they are known as TMSe non-eliminators (EFSA, 2023). Although more evidence is needed, it is suggested by EFSA (2023) that these individuals may be more susceptible to selenium toxicity due to the reduced ability to excrete excess selenium in urine. Selenosugars 1 and 3 are predominant in TMSe non-elimator urine, whilst TMSe is more prominent in TMSe eliminators (EFSA, 2023). Selenosugar 2 Is usually only detected when selenium intake is high (EFSA, 2023).

66. Selenate has been shown to be excreted mostly unmetabolized in humans, at a single dose of 50 μg (EFSA,2023; Jäger et al., 2016a).

67. Selenium is also excreted via breast milk, reflecting maternal selenium intake from mostly organic, but sometimes inorganic sources (EFSA, 2014). EFSA (2014) state, from a study by Dorea (2002), that selenium concentration is highest in colostrum, and decreases as lactation progresses, however high variability in maternal plasma/serum selenium to milk selenium has been noted therefore excretion via breast milk may vary on an individual basis.

Biomarkers

68. Selenium can be measured in a range of biological matrices, including serum, plasma, erythrocytes, urine, hair, placental tissue and amniotic fluid. Key biomarkers used in many studies determining selenium status or effects on selenium status include plasma selenium, GPx and SelP.

69. Plasma selenium concentration can be a useful biomarker of selenium intake and exposure, but not functional selenium status as it reflects both selenoproteins (from the functional pool of selenium) and other plasma proteins where SeMet non-specifically substitutes for methionine (EFSA, 2014; Burk et al, 2001; Hurst et al, 2010). Plasma selenium responds to supplementation across a wide range of intakes (20-700 μg/day) and is therefore a useful marker of recent selenium exposure, particularly in studies using SeMet or selenium-enriched yeast supplements (Ashton et al, 2009; Burk et al, 2006). However, plasma selenium concentrations can be influenced by the chemical form of selenium consumed (Fairweather-Tait et al., 2010), inflammation (Nichol et al., 1998; Maehira et al., 2002; Huang et al., 2012), and, to a lesser extent, age, sex and smoking status (Robberecht and Deelstra, 1994) meaning it’s use for determining selenium status requires interpreting carefully (EFSA, 2014). Plasma selenium does not appear to plateau with increasing intake although concentrations of around 70-100 μg/L have been proposed to reflect ‘selenium adequacy’ (Combs, 2001; EFSA, 2014).

70. Additionally, SelP is widely used as a biomarker of selenium status along with GPx activity as a functional indicator of selenium sufficiency, although GPx reaches a plateau at higher selenium intakes (EFSA, 2014; Labunskyy et al, 2014; Kyriakou, 2026) .

71. Glutathione peroxidase (GPx) activity is a biomarker of selenium function and reaches saturation at relatively low selenium intakes. Plasma GPx3 activity generally plateaus at selenium intakes of around 40-60 μg/day (EFSA,2014; Yang et al., 1987; Duffield et al., 1999; Xia et al., 2005; Xia et al., 2010), while maximum platelet and whole-blood GPx activity has been associated with plasma selenium concentrations in supplementation studies of approximately 95-115 μg/L (Alfthan et al., 1991;EFSA, 2014). As GPx becomes saturated before other functional selenium pools (e.g. SelP), EFSA (2014) considered it less informative than SelP for defining selenium requirements.

72. EFSA (2014) identified SelP as the most informative biomarker of selenium status because of its role in selenium transport and homeostasis, suggesting it more accurately reflects selenium status of the whole organism. SelP represents a saturable pool of selenium , with maximum concentrations generally associated with plasma selenium concentrations of approximately 90-140 μg/L in supplementation studies (EFSA, 2014; Hurst et al., 2013). EFSA (2014) considered the levelling-off of SelP to indicate adequate selenium supply to tissues and saturation of the functional selenium pool, making it the preferred biomarker for deriving selenium dietary reference values.

73. EFSA (2014) noted that maternal selenium biomarkers such as plasma selenium concentrations and GPx activity decline during pregnancy. These changes may reflect physiological adaptations including an increase in plasma volume and maternal-foetal selenium transfer, and are therefore difficult to interpret in relation to selenium requirements in pregnancy.

Interaction With Other Nutrients

74. Selenium is likely to interact with other nutrients involved in the antioxidant-pro-oxidant balance of the cell due to its role in the antioxidant network (EFSA, 2014). Data are limited for interactions between selenium and ascorbic acid, however selenium and iodine play key roles in thyroid hormone metabolism and deficiencies of both, impacting thyroid function, have been observed in animals and humans (EFSA, 2014).

Mechanism(s) of Toxicity

75. The SCF (2000) and EVM (2003), note that molecular mechanisms of selenium toxicity are unclear. Many mechanisms have been reviewed (i.e. redox cycling of auto-oxidisable metabolites, glutathione depletion, inhibition of protein synthesis, depletion of S-adenosyl-methionine and the replacement of sulphur in sulphydryl groups of proteins and cofactors), however the mechanisms may vary depending on the selenium compound, and several may operate at once.

76. EFSA (2023) reviewed the available evidence and reported that the most common mode of action for selenium toxicity is likely oxidative stress, which interrupts cellular and mitochondrial functions. Hydrogen selenide is a key intermediate in selenium metabolism and is required for selenoprotein synthesis, however once this pathway is saturated, excess selenium metabolites may undergo oxidation, leading to the formation of reactive oxygen species (ROS) causing cellular and subcellular damage, and cell death.

77. Another suggested mechanism by EFSA (2023) involves inhibition of methylation, the primary detoxification pathway for selenium, which enables the accumulation of metabolites such as hydrogen selenide and other selenides, contributing to hepatic toxicity. Studies in mice suggest that excessive exposure to SeCys can impair methylation through inhibition of methionine adenosyltransferase (EFSA, 2023; Rayman et al., 2008a).

78. Other mechanisms that have been reported include the selective accumulation of selenium in growth hormone-producing cells of the pituitary gland, which has been proposed as a possible contributor to the growth reductions observed in experimental animals (SCF, 2000; EFSA, 2023). In addition, high selenite exposure has been shown in vitro to activate endoplasmic reticulum stress and increase ROS production, resulting in endothelial dysfunction (EFSA, 2023; Zachariah et al., 2021).

79. Finally,  EFSA (2023) noted that substitution of sulphur with selenium in proteins may disrupt their structure and function leading to toxic effects. This mechanism has been proposed to be the underlying mechanism for the characteristic features of selenosis such as hair loss and nail damage through alterations in keratin structure.

80. Overall, EFSA (2023) concluded that multiple modes of action are likely to operate in parallel, with their importance dependent on selenium dose and chemical form.

Toxicity

81. Study data regarding acute toxicity, carcinogenicity, genotoxicity or reproductive effects were not included in the 2023 EFSA opinion, with their focus being chronic toxicity, epidemiology studies and human data. A further search of the literature was conducted by the Secretariat for additional studies published between 1st January 2022 and 1st January 2026, linking selenium to toxic maternal effects, however no additional studies were identified. Therefore, previous evaluations describing the acute toxicity, carcinogenicity, genotoxicity or reproductive effects of selenium have been used to provide information for these sections e.g. EVM (2003), SCF (2000) etc.

Acute/Subacute Toxicity

82. Symptoms of acute selenium toxicity, or ‘acute selenosis’, in humans include hypersalivation, emesis, garlic aroma on the breath, vomiting, diarrhoea, hair loss, neurological disturbances, hypotension, tachycardia, pulmonary oedema, and fatigue (EVM, 2003; EFSA, 2023).

83. The SCF (2000) reported that symptoms of selenium toxicity have been observed following a single dose of 250 mg (250,000 µg) selenium. The SCF also cited reports of toxicity associated with the consumption of misformulated supplements containing 27 to 31 mg (27,000 to 31,000 µg) selenium per tablet, with symptoms developing after repeated daily use over a period ranging of 11 days to 2 months (SCF, 2000; Jensen, Clossen and Rothenberg, 1984). In comparison, the EVM (2003) stated that acute selenium poisoning appears to occur at doses >0.5 mg/kg bw (>500 µg/kg). In animals, selenium is described by the EVM (2003) as having moderate to high acute toxicity, with effects seen in the nervous system, liver and lungs.

Reproductive toxicity

84. The SCF (2000) and EVM (2003) reviewed animal data and concluded that selenium may cause adverse effects on reproductive parameters such as the oestrous cycle, while multigenerational studies showed that high levels of selenium exposure may lead to reduced weight of offspring and reduced survival after birth.

85. Adverse reproductive and developmental effects have been observed in rodents but have typically been associated with maternal poisoning and nutritional deprivation (SCF, 2000). In mice, increased pre-weaning mortality, runt offspring and impaired breeding was reportedly observed at 3 ppm selenium in drinking water (Schroeder and Mitchener, 1971; SCF, 2000), while in swine, reduced conception rates and reduced offspring survival was observed at 10 ppm selenium in feed (Wahlström and Olson, 1959; SCF, 2000). Selenomethionine administered at  >77 μmol/kg lead to foetal malformations and reduced foetal growth in hamsters (Ferm et al, 1990; SCF, 2000). Equivalent doses in µg/kg bw were not available for these studies.

86. The SCF and EVM also noted that teratogenic effects have been reported in several animal species including birds, fish, sheep, pigs and hamsters, but have not been reported in studies carried out in macaque monkeys despite evidence of maternal toxicity (Tarantal et al,1991). However, developmental effects from mammalian studies occurred mostly at levels that resulted in maternal toxicity therefore limiting conclusions regarding teratogenicity.

87. Tarantal et al. (1991) found no evidence of teratogenicity in macaque monkeys following prenatal exposure to L-selenomethionine ( 0, 25, 150, or 300 µg /kg of L-selenomethionine (Se) daily during organogenesis (gestational days 20-50). Dose related maternal toxicity was observed, with adverse effects including anorexia, vomiting and reduced body weight. One growth-retarded foetus was observed on gestational day 131 in a dam exposed to 25 µg/kg bw/day. At 300 µg/kg bw/day, one early embryonic death (gestational day 35) and two foetal deaths (gestational day 68 and gestational day 123) were reported, however, the authors state that pregnancy loss was not significantly different from concurrent or historical controls. Necropsy on gestational day 100 ± 2 found no statistically significant treatment related findings. A unilateral cortical cataract was identified in one infant exposed to 150 µg/kg bw/day, although this was considered likely to be a spontaneous occurrence.

88. The EVM (2003) review highlighted evidence suggesting that pre-weanling animals may exhibit greater sensitivity to selenium toxicity than adults. Individual study citations for reproductive studies were not provided in the EVM (2003) review.

89. Additionally, the EVM note a prominent data gap in that epidemiological studies are yet to assess reproductive toxicity.

Carcinogenicity and Genotoxicity

90. The EVM describe how selenium compounds used in food and supplements are not carcinogenic, with the exception of selenium sulphide which is carcinogenic in rats and mice. In vivo mutagenicity tests showed variable results, however compounds were mostly negative with an increase in chromosomal aberrations only seen at lethal doses of sodium selenite in hamster bone marrow (EVM, 2003).

Chronic Toxicity

91. Symptoms of chronic selenium toxicity in humans, or ‘chronic selenosis’, include changes to the nails (brittle, thickened, streaks or spots), hair (brittle, alopecia), skin lesions, neurological effects (fatigue, peripheral hypoaesthesia, hyperreflexia) discolouration and decay of the teeth and garlic odour on the breath (EVM, 2003; EFSA, 2023).

92. In animals, chronic exposure has resulted in reduced growth, weight gain, liver changes, anaemia, pancreatic enlargement, and in some domestic animals, neurotoxicity at selenium exposures above 0.03 to 0.4 mg/kg bw (30 to 400µg/kg bw) (SCF, 2000).

