www.thelancet.com Vol 379 March 31, 2012
Lancet 2012; 379: 1256–68
February 29, 2012
Faculty of Health and Medical
Sciences, University of Surrey,
(Prof M P Rayman DPhil [Oxon])
Prof Margaret P Rayman, Faculty
of Health and Medical Sciences,
University of Surrey,
Guildford GU2 7XH, UK
Selenium and human health
Margaret P Rayman
Selenium is incorporated into selenoproteins that have a wide range of pleiotropic eff ects, ranging from anti oxidant
and anti-infl ammatory eff ects to the production of active thyroid hormone. In the past 10 years, the discovery of disease-
associated polymorphisms in seleno protein genes has drawn attention to the relevance of selenoproteins to health.
Low selenium status has been associated with increased risk of mortality, poor immune function, and cognitive decline.
Higher selenium status or selenium supplementation has antiviral eff ects, is essential for successful male and female
reproduction, and reduces the risk of autoimmune thyroid disease. Prospective studies have generally shown some
benefi t of higher selenium status on the risk of prostate, lung, colorectal, and bladder cancers, but fi ndings from trials
have been mixed, which probably emphasises the fact that supplementation will confer benefi t only if intake of a
nutrient is inadequate. Supplementation of people who already have adequate intake with additional selenium might
increase their risk of type-2 diabetes. The crucial factor that needs to be emphasised with regard to the health eff ects of
selenium is the inextricable U-shaped link with status; whereas additional selenium intake may benefi t people with
low status, those with adequate-to-high status might be aff ected adversely and should not take selenium supplements.
A decade ago, investigators believed that identifi cation of
optimal selenium status would benefi t health. However,
since then, excessive zeal for increasing selenium intake
has at times had adverse consequences—a reminder that
selenium was fi rst known as a toxic element.1 This Review
updates an earlier one2 and discusses present contro-
versies, especially the eff ect of selenium on cancer and
type-2 diabetes, with emphasis on clinically relevant
Appendix pp 1–4 provides additional rele-
Role of selenium: selenoproteins
In human beings, the nutritional functions of selenium are
achieved by 25 selenoproteins that have selenocysteine at
their active centre.3 The insertion of selenocysteine to
form a selenoprotein is specifi ed by the UGA codon in
mRNA under specifi c conditions, but many other inter-
acting factors are necessary.3,4 In low selenium supply, the
synthesis of some selenoproteins (eg, glutathione peroxi-
dase, GPx4) is prioritised over that of others.4 Many
selenoproteins are important enzymes and their impor-
tance to human health is shown by the eff ect of single
nucleotide polymorphisms (SNPs) in selenoprotein genes
on disease risk or mortality (table 1
In contrast to many other micronutrients, the intake of
selenium varies hugely worldwide, ranging from
defi cient (associated with selenium-defi ciency diseases;
appendix p 5) to toxic concentrations that cause garlic
breath, hair and nail loss, disorders of the nervous
system and skin, poor dental health, and paralysis.22
Dietary selenium intake ranges from 7 μg per day to
4990 μg per day, with mean values of 40 μg per day in
Europe and 93 μg per day (in women) to 134 μg per day
(in men) in the USA.1,23,24 Selenium-containing supple-
ments add to these intakes, especially in the USA where
50% of the population takes dietary supplements.24
Selenium status, as measured by plasma or serum
selenium, varies by country and corresponds to intake.2
Intakes are high in Venezuela, Canada, the USA, and
Japan, and much lower in Europe, particularly in
eastern Europe. China has areas of both selenium
defi ciency and excess. Intakes in New Zealand, which
were formerly low, have improved after increased
importation of high-selenium Australian wheat.1
Recom mendations for selenium intake average 60 μg
per day for men and 53 μg per day for women.25
Reasons for the variability in intake relate not only to the
selenium content of the soil on which crops and fodder
are grown, but also to factors that determine the availability
of selenium to the food chain (panel
panel), including selenium
Search strategy and selection criteria
I searched PubMed and the Cochrane Library for publications
from January, 1990, to February, 2011. I used the search
terms “selenium”, “selenoprotein”, and the names of the
individual selenoproteins in combination with the terms
“polymorphism”, “Keshan disease”, “Kashin-Beck disease”,
“mortality”, “immune function”, “immunity”, “regulatory
T cells”, “Tregs”, “virus”, “antiviral-eff ect”, “HIV”, “brain”,
“seizures”, Parkinson’s disease”, “cognitive decline”,
“dementia”, “Alzheimer’s disease”, “fertility”, “miscarriage”,
“preeclampsia”, “preterm birth”, “autoimmune thyroid
disease”, “cardiovascular disease”, “coronary heart disease”,
“type 2 diabetes”, “cancer”, “thyroid cancer”, “colorectal
cancer”, “prostate cancer”, lung cancer”, and “bladder cancer”.
Searches were also based on author name in their known
specialist areas. Further articles were included from personal
knowledge, reference lists, and review articles. Information
presented at the international symposium on selenium in
biology and medicine in 2010, in Kyoto, Japan, was also a
useful source. Many review articles have been included
because they capture a range of useful articles that cannot be
cited individually, in view of the overall limitation on the
number of references. Two additional papers published in
May and June, 2011, were added.