Epidemiology and Human Data

93. This  section describes the key human and epidemiological studies identified within previous evaluations that were considered when setting ULs for selenium. Where the same study has been described in more than one evaluation, it is described only once under the earliest evaluation in which it appears, to avoid duplication.

94. In their 2023 evaluation, EFSA state that current literature suggests that there are no obvious sex differences in selenium metabolism, therefore the data presented below, as taken from previous evaluations, includes studies in both male and female subjects.

95. Some full texts were unable to be retrieved (Yang et al, 1989a; Yang and Zhou, 1994) due to their age and a lack of digitisation of the journal (‘Journal of Trace Elements and Electrolytes in Health and Disease’) therefore studies have been described using the abstract only, or by using text from evaluations that have reviewed these papers.

Key studies identified by the SCF (2000)

Yang et al., 1983

96. Yang et al investigated endemic selenium intoxication in Enshi County, China, which was highly prevalent between 1961 and 1964. Selenium intake in the areas studied was classified as high (with or without endemic selenosis), adequate, or deficient. Morbidity was approximately 49% in 248 inhabitants living in heavily affected villages. Selenium toxicity was suspected because of selenium from coal (average selenium content > 300 µg/g; with one sample exceeding 80,000 µg/g) entering soil by weathering and being taken up by crops, and prolonged drought causing increased consumption of vegetables and maize in place of failed rice crops.

97. Symptoms  of selenosis included skin lesions, mottled teeth, nervous system abnormalities (observed in only one heavily affected village; 18 of 22 inhabitants), and characteristic signs of selenosis such as brittle hair and nails, hair loss, and spots and streaks on nail surfaces.

98. Samples of hair, blood, urine, soil, water, vegetables and cereals were analysed for their selenium concentration using distillation and titration methods (samples from 1966 only) and microfluorometric techniques. Selenium intake was estimated from information collected on dietary habits and by measuring the selenium content of cereal and vegetables

99. Symptoms of selenosis were seen at daily selenium intakes ranging from 3,200 to 6,690 µg, with an average of 4,990 µg. In high selenium areas without occurrence of selenosis, daily selenium intake was estimated to range from 240 to 1,510 µg , with an average of 750 µg. Residents suffering with selenosis recovered once diets were modified. Additionally, a case report was also described, in which a middle-aged hemiplegic female (~40 years) died reportedly due to selenosis, showing motor and sensory abnormalities, however a lack of autopsy and clinical history meant the cause of symptoms could not be determined.

Yang et al 1989a, 1989b

100. A cross sectional study of 349 individuals was conducted in the Enshi County in China, where individuals (male and female) aged 1-71 years, who lived in areas with either ‘low’, ‘medium’, or ‘high’ selenium levels in the soil, were evaluated for clinicial and biochemical signs of selenium toxicity (SCF, 2000; EFSA, 2023). Dose response relationships were determined from studying selenium intake via food and measuring selenium in tissues including whole blood, urine, hair, finger and toe nails (SCF, 2000).

101. The average daily intakes were estimated  at 70, 195 and 1,438 µg, and 62, 198, and 1,288 µg for adult male and females respectively, at low, medium and high selenium soil levels (average body weight: male 55 kg, female 53 kg) (SCF, 2000; EFSA, 2023).

102. Individuals were examined for clinical signs of selenosis (hair or nail loss, nail abnormalities, mottled teeth, skin lesions and changes in peripheral nerves). Where clinical signs of selenosis were present, subjects were classified as ++ or +. At blood selenium concentrations below 1,000 µg/L (~853 µg Se intake/day according to regression equation, figure 1 Yang et al. 1989a), no clinical signs of selenosis were observed. At 1,000 to 2,000 µg/L (grouped as 1,000 – 1,250, 1,250 – 1,500 and 1,500 – 2,000 µg/L), mild toxicity (selenosis +) was prevalent in 10-35% of subjects, increasing to 45% at blood concentrations above 2,000 µg/L. More severe toxicity (selenosis ++) varied between 3 and 7% within the same groups. It was concluded that no dose response relationship was seen (SCF, 2000).

103. Subjects with persistent symptoms had blood levels approximately 1,050 – 1,850 µg/L (913 -1,907 µg Se/day intake).  Additional effects observed include abnormal clotting (in 45% of subjects above 1,000 µg/L, and 1 subject <1,000 µg/L), a reduction in plasma selenium to red cell selenium ratio (at concentrations >900 µg/L, or 750 µg Se intake/day), and a decreased concentration of glutathione in blood (at concentrations >850 µg Se intake/day) (SCF, 2000).

104. The authors (Yang et al, 1989b) concluded from their studies that a daily selenium intake of 750 to 850 µg may be a safe level and proposed 400 µg/day as a maximum daily safe intake taking ‘variable factors’ into account. Additionally, Yang et al. (1989b) suggested that the manifestation and severity of selenosis are influenced not only by selenium exposure but also individual susceptibility, as individuals with the highest blood selenium concentrations did not necessarily exhibit the most pronounced signs of toxicity.

Yang and Zhou, 1994

105. Yang and Zhou conducted a follow up study on 5 adults (aged >30) showing signs of selenosis from the original cross-sectional study by Yang et al, 1989 in the Enshi County in China. Their aim was to reexamine blood selenium levels and clinical signs of selenosis to identify any correlation (Yang and Zhou, 1994; SCF, 2000; EFSA, 2023)

106. Symptoms of selenosis disappeared after subjects changed their diets, with average blood levels of selenium decreasing from 1346 ± 366  to 968 ± 115 µg/L. A regression equation was applied to convert these levels into a corresponding intake of 1270 ± 450 to 819 ± 126 µg Se/day, the latter of which was suggested as the mean NOAEL (rounded to 800 µg Se/day)(Yang and Zhou 1994; SCF,2000; EVM, 2003). A maximum safe dietary intake was set at 400 µg Se/day. This was determined from the lower limit of the 95% confidence interval of 600 µg Se/day however was lowered ‘for safety’ (Yang and Zhou,1994).

107. In their evaluation, the IOM described limitations of the study, such as  a very small number of subjects (n=5) and that Chinese populations may not be representative of the wider population (e.g. may have differences in sensitivity to selenium compared to other populations)(IOM, 2000).

Additional studies considered by the SCF (2000)

108. The SCF considered additional studies by Longnecker et al. (1991), Clark et al. (1996) and Brätter and Negretti de Bräatter (1996) to support the UL of 300 μg/day for adults, but were not used directly in its derivation. These studies were also summarised by EFSA in their 2023 opinion.

109. In a study of 142 adults living in seleniferous areas of South Dakota and Wyoming, Longnecker et al. (1991) reported selenium intakes ranging from 68 to 724 μg/day, with around half of participants exceeding 200 μg/day, and 12 exceeding 400 μg/day, assessed using a 48 h duplicate-plate food collection technique. No clinical signs of selenosis were observed, even at the highest intakes of approximately 720 μg/day, and although alanine aminotransferase activity positively correlated with intake, it remained within the reference range and was not considered biologically relevant. Similarly, in the Nutritional Prevention of Cancer trial, supplementation with 200 μg/day selenium yeast  (total dietary intake approximately 300 μg/day) was not associated with dermatological signs of toxicity (Clark et al., 1996). In a cross-sectional study of 125 lactating women in Venezuela with estimated selenium intakes ranging from 90 to 980 μg/day (estimated from concentrations of selenium in breast milk), Brätter and Negretti de Brätter (1996) also reported no evidence of nail or hair changes. Collectively, the SCF (2000) used these studies to support the absence of toxicity at relatively high intakes but were not used as the basis for setting the UL.

Key studies from EFSA (2023)

110. EFSA describe 5 randomised control trials that provide data on signs and symptoms of selenium toxicity in adults from the monitoring of adverse events. Of these trials, the Selenium and Vitamin E Cancer Prevention Trial (SELECT) by Lippman et al (2009) was chosen from which to set an UL for selenium as described in the ‘previous evaluations’ section of this paper.

Lippman et al, 2009

111. The SELECT trial evaluated the effect of 200 µg/day SeMet supplementation in 8,752 men in the United States, compared with 8,696 men who received a placebo. Participants were men without prior prostate cancer, aged ≥50 years (African American) or ≥55 years (others) and received supplementation for a median of approximately 5.5 years. Mean baseline serum selenium concentrations were 135 µg/L (selenium group) and 138 µg/L (placebo group), suggesting an estimated intake of 130 µg/day which is higher than typical European intake. After 4 years, levels increased to 252 µg/L in the selenium group whilst the placebo group saw only marginal increase to 140 µg/L (EFSA, 2023; Lippman et al, 2009).

112. The primary endpoint of the study was prostate cancer, however as part of this trial adverse effects known to be associated with the study supplements were recorded and compared between placebo and selenium supplement groups.

113. The side effects included in the final paper were the highest grading of a subjects symptom reported, with effects being reported every 6 months during a study site visit or substitute phone call. Adverse effects assessed included alopecia, dermatitis (grades 1-2 mild to moderate, and 3-4 severe to life threatening), halitosis, nail changes, fatigue (grades 1-2, and 3-4) and nausea (grades 1-2, and 3-4). The National Cancer Institute Common Toxicity Criteria (Version 2) were used for grading of all listed effects except halitosis and dermatitis which were graded based on the study protocol (EFSA, 2023; Lippman et al, 2009).

114. Compared with the placebo group, selenium supplementation was associated with a significantly increased risk of alopecia (RR 1.28; 99% CI1.01, 1.62). A small increase in risk of developing nail changes was observed (selenium group: RR 1.04; 99% CI 0.94, 1.16). Higher risks were also observed in the selenium group for dermatitis, halitosis, fatigue (grades 1-2), and nausea (grades 1-2). However, most estimates were not statistically significant: dermatitis (grades 1-2: RR 1.17; 99% CI 1.00, 1.35; grades 3-4: RR 1.74; 99% CI 0.56, 5.44), halitosis (RR 1.17; 99% CI 0.99, 1.38), fatigue (grades 1-2: RR 1.09; 99% CI 0.95, 1.26), and nausea (grades 1-2: RR 1.19; 99% CI 0.94, 1.52). Less than 5% of individuals were lost to follow up, >2 years before analysis (EFSA,2023; Lippman et al, 2009).

115. EFSA (2023) noted this study showed an increased risk of adverse effects from selenium toxicity, particularly alopecia, at total intakes around 330 μg/day, compared to 130 μg/day.

Additional studies considered by EFSA (2023)

116. As described in EFSA (2023), four additional RCTs (Algotar et al , 2013b; Winther et al , 2015; Thompson et al , 2016; Fairris et al , 1989) comprising between 69 and 1621 subjects generally did not demonstrate consistent or dose-related evidence of selenium toxicity across intakes ranging from 200–600 μg/day. In the Negative Biopsy Trial (Algotar et al , 2013b) no increase in selenosis endpoints (e.g. brittle hair or nails) were observed at selenium and Se yeast intakes up to 400 μg/day in adult men with increased risk of prostate cancer. Similar findings were reported in the Selenium and Celecoxib trial (Thompson et al. 2016) in the US, where no increased risk of hair or nail effects was identified at 200 μg/day Se yeast in participants with a history of colorectal adenoma. In the Danish Prevention of Cancer by Intervention with Selenium study (Winther et al., 2015), adverse effects such as hair loss and nail changes were reported to occur similarly in selenium (yeast; 100, 200 and 300 μg/day) and placebo groups and showed no dose-response relationship. Additionally, no signs of toxicity were reported in a small psoriasis trial (Fairris et al. 1989) where patients received Se yeast supplements of 600 μg/day (n =22), although, study results were not reported in the publication of this study.

117. EFSA determined that these studies provide limited inconsistent evidence for selenium toxicity, with limitations including small sample sizes, high dropout rates, and incomplete or poorly described adverse event monitoring. While one RCT (SELECT) indicated an increased risk of alopecia and related effects at intakes of approximately 330 μg /day, this finding was not consistently replicated across the other trials. EFSA considered the SELECT trial was the most robust study for setting the UL.

118. In their opinion, EFSA (2023) also considered three cross-sectional studies (Lemire et al, 2012; Chawla et al, 2020; Martens et al, 2015), comprising 41 to 680 participants, that examined selenium toxicity in seleniferous areas of Brazil, India and the United States, in addition to the study by Yang et al. (1989a). In these studies, plasma selenium levels ranged from 50 to 950 μg/L, with median values around 135 to 250 μg/L.

119. In the Brazilian Amazonas study (Lemire et al. 2012), mild nail changes were reported at plasma selenium concentrations ≥ 328 μg/L, although these effects were also observed at lower levels. In the Indian study (Chawla et al.2020), hair and nail abnormalities were reported at median serum selenium concentrations of approximately 250 μg/L compared with around 155 μg/L in unaffected individuals, with increased risks observed above the study population mean of approximately 171 μg/L however, there was considerable overlap in selenium status between affected and unaffected individuals. In children, selenium intakes of approximately 155 μg/day (exceeding age-specific ULs of 60 - 90 μg/day) were not associated with clinical signs of selenosis (Martens et al, 2015).

120. EFSA also describe a case report of a 55-year-old woman experiencing symptoms of selenosis after consumption of 10 to 15 nuts per day for 20 days resulting in a selenium serum concentration of 512 μg/L. Symptoms resolved after consumption of the nuts was ceased.

121. Overall, EFSA concluded that selenium toxicity was being observed at levels below the NOAEL determined by Yang et al 1989a, highlighting the wide variability in the susceptibility of individuals to selenium toxicity.

Additional Literature Search

122. A further search of the literature was conducted for articles published between 1st January 2022 and 1st January 2026, linking selenium to toxic maternal effects. Search terms for the literature search can be found in Annex A. No new studies were identified that directly linked selenium to toxicity in pregnant women or women of childbearing age. However, 6 epidemiological studies were identified that reported associations between selenium status and outcomes including reduced birth weight, increased risk of gestational diabetes mellitus (GDM), and reduced placental weight. These studies are based on observational data and associations, and do not directly provide clear evidence of adverse effects from high dietary or supplementary selenium intakes. Therefore, the studies are described below to capture the available human evidence for consideration by the Committee as part of the overall weight of evidence.

123. In addition, 3 recent studies in fish species reported developmental, neurological and reproductive effects from selenium exposure. These studies are described briefly as supporting evidence as they provide information on potential developmental and reproductive effects of excess selenium exposure. The study limitations and applicability to humans are outlined when described below.

Epidemiology and Human Data

Reduced placental weight

124. Kinjo et al (2024) examined the association between placental weight and heavy metal exposure using blood samples of 73,005 women who delivered a single live birth from 15 varying geographical areas of Japan from the Japan Environment and Children’s Study (JECS; Kawamoto et al., 2014). Women with data missing from key characteristic questions ( e.g. age, weeks of pregnancy, sec of infant) and with pregnancy conditions (e.g. foetal growth restriction, hypertensive disorder of pregnancy) were excluded. The study did not exclude pregnancies affected by a range of conditions, including foetal anomalies, second trimester bleeding, collagen disorders, renal, hepatic or cardiac disease, haematological conditions, and cancers.

125. A 2 mL subsample of maternal blood, taken from a 33 mL sample retrieved during the second or third trimester, was used to determine whole blood selenium concentration using an Agilent 7700 Inductively Coupled Plasma Mass Spectrometer (ICP-MS). The authors grouped participants into quartiles (Q1-4) based on selenium levels in whole blood, and used Z scores and multivariable logistic regression analysis to explore the association between blood selenium levels and placental weight.

126. The median selenium concentration detected in maternal blood was 168 ng/g (equivalent to 168 µg/kg), however whole blood concentrations ranged from 83 to 412 ng/g (equivalent to 83 to 412 µg/kg) across the 4 quartiles. The authors conducted multivariable regression analysis for the association between maternal whole blood concentrations of heavy metals and Z score, adjusted for neonate sex and placental weight by gestational age. Multivariable regression analysis showed a significant inverse association between selenium levels and placental weight (Z score; CoEf −0.31, standard error = 0.028, p < 0.001). However, a decreased odds ratio from 1 in Q1 (lowest selenium whole blood concentration; 83 to 155 ng/g, equivalent to 83 to 155 µg/kg) to 0.82 in Q4 (highest selenium whole blood concentration; 182 to 412 ng/g, equivalent to 182 to 412 µg/kg) demonstrated reduced odds of low placental weight with increasing selenium exposure. The authors concluded that selenium itself may not be responsible for decreasing placental weight, and that it may involve multiple metals acting on placental weight.

127. Similarly, Grundeken et al (2024) assessed the impact of essential elements on the placenta of 406 Swedish women (aged 20 to 45 years), using samples and data from the Swedish birth cohort “Nutritional impact on Immunological maturation during Childhood in relation to the Environment” which was undertaken between 2015 and 2018. Although the cohort consisted of 655 pregnancies, only 406 were included in the authors’ main analysis as these women had complete data on placental trace element concentrations, placental weight, birth weight, and relevant covariates. A sub sample of 391 women was used to estimate placental accumulation, by comparing placental concentrations with maternal blood erythrocyte fractions.

128. Placentas for analysis were collected at birth, and triangular cross-sectional samples were taken from the cord insert to outer edge of the placenta. One third was used for DNA extraction and the remainder for trace element analysis using ICPMS Agilent 7900. Venous blood samples (6 mL) were collected from women at gestational week 29 and separated into erythrocyte and plasma fractions, which were diluted in an alkali solution before use. Placental weight (g; weighed with membranes and umbilical cord) and birth weight (g) were measured after birth by midwives and recorded to the nearest gram. Placental efficiency was estimated using the fetoplacental weight ratio. Relative telomere length (TL) and mitochondrial DNA copy number (mtDNAcn) in placental tissue were measured by quantitative polymerase chain reaction (qPCR) in a sub-sample of 285. Associations between individual exposures and outcomes were assessed using linear or spline regression models, while joint effects and interactions were examined using Bayesian kernel machine regression (BKMR). All models were adjusted for sex, maternal smoking, and either maternal age or body mass index (BMI).

129. Median placental selenium concentration was 166 μg/kg.  A nonlinear association between selenium and placental weight was observed. Adjusted regression showed an inverse association between selenium levels above 147 μg/kg and placental weight (B: −158g per doubling; 95 % CI: −246, −71), and no association below this level. Other observations included a weak inverse association with gestational age at birth (rs: −0.11; p = 0.029), and birth weight (rs: −0.17; p = 0.001). Additionally, placental selenium slightly decreased with increasing maternal BMI (rs: −0.11; p =0.024).

130. In multivariable-adjusted linear regression models, selenium was not associated with TL or mtDNAcn in the placenta.

131. The authors identify that the intake of selenium in women studied was below recommended intakes (~40 μg/day), and that differences in selenium status between women were still associated with differences in placental outcomes.  The observed inverse association between higher placental selenium concentrations and placental weight was suggested to potentially indicate a change in selenium transport between the placenta and foetus in late gestation, however the underlying mechanism is uncertain.

Lower Birth Weight

132. Mehta et al (2023) investigated selenium biomarkers measured in maternal delivery and venous cord (VC) blood , including whole blood (WBSe), serum selenium, SelP, and their association birth weight and infant growth outcomes. Growth outcomes included age-adjusted length, weight, head circumference and weight-for-length z-scores at birth, 1 and 2 years of age. However, to align with the scope of this paper, only the results relating to selenium concentrations and birth weights are considered. Low birth weight was defined as birth weight < 2500 g. Data were obtained from a randomised controlled trial initially undertaken to assess the effects of supplementation with vitamin D3 on pregnant and lactating women in Bangladesh. The original trial included 1298 women with uncomplicated single pregnancies between 17 and 24 weeks of gestation, aged 18 years and above. The authors note that a sub-sample of 1021 was analysed for at least one selenium biomarker, however analytical sample sizes varied across different biomarkers and time points ranging from 435 to 780.

133. Secondary analysis was based on availability of selenium samples (one measurement of selenium and/or selenoproteins at delivery in mothers, or in venous cord blood of neonates) and measured as part of a panel of metals for the primary study. Venous cord blood was collected within 30 minutes of delivery. Whole blood selenium at delivery and venous cord blood were analysed using inductively coupled plasma mass spectrometry. Serum selenium was measured using inductively coupled plasma-dynamic reaction cell mass spectrometry at delivery. Plasma SelP was measured at delivery and in venous cord blood using a selenotest ELISA assay. Maximum selenium levels were reported, with whole blood selenium ranging from approximately 190 to 230 µg/L in delivery and VC blood, and 110 µg/L in serum samples. The authors state that selenium levels were therefore within normal ranges and did not indicate high exposure. Birth weight was measured using a digital scale. Results were analysed using regression analyses.

134. Delivery and venous cord whole blood selenium were both negatively associated with birth weight, with statistically significant associations (WBSe delivery: adjusted β −26.6; 95 % CI –44.3, −8.9; p = 0.003; WBSe VC: adjusted β −19.6, 95 % CI –33·0, −6.1; p = 0.005). Serum selenium showed no significant association in adjusted models at birth. In contrast, delivery SelP levels (adjusted β: −37.5, 95 % CI –73·0, −2.0; p = 0.04) and venous cord blood (adjusted β: 82.3, 95 % CI 30.0, 134.7; p = 0.002) showed inconsistent associations with birth weight,  with maternal concentrations negatively associated with birth weight, and venous cord blood concentrations showing a positive association. A small increase in risk of low birth weight at higher whole blood selenium concentrations was also identified.

135. Overall, findings were observed in a population with selenium concentrations within normal range, and the reported effects were significant but small. This suggests that adverse effects such as lower birth weights, are unlikely to be the result of high selenium intakes. The authors conclude that ‘mechanisms for the associations between birth weight and Se remain unclear’.

136. Álvarez-Silvares et al (2023) acknowledged that the effect of metals, such as selenium, on foetal growth should be investigated highlighting that investigations into the effects of metal exposure on human health have been conflicting, with effects on foetal growth not yet clear. The authors therefore investigated the association of placental metals, including selenium, and neonatal weight in a cohort of 79 low obstetric risk pregnant women aged 19 to 42 years from northwestern Spain.

137. Participants were recruited between October and December, 2017. Women were excluded if they were under 18 years of age, had a twin pregnancy, had a chronic disease diagnosed before pregnancy, experienced preterm labour (<37 weeks), were followed up exclusively at another centre, had antenatal care at the University Hospital of Ourense but gave birth outside the Ourense healthcare area, or did not consent to participate after reading the consent form.

138. Placental samples (excluding umbilical cord) were collected at delivery. Inductively coupled plasma mass spectrometry was used to determine the levels of trace metals in the placental tissue. Multivariate statistical analyses were performed using linear regressions, including generalised linear models, and generalised additive models (GAM) where linearity was not met. Models were adjusted for body mass index and maternal age at the start of pregnancy, previous birth history, gestational age, and maternal exposure to smoking. Significance was set at p < 0.05.

139. Placental selenium ranged from 0.744 to 1.202 µg/g dry weight, with a mean of 0.969µg/g dry weight. The authors reported a significant association between higher placental selenium concentrations and lower neonatal birthweight. However, the GAM suggest a weak negative association, with data points substantially varied and just within significance ( p = 0.049). This, along with the small sample size (n = 79) limits confidence in the strength of the association. The lack of data on additional placental parameters and dietary and/or supplemental selenium intake limits interpretation of these findings as placental selenium levels represent internal exposures only, and cannot be linked to external intakes. Therefore, consistent with Mehta et al (2023), no clear or plausible mechanism linking selenium exposure to reduced birth weight can be established.

Increased Risk of Gestational Diabetes Mellitus (GDM)

140. Li et al (2025) investigated associations between heavy metals, trace elements and gestational diabetes mellitus (GDM). The authors note that GDM has been associated with adverse pregnancy and birth outcomes, including preterm birth, foetal or neonatal growth abnormalities, spontaneous abortion and maternal morbidity, and aimed to determine whether exposure to specific metals and trace elements, including selenium, in Black and Hispanic communities were associated with the development of GDM.

141. The study included 1256 pregnant women (average age at delivery of 28 years) from the prospective Boston Birth Cohort, a predominantly Black (58%) and Hispanic (22%) population in the United States. Selenium was measured in venous blood samples, collected within 24 to 72 hours of delivery, and analysed using inductively coupled plasma mass spectrometry.

142. Medical records were used to verify GDM cases, where GDM was defined as at least two of the following plasma glucose values being met in line with American Diabetes Association standards: fasting glucose ≥ 5.3, 1 h ≥ 10.0, 2 h ≥ 8.6, and 3 h ≥ 7.8 mmol/L in response to a 100 g oral glucose load, along with no diagnosis of pregestational diabetes mellitus. Although dietary intake was reported and assessed, intakes of individual elements being measured were not tracked. Poisson regression models with adjustment for covariates were used for determining associations between individual elements and the risk of GDM. Significance was set at p <0.05.

143. Results showed that selenium levels in Hispanic women (geometric mean: 262.31 μg/L) were lower than women of other race and ethnicity. Overall, selenium was not associated with GDM risk in either the adjusted or unadjusted analyses. However, in Hispanic women, modified Poisson regression models found a positive association between selenium concentration and GDM risk (risk ratio: 4.43, 95% CI: 2.25, 8.69) compared to non-Hispanic Black women (RR: 1.18, 95% CI: 0.65, 2.12) (P interaction = 0.01). The authors propose population heterogeneity as an explanation for the null and positive associations seen in Hispanic participants. Additionally, it was suggested that Hispanic populations may be more sensitive to selenium, as blood selenium levels in Hispanic participants were lower than other racial and ethnic groups.

144. A positive association was only observed in Hispanic women and was not observed in the overall study population, suggesting potential differences between populations in the relationship between selenium and GDM. However, the mechanisms for this are unclear. As selenium concentrations were measured after delivery (24 to 72 hours) and information on individual selenium intake was not available, further research would be required to determine whether findings reflect differences in exposure, selenium metabolism, susceptibility, or other variables.

145. Overall, the available additional human data provides limited evidence for the direct toxicity of excess selenium in the maternal diet. Although several studies reported associations between selenium biomarkers and outcomes such as reduced placental weight, lower birth weight, and gestational diabetes mellitus, the findings were inconsistent and no clear mechanisms of toxicity have been established. These studies have a number of limitations, including lack of data on dietary and supplemental selenium intakes and potential for confounding e.g. interactions with other trace elements and metals.It should also be noted that although outcomes such as reduced placental weight or small differences in birth weight were reported, this does not necessarily represent adverse toxic effects.

Animal Data

146. In zebrafish embryos, exposure to selenium concentrations of 0.125 to 1 µM resulted in concentration related increases in mortality and decreased hatchability. At 1 µM, severe deformities were observed in the 50% of embryos that survived. Larvae exposed to 0.5 µM showed impaired locomotor activity and reduced responsiveness to stimuli, suggesting effects on neural development (Zhao et al, 2022). In a separate zebrafish study, exposure to 12.5–100 µg Se/L selenite for 120 days reduced spawning capacity and gonadosomatic index (GSI), delayed oocyte maturation and increased apoptosis in both oocytes and offspring suggesting reproductive toxicity following maternal exposure (Cheng et al, 2023).

147. Similarly, Covington et al (2025) reported reduced survival and an increased incidence of developmental abnormalities in Yellowstone cutthroat trout offspring following maternal selenium exposure. Maternal whole-body selenium concentrations ranged from 2.6 to 25.7 mg/kg dry weight (dw), corresponding to embryo concentrations of 3.4–47.6 mg/kg dw, with effects (survival and abnormality endpoints) observed at egg selenium concentrations of approximately 35 mg/kg dw.

148. Although these findings demonstrate that excessive selenium exposure may lead to adverse developmental and reproductive effects in fish, these species are known to be sensitive to selenium toxicity,with effects often associated with bioaccumulation and maternal transfer to eggs (Uddin et al, 2024). Therefore the relevance of the findings in these studies and their applicability to humans is limited.

Exposure Assessment

Dietary

149. Selenium is found in a variety of foods (Mazmanyan and Gasparyan, 2024; EFSA, 2014, 2023; NHS, 2020; SACN, 2013) such as nuts and seeds, offal, fish and crustaceans, milk and dairy products, meat and meat products, grains and vegetables and legumes. Annex B, Table B1 provides a summary of the selenium content in these specific foods.

150. In EFSA’s intake assessment of selenium in European populations, the main food groups contributing to selenium intake were found to be milk and dairy products, meat and meat products, and grain-based products and fish and fish products (EFSA, 2014).

151. To determine chronic exposure to selenium from the diet a nutrient assessment was conducted utilising the UK Nutrient Databank (NDB) NDB and NDNS years 1-11 data. The NDB contains extensive information on the nutrient content of foods, including selenium, and can be used with the NDNS food consumption data to estimate dietary exposure to specific nutrients. Mean and 97.5th percentile estimates have been provided on a chronic basis, using a population-based exposure assessment approach for women of childbearing ages (16-49 years). This assessment used food groups based on those established for the 2014 Total Diet Study (TDS) for metals and other elements, including selenium.

152. The exposure estimates for women of childbearing age to selenium from the diet are presented in Table 2 for each food group and all food groups combined (referred to as “All” in Table 2). It should be noted, that where food groups are combined, this does not amount to the sum of exposures for all food groups but is the provision of statistical outputs from the distribution of exposure of consumers to selenium across all food groups.

153. Based on consumption of all food groups, the mean dietary exposure to selenium for women of childbearing age is 42 µg/person/day and 0.63 µg/kg bw/day. The high level (97.5th percentile) dietary exposure to selenium for women of childbearing age is 89 µg/person/day and 1.3 µg/kg bw/day.

154. Information on the commodities included within each food group can be found in Annex C, Table C1.

Table 2: Mean and 97.5th percentile (P97.5) chronic exposure to selenium from the diet for women of childbearing age (16-49 years) (NDNS years 1-11).

Food groups	Number of consumers exposed to selenium from the food group	Mean ( µg/ person/ day*)	P97.5 (µg/ person/ day*)	Mean (µg/ kg bw/day*)	P97.5 (µg/ kg bw/day*)
Bread	2432	4.1	11	0.060	0.17
Misc Cereals	2508	6.9	20	0.10	0.30
Carcasse meat	1728	2.8	13	0.040	0.20
Offal	93	0.26	3.7	0.0039	0.055
Meat products	1750	2.5	11	0.036	0.16
Poultry	2004	5.9	21	0.088	0.33
Fish and seafood	1322	8.3	47	0.13	0.67
Fats and oils	454	0.019	0.18	0.00028	0.0025
Eggs	1393	4.3	21	0.063	0.32
Sugars and confectionary	1476	0.31	1.7	0.0047	0.027
Green vegetables	1410	0.22	1.2	0.0033	0.018
Potatoes	677	0.23	2.0	0.0035	0.028
Other vegetables	1977	1.2	6.7	0.017	0.11
Canned vegetables	1059	0.61	3.5	0.0091	0.056
Fresh fruit	512	0.074	0.75	0.0011	0.011
Fruit products	657	0.11	1.1	0.0018	0.016
Non-alcoholic beverages	1278	0.20	1.3	0.0030	0.020
Milk	1639	0.79	3.7	0.012	0.055
Dairy products	2089	1.6	6.1	0.024	0.093
Nuts and seeds	727	0.58	5.5	0.0090	0.085
Alcoholic beverages#	2	0.00091	0	0.000010	0
Meat alternatives	201	0.13	1.7	0.0019	0.027
Snacks	1255	0.15	1.2	0.0022	0.017
Desserts	824	0.29	2.3	0.0043	0.033
Condiments	2280	0.51	2.1	0.0076	0.031
**All	2556	42	89	0.63	1.3

*Rounded to 2 significant figures.

** This is not a sum of all the individual groups.

There are very few consumers exposed to selenium from the alcoholic beverage food group as the large majority of the products within the group do not contain selenium. For this reason, it is possible for the 97.5th percentile to be 0.

Supplements

155. The exposure assessment for supplements assumes that a woman of childbearing age (16-49 years) would consume only one type of supplement and would follow the instructions on either the packaging or the website for the dosage (exposure assessment is based on the highest reported dosage).

156. An internet search was performed to collect information on commercially available selenium supplements on sale in the UK only. Details of the supplements identified from online research are presented in Annex E, Table E1.  Multi-ingredient dietary supplements (MIDS) aimed at women and pregnant women (including pre-conception and post-partum) that are not specifically selenium supplements but contain selenium have been included.

157. The selenium contents in the supplements that were considered in this assessment are summarised in Annex D, Table D1. These contained between 16.5 and 600 µg of selenium per daily dose. MIDS products generally contained lower concentrations of selenium than the selenium only supplements and therefore lower daily doses, especially those aimed at pregnant or breastfeeding women, or those looking to conceive.

158. Based on the range of recommended daily doses (16.5 - 600 µg of selenium), an exposure assessment has been performed using the average bodyweight of women of childbearing age from the NDNS (70.3kg) (years 1-11).  These exposure estimates, from supplements, are presented in Table D2, Annex D. The range of exposure estimates from supplements alone are 0.23 to 8.5 µg/kg bw/day. These values have been rounded to 2 significant figures.

Combined Exposure Scenarios

159. It is possible that women of childbearing age may take a supplement containing selenium at varying dosages alongside their dietary intake of selenium. The scenarios included in this section aim to estimate exposure to selenium from the diet and from supplements for a range of consumers, including average and high-level consumers. These scenarios are as follows: - Scenario 1: Mean consumer range – combines the range of exposure estimates for selenium intake from supplements with the mean dietary exposure estimate for selenium (‘All’ in Table 2), for women of childbearing age on a per day and per kg bodyweight per basis. - Scenario 2: High-level consumer – combines the high-level exposure estimate for selenium intake from supplements (highest recommended daily dose) with the high-level dietary exposure estimate (97.5th percentile) for selenium (‘All’ in Table 2), for women of childbearing age on a per day and per kg bodyweight per basis.

160. The exposure estimates for these scenarios are presented in Table 3.

Table 3: Exposure estimates for scenarios 1 and 2 (combining exposure estimates to selenium from the diet with exposure estimates from supplements)

Scenario	Exposure estimates µg/ person/ day*	Exposure estimates µg/ kg bw/day*
Scenario 1: average consumer range (range of supplementation and average diet)	59 – 640	0.86 – 9.1
Scenario 2: high-level consumer (highest supplementation and diet)	690	9.8

*Rounded to 2 significant figures.

Routes Other Than Food

161. Additional exposures to selenium from drinking water, soil and air were considered.

162. EFSA state that selenium is predominantly present in water as inorganic compounds mainly as selenate. As per the Directive (EU) 2020/2184, the parametric value for selenium in drinking water is 20 µg/L in the EU which can be increased to 30 µg/L in areas where geological conditions could lead to high selenium levels. The maximum legal parameters for selenium in the UK are 10 µg/L as defined in The Water Supply (Water Quality) Regulations 2016. Common estimates for drinking water consumption are that an adult may consume approximately 2L per day, meaning exposure could be up to an additional 0.28ug/kg bw/day (WHO, 2011)

163. Regarding exposure from air, information from a review on the toxicity of elemental selenium and selenium substances after pulmonary exposure; and environmental and occupational air concentrations, suggests that exposure to selenium from air is very low (2-400 ng/day) (Hadrup et al, 2025).

164. Selenium levels in soil from the UK Soil Observatory (UKSO) are reported as 0.92 to 1.46 mg/kg (90th percentile) (UKSO, 2026); this is also low.

165. Overall, exposure from supplements and the diet are the main sources of selenium. As water contributes a lesser amount and exposure from air and soil is negligible, their contribution to overall selenium exposure in women of childbearing age has not been considered further in this assessment.

Uncertainties

166. The following are uncertainties of the exposure assessment.

167. Some assumptions have been made where supplement dosage recommendations are varied. In cases where the dosage is given as a range (e.g. 1-3 capsules per day), the highest possible amount has been assumed to be consumed (e.g. 3 capsules per day). Also, where there is a discrepancy between the packaging image and website instructions, the highest dosage has been assumed e.g. if the website suggests a serving of 2 capsules a day, but the packaging suggests 3, then the highest suggestion has been used.

168. For many supplements listed the warnings provided on the website/ packaging recommend that pregnant or breastfeeding women consult a doctor before taking them, however, there is no guarantee that all of them do.

169. For the dietary intake assessment for selenium, a population-based assessment was performed. Within this dietary intake assessment, there were values of 0 for the 97.5th percentile estimates for “Alcoholic beverages”. This is possible because the large majority of alcoholic drinks do not contain selenium and so there are very few consumers exposed to selenium from this food group.

170. The “Alcoholic beverages” food group has been included in the assessment as the exposure assessment uses women of childbearing ages as a proxy and this consumption may differ to those who are pregnant or trying to conceive.

171. The NDNS does not include pregnant or breastfeeding women and so the nutrient assessment has been performed for women aged 16-49 (women of childbearing age). The diet of this population group may not be fully representative of the maternal diet.

172. In NDNS there is a reported 30% energy intake underestimation. There are many possible reasons for this, resulting from both misreporting and the survey design. However, exposure assessments at the 97.5th percentile are undertaken to ensure that high consumers are accounted for in the assessment, including those who may have mis-reported their energy intake.

173. The following uncertainties are applicable to studies identified from the evidence base including the additional literature search, post EFSA 2023.

174. There are limited human toxicology data in the target population (pregnant women). Much of the available evidence is derived from studies involving men, children or adults above childbearing age. This limits the applicability of findings to pregnant women and women of childbearing age.

175. Many studies have relatively small study populations. This may reduce statistical power and increase uncertainty in associations reported.

176. Studies investigating health effects of selenium in pregnant women or women of childbearing age often assess internal exposure through biomarkers (e.g. blood, serum or placental selenium concentrations) and don’t always report or measure selenium intakes from food or supplements. This limits the ability to evaluate the relationship between external exposure, internal dose, and any observed adverse effects.

177. There is potential confounding from co-exposure to other metals. Co-exposure to different metals is difficult to control, therefore it is difficult to isolate the specific contribution of selenium to the outcomes observed.

178. The mechanism(s) of toxicity for selenium are not fully understood and can vary according to the chemical form of selenium ingested. This increases uncertainty in interpretations of toxicological findings and associations.

179. Some studies rely on self-reporting of symptoms, which may be subject to reporting bias which may affect reliability of the findings.

180. Individual susceptibility to selenium toxicity may vary due to differences in genetics, nutritional status and selenium metabolism. Therefore, some individuals may experience adverse effects at lower selenium exposures than those identified in the available studies.

Risk Characterisation

181. For the purpose of this risk characterisation, a conservative approach has been taken, and exposures have been compared with the lowest UL as set by EFSA (2023) of 255 μg Se/day for adults ≥18 years, including pregnant and lactating women. This UL covers selenium intake from all dietary sources, including forms authorised for addition to foods and use in food supplements. The UL was derived from a LOAEL of 330 μg Se/day (130 μg/day from background diet plus supplemental intake of 200 μg/day) from the SELECT trial (Lippman et al, 2009), with the critical endpoint selected as alopecia.

182. Estimated exposures from food are 42 and 89 µg/person/day at the mean and the 97.5th percentile, respectively (Table 2). These exposures are below the UL of 255 μg Se/day established by EFSA and are therefore not considered to be a concern for pregnant women, women of childbearing age, or foetal development.

183. Combined exposures from food and supplementation range from 59 to 640 µg/ person/ day for the average consumer (scenario 1), and up to 690 µg/ person/ day for the high consumer (scenario 2) (Table 3). While some average consumers may have exposures that remain below the UL, exposures at the upper end of the average consumer range exceed the UL by approximately 2.5-fold. Combined exposures for high consumers (scenario 2) exceed the UL by approximately 2.7-fold.

184. Overall,  there are no estimated exceedances of the UL from dietary exposure alone. Estimated exceedances of the UL arise from combined exposures to selenium from food and supplements, driven by high levels of supplementation. Therefore, pregnant women and women of childbearing age who regularly consume high dose selenium supplementation may be at increased risk of adverse effects associated with excess selenium intake (selenosis). However, there is no current UK NHS recommendation for pregnant women to take selenium supplementation during pregnancy, therefore the proportion of pregnant women consuming high dose selenium supplements is uncertain.

Discussion

185. Selenium is an essential trace element required for the synthesis of selenoproteins that are involved in antioxidant defence, thyroid hormone metabolism, and selenium transport (EVM, 2003; Labunskyy et al., 2014), functions that are important during pregnancy for maternal and foetal health.

186. There is a narrow margin between adequate intake and excessive exposure with high exposures resulting in adverse effects, including selenosis, characterised by changes in hair (alopecia; hair loss, brittleness) and nails (dry, thickened, brittle, discoloured), garlic‑like breath, skin lesions and neurological effects (EVM, 2003; CRN, 2025; IOM, 2000; Hubalewska-Dydejczyk et al, 2020; SCF, 2000).

187. The mechanisms underlying selenium toxicity are not fully established and may differ depending on the form of selenium compound involved (SCF, 2000; EVM, 2003). Proposed mechanisms include oxidative stress, depletion of glutathione, inhibition of protein synthesis, impaired methylation pathways, and substitution of sulphur by selenium in proteins (SCF, 2000; EVM, 2003; EFSA, 2023). EFSA (2023) considered oxidative stress to be the most likely mechanism of toxicity, where excess selenium metabolites generated after saturation of selenium pathways result in formation of reactive oxygen species, leading to cell damage and cell death.

188. Direct evidence of selenium induced toxicity in pregnant women remains limited. No studies identified in the additional literature search directly linked excessive selenium exposure from the diet and supplements with toxicity in pregnant women or women of childbearing age. Although several observational studies reported associations between selenium status and outcomes such as reduced birth weight, reduced placental weight and gestational diabetes mellitus, these studies do not establish causality and do not provide clear evidence that adverse outcomes were attributable to excessive selenium intakes from food or supplements. Such studies were impacted by confounding variables, small study populations, potential reporting bias and limited reporting of selenium intakes from the diet.

189. Studies in fish species indicate that excessive selenium exposure can adversely affect reproduction and development. However, the relevance of these findings to humans is uncertain. Fish species are known to be highly sensitive to selenium toxicity because of bioaccumulation and maternal transfer of selenium to eggs, limiting the direct applicability of these findings to human pregnancy. Although these studies support the biological plausibility of developmental toxicity following excessive selenium exposure, they provide limited evidence for risks in humans.

190. Human evidence informing selenium toxicity thresholds is also subject to limitations. EFSA (2023) noted that many intervention studies were characterised by small sample sizes, high dropout rates and incomplete adverse event reporting. These limitations result in uncertainty when determining the selenium intake at which adverse effects may occur.

191. EFSA (2023) reviewed the NOAEL of 850 μg Se/day derived from Yang et al (1989b), which had previously been used by the SCF (2000) in establishing health-based guidance values. EFSA concluded that evidence of selenium toxicity had been reported at intake levels below this level, indicating that the NOAEL may not adequately reflect current knowledge regarding selenium toxicity. However, the Panel also considered that studies published since Yang et al (1989b) were not suitable for deriving a new NOAEL. EFSA therefore established an UL of 255 μg Se/day for adults, including pregnant and lactating women, using a LOAEL of 330 μg Se/day from the SELECT trial based on alopecia, an early and recognised symptom of selenosis. While the SELECT trial reported increased incidences of alopecia and related effects at selenium intakes of approximately 330 μg/day, these findings were not consistently observed across other studies. Nevertheless, EFSA considered the SELECT trial to represent the most robust dataset available for deriving the UL. An uncertainty factor of 1.3 was applied to account for the use of a LOAEL rather than a NOAEL and the limited data available in women. The Panel considered the UL applicable to pregnant women as available evidence did not indicate sex-specific differences in susceptibility to selenium toxicity. Additionally, the EFSA review highlighted how there may be considerable inter-individual variability in susceptibility to selenium toxicity.

192. Estimated selenium exposures from food alone are 42 and 89 μg/person/day at the mean and 97.5th percentile, respectively, and are therefore well below the EFSA UL of 255 μg/day. These exposure estimates do not indicate a concern for pregnant women, women of childbearing age or foetal development. In contrast, combined exposures from food and supplements range from 59 to 640 μg/person/day for average consumers and up to 690 μg/person/day for high consumers. Exposures at the upper end of these ranges exceed the UL by approximately 2.5 to 2.7-fold, indicating that the potential for excessive exposure comes primarily from supplementation rather than dietary intake. Therefore, women who regularly consume high-dose selenium supplements may be at increased risk of adverse effects associated with excess selenium intake. However, there is currently no UK NHS recommendation for routine selenium supplementation during pregnancy, and the proportion of pregnant women consuming high dose selenium supplements is unknown. It would therefore be prudent for pregnant women to consult an appropriately qualified health professional before consuming these supplements.

Conclusion

193. Dietary selenium exposures in women of childbearing age are well below the EFSA UL of 255 μg/day and therefore unlikely to be of concern for maternal health.

194. Combined exposures from food and high dose supplements may exceed the UL by up to 2.7-fold, suggesting that supplementation rather than dietary intakes of selenium are of concern.

195. Overall, direct evidence of selenium toxicity in pregnant women is limited however despite the lack of sex specific data for women, EFSA (2023) suggest from their review of the literature that there are no sex differences in susceptibility to selenium toxicity.

196. The available evidence supports the conclusion that adverse effects are more likely to occur from high dose selenium supplementation rather than from habitual dietary intake. Should pregnant women wish to consume selenium supplements during pregnancy, they should consult an appropriately qualified health professional.

Questions on which the views of the Committee are sought

1. Does the Committee have any comments on the discussion paper?
2. Does the Committee agree on the use of the more conservative EFSA 2023 UL of 255 μg/person/day, or should the previous EVM UL of 400 μg /person/day be used for risk characterisation?
3. Does the Committee agree with contents and structure of the discussion paper?
4. Does the Committee have any other comments

Secretariat

August 2026

List of Abbreviations and Technical terms

Abbreviations	Technical term
ADI	Acceptable Daily Intake
ADI	Acceptable Daily Intake
ADME	Absorption, Distribution, Metabolism and Excretion
AI	Adequate Intake
ATSDR	Agency for Toxic Substances and Disease Registry
apoER2	Apolipoprotein E Receptor 2
BKMR	Bayesian Kernel Machine Regression
CI	Confidence Interval
COT	Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment
CRN	Council for Responsible Nutrition
DMSe	Dimethylselenide
DRI	Dietary Reference Intake
EFSA	European Food Safety Authority
EVM	Expert Group on Vitamins and Minerals
FAO	Food and Agriculture Organization of the United Nations
FSA	Food Standards Agency
GAM	Generalised Additive Model
GDM	Gestational Diabetes Mellitus
GPx	Glutathione Peroxidase
GPx1	Glutathione Peroxidase 1
GPx3	Glutathione Peroxidase 3
GSI	Gonadosomatic Index
HBGV	Health-Based Guidance Value
ICP-MS	Inductively Coupled Plasma Mass Spectrometry
IOM	Institute of Medicine
INMT	Indolethylamine N-Methyltransferase
LOAEL	Lowest Observed Adverse Effect Level
MeSeCys	Se-Methyl-Selenocysteine
MIDS	Multi-Ingredient Dietary Supplements
mtDNAcn	Mitochondrial DNA Copy Number
NDB	Nutrient Databank
NDA Panel	Panel on Nutrition, Novel Foods and Food Allergens
NDNS	National Diet and Nutrition Survey
NHS	National Health Service
NHMRC	National Health and Medical Research Council
NOAEL	No Observed Adverse Effect Level
NRV	Nutrient Reference Value
PCR	Polymerase Chain Reaction
P97.5	97.5th Percentile
RCT	Randomised Controlled Trial
RNI	Reference Nutrient Intake
RR	Relative Risk
SACN	Scientific Advisory Committee on Nutrition
SCF	Scientific Committee on Food
Se	Selenium
SeCys	Selenocysteine
SelP	Selenoprotein P
SeMet	Selenomethionine
TMSe	Trimethylselenonium Ion
TDS	Total Diet Study
TL	Telomere Length
WBSe	Whole Blood Selenium
UKSO	UK Soil Observatory
UL	Tolerable Upper Intake Level
VC	Venous Cord
WHO	World Health Organization
UK	United Kingdom

References

Adebayo, A.H., Yakubu, O.F. & Bakare‑Akpata, O. (2020) Uptake, Metabolism and Toxicity of Selenium in Tropical Plants. In: Importance of Selenium in the Environment and Human Health. IntechOpen. DOI: 10.5772/intechopen.90295 Uptake, Metabolism and Toxicity of Selenium in Tropical Plants IntechOpen

Agency for Toxic Substances and Disease Registry (ATSDR) (2013) Toxicological profile for selenium. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service. Selenium Toxicological Profile ATSDR

Alfthan G, Aro A, Arvilommi H and Huttunen JK, (1991) Selenium metabolism and platelet glutathione peroxidase activity in healthy Finnish men: effects of selenium yeast, selenite, and selenate. American Journal of Clinical Nutrition, 53, 120-125.

Algotar, A.M., Stratton, M.S., Ahmann, F.R., Ranger-Moore, J., Nagle, R.B., Thompson, P.A., Slate, E., Hsu, C.H., Dalkin, B.L., Sindhwani, P., Holmes, M.A., Tuckey, J.A., Graham, D.L., Parnes, H.L., Clark, L.C. and Stratton, S.P. (2013b) ‘Phase 3 clinical trial investigating the effect of selenium supplementation in men at high-risk for prostate cancer’, Prostate, 73(3), pp. 328–335. https://doi.org/10.1002/pros.22573.

Álvarez-Silvares, E., Fernández-Cruz, T., Bermúdez-González, M., Rubio-Cid, P., Almeida, A., Pinto, E., Seoane-Pillado, T. and Martínez-Carballo, E. (2023). Placental levels of essential and non-essential trace elements in relation to neonatal weight in northwestern Spain: application of generalized additive models. Environmental Science and Pollution Research, 30, pp. 62566–62578.

Amazon (2026) Selenium Supplements 200mcg (40mg L-Selenomethionine).  Selenium Supplements 200mcg - 365 Tablets - 1 Year Supply - Bioavailable Sodium Selenite Form - 1-a-Day - Small in Size (6mm) - Vegan & Non-GMO Selenium Tablets : Amazon.co.uk: Health & Personal Care Viewed 22/06/2026.

Amazon (2026) Selenium 200 mcg Tablets.  Selenium 200mcg Tablets - 180 Vegan Society Registered Pills, for Thyroid and Immune Health, Hair, Skin and Nails - 6 Month Supply Tablets Not Capsules - Made in The UK By YrHealth : Amazon.co.uk: Health & Personal Care Viewed 22/06/2026.

Ashton K, Hooper L, Harvey LJ, Hurst R, Casgrain A and Fairweather-Tait SJ, (2009) Methods of assessment of selenium status in humans: a systematic review. American Journal of Clinical Nutrition, 89, 2025S-2039S.

Bates, B.; Collins, D.; Jones, K.; Page, P.; Roberts, C.; Steer, T.; Swan, G.(2020) National Diet and Nutrition Survey Results from years 9, 10 and 11 (combined) of the Rolling Programme (2016/2017 to 2018/2019) National Diet and Nutrition Survey

Bates, B.; Cox, L.; Nicholson, S.; Page, P.; Prentice, A.; Steer, T.; Swan, G. (2016) National Diet and Nutrition Survey Results from Years 5 and 6 (combined) of the Rolling Programme (2012/2013 – 2013/2014) Main heading

Bates, B.; Lennox, A.; Prentice, A.; Bates, C.; Page, P.; Nicholson, S.; Swan, G. (2014) National Diet and Nutrition Survey Results from Years 1, 2, 3 and 4 (combined) of the Rolling Programme (2008/2009 – 2011/2012) Main heading

BioCare (2026) BioCare Selenium [CT2] vegetable capsules. Selenium Supplement Capsules BioCare Viewed 22/06/2026.

Brätter, P. & de Brätter, V.E.N. (1996) Influence of high dietary selenium intake on the thyroid hormone level in human serum. Journal of Trace Elements in Medicine and Biology, 10(3), pp. 163–166. doi:10.1016/S0946-672X(96)80027-7. Influence of High Dietary Selenium Intake on the Thyroid Hormone Level in Human Serum - ScienceDirect

BodybBio (2026) Liquid Selenium Supplement Liquid Minerals BodyBio – BodyBio UK Liquid Selenium Supplement. Viewed 22/06/2026.

Broadley, M.R., White, P.J., Bryson, R.J., Meacham, M.C., Bowen, H.C., Johnson, S.E., Hawkesford, M.J., McGrath, S.P., Zhao, F.J., Breward, N., Harriman, M. and Tucker, M. (2006). Biofortification of UK food crops with selenium. Proceedings of the Nutrition Society, 65(2), pp.169–181. Biofortification of UK food crops with selenium Proceedings of the Nutrition Society Cambridge Core

Boots (2026) Boots A-Z Vitamins and minerals. Boots A-Z 180 Tablets (6 month supply) - Boots. Viewed 22/06/2026.

Boots (2026) Boots Selenium with Vitamins A, C and E. Boots Selenium with Vitamins A, C and E 60 Tablets (2 month supply) - Boots. Viewed 22/06/2026.

Boots (2026) Proceive Pregnancy Supplement Trimester 1. Proceive Pregnancy Supplement Trimester 1 Capsules 60s - Boots. Viewed 22/06/2026.

Burk R.F, Hill K.E and Motley A.K (2001) Plasma selenium in specific and non-specific forms. Biofactors, 14, 107-114.

Burk, R.F., Norsworthy, B.K., Hill, K.E., Motley, A.K. and Byrne, D.W. (2006) ‘Effects of chemical form of selenium on plasma biomarkers in a high-dose human supplementation trial’, Cancer Epidemiology, Biomarkers & Prevention, 15(4), pp. 804–810. https://doi.org/10.1158/1055-9965.EPI-05-0950

Burk, R.F., Olson, G.E., Hill, K.E., Winfrey, V.P., Motley, A.K. and Kurokawa, S. (2013) ‘Maternal-fetal transfer of selenium in the mouse’, FASEB Journal, 27(8), pp. 3249–3256. https://doi.org/10.1096/fj.13-231852

Cardoso, B.R., Duarte, G.B.S., Reis, B.Z. and Cozzolino, S.M.F. (2017). Brazil nuts: Nutritional composition, health benefits and safety aspects. Food Research International, 100(2), pp.9–18. https://doi.org/10.1016/j.foodres.2017.08.036

Chawla, R., Filippini, T., Loomba, R., Cilloni, S., Dhillon, K.S. and Vinceti, M. (2020) ‘Exposure to a high selenium environment in Punjab, India: Biomarkers and health conditions’, Science of the Total Environment, 719, 134541. https://doi.org/10.1016/j.scitotenv.2019.134541

Cheng, R., Zhang, Z., Zhan, C., Qin, T., Wang, L. and Zhang, X. (2023). Environmentally relevant concentrations of selenite trigger reproductive toxicity by affecting oocyte development and promoting larval apoptosis. Environmental Pollution, 316(Pt 1), 120648. https://doi.org/10.1016/j.envpol.2022.120648

Clark, L.C., Combs, G.F. Jr, Turnbull, B.W., Slate, E.H., Chalker, D.K., Chow, J., Davis, L.S., Glover, R.A., Graham, G.F., Gross, E.G., Krongrad, A., Lesher, J.L. Jr, Park, H.K., Sanders, B.B. Jr, Smith, C.L., Taylor, J.R. (1996) Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. Journal of the American Medical Association 276, 1957-1963.

Combs, G.F. Jr (2015) ‘Biomarkers of selenium status’, Nutrients, 7(4), pp. 2209–2236. doi:10.3390/nu7042209.

Combs GF, Jr., (2001). Selenium in global food systems. British Journal of Nutrition, 85, 517-547.

Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) (2019) Overarching statement on the potential risks from contaminants in the diet of infants aged 0 to 12 months and children aged 1 to 5 years. London: Committee on Toxicity. Overarching Statement

Council for Responsible Nutrition (CRN) (2025) Vitamin and Mineral Safety Review: Selenium. Washington, DC: Council for Responsible Nutrition. Vitamin and Mineral Safety Review: Selenium

Covington, S.M., Naddy, R.B., Prouty, A.L., Petach, M. and Schmeda, G. (2025) Effects of in situ selenium exposure and maternal transfer on survival and abnormalities of Yellowstone cutthroat trout (Oncorhynchus clarkii bouvieri) fry. Integrated Environmental Assessment and Management, 21(3), pp.674–687. https://doi.org/10.1093/inteam/vjaf013.

Cytoplan (2026) Cytoplan women’s wholefood multivitamin capsules. Women’s Wholefood Multi - Multivitamin Tailored for Women Viewed 22/06/2026.

Dorea, J.G. (2002) ‘Selenium and breast-feeding’, British Journal of Nutrition, 88(5), pp. 443–461.

Drake, V.J. (2024) Selenium. Linus Pauling Institute, Oregon State University. Selenium Linus Pauling Institute Oregon State University (Updated November 2024).

Duffield AJ, Thomson CD, Hill KE and Williams S, (1999). An estimation of selenium requirements for New Zealanders. American Journal of Clinical Nutrition, 70, 896-903.

Duffield-Lillico AJ, Slate EH, Reid ME, Turnbull BW, Wilkins PA, Combs GF Jr, Park HK, Gross EG, Graham GF, Stratton MS, Marshall JR, Clark LC (2003) Nutritional Prevention of Cancer Study Group. Selenium supplementation and secondary prevention of nonmelanoma skin cancer in a randomized trial. J Natl Cancer Inst;95:1477–81.

EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA) (2014) ‘Scientific Opinion on Dietary Reference Values for selenium’, EFSA Journal, 12(10), Article 3846, 67 pp. Scientific Opinion on Dietary Reference Values for selenium - - 2014 - EFSA Journal - Wiley Online Library

EFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA) (2023) ‘Scientific opinion on the tolerable upper intake level for selenium’, EFSA Journal, 21(2), Article e07704. Scientific opinion on the tolerable upper intake level for selenium - - 2023 - EFSA Journal - Wiley Online Library

Eidon Ionic Minerals (2026) Eidon Ionic Minerals, Selenium, Liquid Concentrate, 2 oz (60 ml) Eidon Ionic Minerals, Selenium, Liquid Concentrate, 2 oz (60 ml)  Viewed 22/06/2026.

European Parliament and Council of the European Union (2002) Directive 2002/46/EC of the European Parliament and of the Council of 10 June 2002 on the approximation of the laws of the Member States relating to food supplements. Official Journal of the European Communities, L183, pp. 51–57. Directive - 2002/46 - EN - EUR-Lex

Expert Group on Vitamins and Minerals (EVM) (2003) Safe Upper Levels for Vitamins and Minerals. London: Food Standards Agency. Safe Upper Levels for Vitamins and Minerals

Fairris, G.M., Lloyd, B., Hinks, L., Perkins, P.J. and Clayton, B.E. (1989) ‘The effect of supplementation with selenium and vitamin E in psoriasis’, Annals of Clinical Biochemistry, 26(1), pp. 83–88. https://doi.org/10.1177/000456328902600113

Fairweather-Tait SJ, Collings R and Hurst R, 2010. Selenium bioavailability: current knowledge and future research requirements. American Journal of Clinical Nutrition, 91, 1484S-1491S.

Ferm VH, Hanlon DP, Willhite CC, Choy WN and Book SA (1990). Embryotoxicity and dose-response relationships of selenium in hamsters. Reprod Toxicol 4: 183-190.

Food Standard Agency (2014) Total diet study: Metals and other elements.  [ARCHIVED CONTENT]

Free Soul (2026) Free Soul Hair Skin and Nails Gummies. Hair, Skin & Nails Gummies – Free Soul. Viewed 22/06/2026.

Free Soul (2026) Free Soul Women’s Multivitamins with collagen. Multivitamins with Collagen – Free Soul. Viewed 22/06/2026.

Grape Tree (2026) Grape Tree Selenium 200mg. Grape Tree Selenium 200mg 240 Tablets Grape Tree. Viewed 22/06/2026.

Grundeken, M., Gustin, K., Vahter, M., Delaval, M., Barman, M., Sandin, A., Sandberg, A-S., Wold, A.E., Broberg, K. and Kippler, M. (2024). Toxic metals and essential trace elements in placenta and their relation to placental function. Environmental Research, 248, 118355. Toxic metals and essential trace elements in placenta and their relation to placental function

Ha, H.Y., Alfulaij, N., Berry, M.J. and Seale, L.A. (2019). From Selenium Absorption to Selenoprotein Degradation. Biological Trace Element Research, 192(1), pp. 26-37. From Selenium Absorption to Selenoprotein Degradation Biological Trace Element Research Springer Nature Link

Hadrup, N., Poulsen, M. and Ravn-Haren, G. (2026) Toxicity of elemental selenium and selenium substances after pulmonary exposure; and environmental and occupational air concentrations: a review. Drug and Chemical Toxicology, 49(2), pp.428–438. https://doi.org/10.1080/01480545.2026.2620397

Holland & Barrett (2026) Bassetts Vitamins Pregnancy + Omega 3 DHA Multivitamins & Multiminerals Strawberry & Orange 30 Pastilles. Bassetts Vitamins Pregnancy + Omega 3 DHA Multivitamins & Multiminerals Strawberry & Orange 30 Pastilles H&B. Viewed 22/06/2026.

Holland & Barrett (2026) Centrum Women. Centrum Women (60 Tablets) H&B. Viewed 22/06/2026.

Holland & Barrett (2026) Holland & Barrett Conception & Pregnancy Support. Conception & Pregnancy Support - 30 Tablets Holland & Barrett Viewed 22/06/2026.

Holland & Barrett (2026) Holland and Barrett Ultra woman.  Holland & Barrett Ultra Woman Tablets 3 Months Supply Bundle H&B Viewed 22/06/2026.

Holland & Barrett (2026) Neu Mum Daily Pregnancy Multivitamin Neubria Neu Mum Daily Pregnancy Multivitamin. Food Supplement 30 Tablets H&B Viewed 22/06/2026.

Holland & Barrett (2026) Seven Seas Omega-3 and Multivitamins Omega-3 & Multivitamins Woman Duo Pack Seven Seas H&B Viewed 22/06/2026.

Huang Z, Rose AH and Hoffmann PR, (2012). The role of selenium in inflammation and immunity: from molecular mechanisms to therapeutic opportunities. Antioxidants and Redox Signalling, 16,705-743.

Hubalewska-Dydejczyk, A., Duntas, L. and Gilis-Januszewska, A. (2020). Pregnancy, thyroid, and the potential use of selenium. Hormones, 19(1), pp.47–53.  Pregnancy, thyroid, and the potential use of selenium Hormones Springer Nature Link

Hurst R, Armah CN, Dainty JR, Hart DJ, Teucher B, Goldson AJ, Broadley MR, Motley AK and Fairweather-Tait SJ, (2010). Establishing optimal selenium status: results of a randomized, double-blind, placebo-controlled trial. American Journal of Clinical Nutrition, 91, 923-931.

Hurst R, Collings R, Harvey LJ, King M, Hooper L, Bouwman J, Gurinovic M and Fairweather-TaitSJ, (2013) EURRECA-Estimating selenium requirements for deriving dietary reference values. Critical Reviews in Food Science and Nutrition, 53, 1077-109.

Inessa (2026) Inessa advanced multivitamin. Inessa Advanced Multivitamin. Viewed 22/06/2026.

Institute of Medicine (IOM) (2000) Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. National Academies Press, Washington, DC.Chapter 7. Pg 284 – 315. Dietary Reference Intakes

Jäger, T., Drexler, H. and Göen, T. (2016a) ‘Human metabolism and renal excretion of selenium compounds after oral ingestion of sodium selenate dependent on trimethylselenium ion (TMSe) status’, Archives of Toxicology, 90(1), pp. 149–158. doi:10.1007/s00204-014-1380-x.

Jäger, T., Drexler, H. and Göen, T. (2016b) ‘Human metabolism and renal excretion of selenium compounds after oral ingestion of sodium selenite and selenized yeast dependent on the trimethylselenium ion (TMSe) status’, Archives of Toxicology, 90(5), pp. 1069–1080. doi:10.1007/s00204-015-1548-z.

Jensen R, Clossen W and Rothenberg R (1984). Selenium intoxication. New York Morb Mortal Wkly Rep 33: 157-158.

Kawamoto, T., Nitta, H., Murata, K., Toda, E., Tsukamoto, N., Hasegawa, M., Yamagata, Z., Kayama, F., Kishi, R., Ohya, Y., Saito, H., Sago, H., Okuyama, M., Ogata, T., Yokoya, S., Koresawa, Y., Shibata, Y., Nakayama, S., Michikawa, T., Takeuchi, A. and Working Group of the Epidemiological Research for Children’s Environmental Health (2014) ‘Rationale and study design of the Japan Environment and Children’s Study (JECS)’, BMC Public Health, 14, 25. Rationale and study design of the Japan environment and children’s study (JECS) BMC Public Health Springer Nature Link

Kinjo, Y., Shibata, E., Askew, D.J., Tanaka, R., Suga, R., Shimono, M., Sakuragi, T., Morokuma, S., Ogawa, M., Sanefuji, M., Hamada, N., Ochiai, M., Ohga, S., Tsuji, M., Kusuhara, K. and Yoshino, K. (2024) ‘Association of placental weight at birth with maternal whole blood concentration of heavy metals (cadmium, lead, mercury, selenium, and manganese): The Japan Environment and Children’s Study (JECS)’, Environment International, 188, 108725. https://doi.org/10.1016/j.envint.2024.108725

Kyriakou, S.I., Tsarna, E., Stachika, N., Dalla, C., Potiris, A., Stavros, S. and Christopoulos, P. (2026). The Impact of Selenium Exposure During Pregnancy on Risk for Miscarriage: A Systematic Review. International Journal of Molecular Sciences, 27(2), 968. https://doi.org/10.3390/ijms27020968

Labunskyy, V. M., Hatfield, D. L., & Gladyshev, V. N. (2014). Selenoproteins: Molecular pathways and physiological roles. Physiological Reviews, 94, 739–777. https://doi.org/10.1152/physrev.00039.2013

Lajin, B. and Francesconi, K.A. (2016) ‘The association between the urinary excretion of trimethylselenonium and trimethylsulfonium in humans’, PLoS ONE, 11(11), e0167013. https://doi.org/10.1371/journal.pone.0158765

Lemire, M., Philibert, A., Fillion, M., Passos, C.J.S., Guimarães, J.R.D., Barbosa, F. Jr and Mergler, D. (2012) ‘No evidence of selenosis from a selenium-rich diet in the Brazilian Amazon’, Environment International, 40, pp. 128–136. https://doi.org/10.1016/j.envint.2011.07.005

Lamberts (2026) Selenium 200µg. Selenium 200µg Lamberts Healthcare. Viewed 22/06/202.

Levander, O.A. (1987) ‘A global view of human selenium nutrition’, Annual Review of Nutrition, 7, pp. 227–250. https://doi.org/10.1146/annurev.nu.07.070187.001303

Lippman SM, Klein EA, Goodman PJ, Lucia MS, Thompson IM, Ford LG, Parnes HL, Minasian LM, Gaziano JM,Hartline JA, Parsons JK, Bearden JD 3rd, Crawford ED, Goodman GE, Claudio J, Winquist E, Cook ED, Karp DD,Walther P, Lieber MM, Kristal AR, Darke AK, Arnold KB, Ganz PA, Santella RM, Albanes D, Taylor PR, ProbstfieldJL, Jagpal TJ, Crowley JJ, Meyskens FL Jr, Baker LH and Coltman CA Jr, 2009. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT).The Journal of the American Medical Association, 301, 39–51. https://doi.org/10.1001/jama.2008.864

Li, Z., Wang, G., Hong, X., Juraschek, S.P., Ngo, L.H., Wang, X. and Zhang, M. (2025). Supporting Information for: Associations of Heavy Metals and Trace Elements with Gestational Diabetes Mellitus in the Boston Birth Cohort. Environmental Science & Technology, 59(38), pp. 20307–20315.

Associations of Heavy Metals and Trace Elements with Gestational Diabetes Mellitus in the Boston Birth Cohort Environmental Science & Technology ACS Publications

Longnecker, M.P., Taylor, P.R., Levander, O. A., Howe, S.M., Veillon, C., McAdam, P.A., Patterson, K.Y., Holden, J.M., Stampfer, M.J., Morris, J.S., Willett, W.C. (1991) Selenium in diet, blood and toenails in relation to human health in a seleniferous area. American Journal of Clinical. Nutrition 53, 1288-94.

Maehira F, Luyo GA, Miyagi I, Oshiro M, Yamane N, Kuba M and Nakazato Y, (2002). Alterations ofserum selenium concentrations in the acute phase of pathological conditions. Clinica Chimica Acta,316, 137-146.

Martens, I.B., Cardoso, B.R., Hare, D.J., Niedzwiecki, M.M., Lajolo, F.M., Martens, A. and Cozzolino, S.M.F. (2015) ‘Selenium status in preschool children receiving a Brazil nut-enriched diet’, Nutrition, 31(11–12), pp. 1339–1343. https://doi.org/10.1016/j.nut.2015.05.005.

Mazmanyan, V. and Gasparyan, A. (2024) Selenium — Sources, Health Benefits, Deficiency, and Side Effects. [online] Food Struct. Selenium — Sources, Health Benefits, Deficiency, and Side Effects

Mehta R, Krupa C, Ahmed T, Hamer DH, Al Mahmud A. Associations between maternal and infant selenium status and child growth in a birth cohort from Dhaka, Bangladesh. Br J Nutr. 2023 Nov 14;130(9):1558-1572. doi: 10.1017/S0007114523000739. Epub 2023 Mar 22.

Metabolics (2026) Ionic Selenium Metabolics Ionic Selenium Supplement 100ml. Viewed 22/06/2026.

National Institutes of Health Office of Dietary Supplements (NIH ODS) (2025) Selenium - Health Professional Fact Sheet. Selenium - Health Professional Fact Sheet

New Leaf Products (2026) Biotin Hair Growth Vitamins 12,000mcg D-Biotin Tablets Enriched with Zinc & Selenium. Biotin+ Vitamin B7 Tablets New Leaf Supplements – New Leaf Products. Viewed 22/06/2026.

NHS (2020) Vitamins and Minerals. [online] NHS. Vitamins and minerals - Others - NHS

NHMRC (2026). Nutrient Reference Values - Public Consultation on Draft Recommendations – Selenium. National Health and Medical Research Council, Australia. Nutrient Reference Values

Nichol C, Herdman J, Sattar N, O’Dwyer PJ, St JORD, Littlejohn D and Fell G, 1998. Changes in theconcentrations of plasma selenium and selenoproteins after minor elective surgery: furtherevidence for a negative acute phase response? Clinical Chemistry, 44, 1764-1766.

Nutracraft (2026) ThyroAid - Thyroid Support Supplement. ThyroAid Thyroid Support Nutracraft. Viewed 22/06/2026.

Nutravita (2026) Nutravita multivitamins. Multivitamin Minerals - Vegan & Vegetarian Multivitamins - Nutravita Nutravita  Viewed 22/06/2026.

Pacheco, A.M. and Scussel, V.M. (2007). Selenium and Aflatoxin Levels in Raw Brazil Nuts from the Amazon Basin. Journal of Agricultural and Food Chemistry, 55(26), pp.11087–11092. https://doi.org/10.1021/jf072434k

Perrone, D., Monteiro, M. and Nunes, J.C. (2015) ‘The Chemistry of Selenium’, in Preedy, V.R. (ed.) Selenium: Chemistry, Analysis, Function and Effects. Cambridge: Royal Society of Chemistry, pp. 3–15. https://doi.org/10.1039/9781782622215-00003

Picciano, M.F. and McGuire, M.K. (2018). Selenium in pregnancy and breastfeeding. In: Lammi-Keefe, C.J., Couch, S.C. and Philipson, E.H. (eds.) Nutrition and Lifestyle for Pregnancy and Breastfeeding. Oxford: Oxford University Press. Selenium in pregnancy and breastfeeding Nutrition and Lifestyle for Pregnancy and Breastfeeding Oxford Academic

Proceive (2026) PROCEIVE® Women Advanced Fertility Supplement Proceive Women. Viewed 22/06/2026.

Rayman, M.P. (2000). The importance of selenium to human health. The Lancet, 356(9225), pp.233–241. https://doi.org/10.1016/S0140-6736(00)02490-9

Rayman MP, Infante HG and Sargent M, (2008). Food-chain selenium and human health: spotlight onspeciation. British Journal of Nutrition, 100, 238-253.

Robberecht H and Deelstra H, (1994). Factors influencing blood selenium concentration values: aliterature review. Journal of Trace Elements and Electrolytes in Health and Disease, 8, 129-143.

Roberts, C.; Steer, T.; Maplethorpe, N.; Cox, L.; Meadows, S.; Page, P.; Nicholson, S.; Swan, G. (2018) National Diet and Nutrition Survey Results from Years 7 and 8 (combined) of the Rolling Programme (2014/2015 – 2015/2016) National Diet and Nutrition Survey

Sainsbury’s (2026) Sainsbury’s Pregnancy Plus Essential Vitamins x30 Sainsbury’s

Schroeder HA and Mitchener M (1971). Toxic effects of trace elements on the reproduction of rats and mice. Arch Environ Health 23: 102-106.

Scientific Advisory Committee on Nutrition (SACN) (2013) SACN Statement on Selenium and Health. London: The Stationery Office. Statement on Selenium and Health

Scientific Committee on Food (SCF) 2000, Opinion of the Scientific Committee on Food on the Tolerable Upper Intake Level of Selenium, European Commission, Brussels. Microsoft Word - Seleniumupplev25finalf.doc

Stoffaneller, R. and Morse, N.L. (2015). A review of dietary selenium intake and selenium status in Europe and the Middle East. Nutrients, 7(3), pp.1494–1537. https://doi.org/10.3390/nu7031494

Supplemented (2026) Selenium 200mcg & ACE RDA Tablets. Selenium 200mcg & ACE RDA Tablets – Supplemented Viewed 22/06/2026.

The Swallow Co. (2026) Swallow multivitamin. Daily Multivitamin for Men & Women The Swallow Co Viewed 22/06/2026.

Tarantal, A.F., Willhite, C.C., Lasley, B.L., Murphy, C.J., Miller, C.J., Cukierski, M.J., Book, S.A. and Hendrickx, A.G. (1991) ‘Developmental toxicity of L-selenomethionine in Macaca fascicularis’, Fundamental and Applied Toxicology, 16(1), pp. 147–160. https://doi.org/10.1016/0272-0590(91)90143-R

The Food Foundation (2026) The Broken Plate 2026. London: The Food Foundation. The Broken Plate

Thompson, P.A., Ashbeck, E.L., Roe, D.J., Fales, L., Buckmeier, J., Wang, F., Bhattacharyya, A., Hsu, C.H., Chow, H.H., Ahnen, D.J., Boland, C.R., Heigh, R.I., Fay, D.E., Hamilton, S.R., Jacobs, E.T., Martinez, M.E., Alberts, D.S. and Lance, P. (2016) ‘Selenium supplementation for prevention of colorectal adenomas and risk of associated type 2 diabetes’, Journal of the National Cancer Institute, 108(12), djw152. https://doi.org/10.1093/jnci/djw152

Uddin, M.H., Ritu, J.R., Putnala, S.K., Rachamalla, M., Chivers, D.P. and Niyogi, S. (2024). Selenium toxicity in fishes: A current perspective. Chemosphere, 364, 143214. https://doi.org/10.1016/j.chemosphere.2024.143214

Viridian (2026) Viridian Selenium Vegicaps. https://viridian-nutrition.com/products/selenium-200ug Viewed 22/06/2026

Vitabiotics (2026) Vitabiotics Pregnacare 60 Vegan Gummies Pregnacare® Gummies Pregnancy Gummies Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Breastfeeding. Pregnacare® Breastfeeding Postnatal Vitamins Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Conception Tablets. Pregnacare® Conception Tablets By Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Him & Her Conception. Pregnacare® Him & Her Conception Tablets By Vitabiotics. Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Liquid. Pregnacare® Liquid Pregnancy Vitamins Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Max. Pregnacare® Max Pregnancy Vitamins By Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Pregnacare Original. Pregnacare® Original Pregnancy Supplement Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Ultra Selenium Tablets. Ultra Selenium Tablets Selenium Supplement Vitabiotics® Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Wellwoman Max. Wellwoman® Max Supplement For Women By Vitabiotics Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Wellwoman Multivitamin Gummies for Women. Wellwoman® Multi - Vitamin Gummies For Women Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Wellwoman Original. Wellwoman Original Multivitamin For Women Vitabiotics® Viewed 22/06/2026.

Vitabiotics (2026) Vitabiotics Wellwoman Plus. Wellwoman® Plus Omega 3-6-9 By Vitabiotics Viewed 22/06/2026.

Wahlstrom, R.C. and Olson, O.E. (1959). The Effect of Selenium on Reproduction in Swine. Journal of Animal Science, 18(1), 141–145.

WeightWorld (2026) WeightWorld Multivitamin and Mineral Multivitamin and Mineral Tablets x 400 27 Nutrients WeightWorld Viewed 22/06/2026.

White, P.J. (2018) ‘Selenium metabolism in plants’, Biochimica et Biophysica Acta (BBA) – General Subjects, 1862(11), pp. 2333–2342. https://doi.org/10.1016/j.bbagen.2018.05.006

Wild Nutrition (2026) Wild Nutrition Food Grown Daily Multi Nutrient for Women. Food-Grown Women’s Daily Multi Nutrient Multivitamins For Women. Viewed 22/06/2026

Winther, K.H., Bonnema, S.J., Cold, F., Debrabant, B., Nybo, M., Cold, S. and Hegedüs, L. (2015) ‘Does selenium supplementation affect thyroid function? Results from a randomized, controlled, double-blinded trial in a Danish population’, European Journal of Endocrinology, 172(6), pp. 657–667. https://doi.org/10.1530/EJE-15-0069.

Woods Health (2026) Selenium + Vitamins A, C and E. Selenium + Vitamins A, C and E Woods Health Supplements And Vitamins. Viewed 22/06/2026

World Health Organization (WHO) (2011). Guidelines for Drinking-water Quality, 4th ed. Geneva: World Health Organization. Guidelines for drinking-water quality: fourth edition incorporating the first, second and third addenda

Xia Y, Hill KE, Byrne DW, Xu J and Burk RF, 2005. Effectiveness of selenium supplements in a low-selenium area of China. American Journal of Clinical Nutrition, 81, 829-834.

Xia Y, Hill KE, Li P, Xu J, Zhou D, Motley AK, Wang L, Byrne DW and Burk RF, 2010.Optimization of selenoprotein P and other plasma selenium biomarkers for the assessment of the selenium nutritional requirement: a placebo-controlled, double-blind study of selenomethionine supplementation in selenium-deficient Chinese subjects. American Journal of Clinical Nutrition,92, 525-531.

Yang, G., Wang, S., Zhou, R., Sun, S. (1983) Endemic selenium intoxication of humans in China. American Journal of Clinical Nutrition 37, 872-881.

Yang G, Yin S, Zhou R, Gu L, Yan B, Liu Y and Liu Y (1989b). Studies of safe maximal daily selenium intake in a seleniferous area in China. Part II J Trace Elem Electrolytes Health Dis 3: 123-130.

Yang G and Zhou R (1994). Further observations on the human maximum safe dietary selenium intake in a seleniferous area of China. J Trace Elem Electolytes Health Dis 8: 159- 165.

Yang GQ, Zhu LZ, Liu SJ, Gu LZ, Qian PC, Huang JH and Lu MD, 1987. Human selenium requirements in China. In: Selenium in Biology and Medicine. Eds Combs GF, Spallholz JE,Levander OA and Oldfield JE. Nostrand Rheinhold/ AVI, New York, USA, 589-607.

Yang, G., Zhou, R., Yin S., Gu, L., Yan, B., Liu, Y., Liu, Y., Li, X. (1989a) Studies of safe maximal daily selenium intake in a seleniferous area in China. I. Selenium intake and tissue selenium levels of the inhabitants. Journal of Trace Elements and Electrolytes in Health and Disease 3, 77-87.

Zachariah, M., Maamoun, H., Milano, L., Rayman, M.P., Meira, L.B. and Agouni, A. (2021). Endoplasmic reticulum stress and oxidative stress drive endothelial dysfunction induced by high selenium. Journal of Cellular Physiology, 236(6), pp.4348-4359. Journal of Cellular Physiology Cell Biology Journal Wiley Online Library

Zhao, G., Hu, J., Gao, M., Zhu, Y. and Hong, Y. (2022). Excessive selenium affects neural development and locomotor behavior of zebrafish embryos. Ecotoxicology and Environmental Safety, 238, 113611. Excessive selenium affects neural development and locomotor behavior of zebrafish embryos - ScienceDirect