See Online for appendix
www.thelancet.com Vol 379 March 31, 2012 1257
speciation, soil pH and organic-matter content, and the
presence of ions that can complex with selenium.22
Health eff ects of selenium
In at least three prospective studies,32–34 high selenium
status has been associated with low overall mortality.
A non-linear association was noted between selenium
status and all-cause and cancer mortality in 13 887 adult
participants followed up for up to 12 years (until the
end of 2000) in the US Third National Health and
Nutrition Examination Survey.32 Figure 3 shows the
updated follow-up of these participants to the end of
2006. Increasing serum selenium concentrations up to
about 135 μg/L were associated with decreased mortality.
In the 9-year longitudinal Epidemiology of Vascular
Ageing (EVA) study33 of 1389 elderly French individuals
living independently, low plasma selenium at baseline
(mean 87 μg/L) was associated with increased overall and
cancer mortality. In the Baltimore Women’s Health and
Aging Study,34 low serum selenium was a signifi cant
independent predictor of all-cause 5-year mortality in
older women living in the community. By contrast, no
association was noted between total death and baseline
Function or health eff ect Health eff ects associated with polymorphisms (or haplotypes) in the
Glutathione peroxidases (GPxs) Family of antioxidant enzymes: remove hydrogen peroxide, lipid
hydroperoxides, and (GPx4) phospholipid and cholesterol hydroperoxides4
Reduces retroviral virulence by preventing viral mutations;5 defi ciency
Cardiometabolic eff ects: metabolic syndrome, CVD, CAD, blood pressure,
restenosis, coronary-artery calcium score, intimamedia thickness, peripheral
vascular disease, thoracic aortic aneurysm, intracerebral haemorrhage.
Cancer: lung, prostate, bladder, primary liver; Keshan disease, GPx1-198Leu
carriers had low blood selenium and low GPx1 activity; Kashin-Beck disease,
GPx activity lower in GPx1-198Leu carriers; autism
GPx2 (gastrointestinal) Antiapoptotic function in colon crypts; helps to maintain intestinal
Antioxidant in extracellular fl uids; kidney is source of GPx3 in plasma;4,8
thyroid protection from hydrogen peroxide in thyrocytes and
Membrane-associated; present at high concentrations in the testis,
where it is essential for sperm motility and viability10–12
Production of active thyroid hormone T3, and reverse T3 (rT3)13
Production of T3 in the thyroid and peripheral tissues13
T3 production in peripheral tissues13
Ischaemic stroke; diff erentiated thyroid cancer
Adenomatous polyps, colorectal adenocarcinomas; colorectal cancer;
breast cancer survival
Free IGF-1 concentrations, muscle strength, lean body mass
Type-2 diabetes and insulin resistance; osteoarthritis and bone-mineral
density; mental retardation (in iodine defi cient areas)
Dio1 (thyroid, liver, kidney, etc)
Dio2 (brain, pituitary, muscle,
BAT, ear, heart, etc)
Dio3 (cerebral cortex, skin,
placenta, pregnant uterus)
Selenoprotein P (SEPP1)
Production of rT3; prevents overexposure of fetus to T3
Contains 10 selenocysteine residues; major contributor to plasma selenium
and a good indicator of selenium status;14 transports selenium from the
liver via the plasma: brain, testis, and kidney have special receptors;14
has some antioxidant function;14 needed for brain; defi ciency causes
spasticity, abnormal movements, and spontaneous seizures in mice;14,15
essential for male fertility; defi ciency causes infertility with kinked and
hypomotile spermatozoa in mice;14,16 correlated with fasting plasma
glucose;17,18 may serve as heavy-metal (eg, mercury) chelator4
Redox active with a wide range of substrates, notably thioredoxin,
required for DNA synthesis4
Controls activity of transcription factors, cell proliferation, apoptosis;
reduction of expression leads to slower tumour-cell growth4
Indispensable for cardiomyocyte viability4
Anti-infl ammatory,19 located in the ER;4 might protect cells from
ER stress-induced apoptosis;4 linked to glucose metabolism and
Located in the ER; may aff ect glycoprotein folding4
Located in the ER; may regulate calcium mobilisation required for early
muscle development; mutations cause myopathies including
Prostate cancer; aff ects selenium status (plasma selenium and plasma
SEPP) and expression of other selenoproteins; colorectal adenoma,
Thioredoxin reductases (TrxR)
Advanced colorectal adenoma; familial amyotrophic lateral sclerosis
TrxR3 (testis-specifi c)
Selenoprotein S (SEPS1)
Gastric cancer: an SNP in TrxR2 interacts with GPx1 (Pro/Leu) to aff ect risk
Risk of pre-eclampsia; risk of CHD, ischaemic stroke; W:H ratio, BMI;
gastric, colorectal, and rectal cancers
15kDa selenoprotein (SEP15)
Selenoprotein N (SelN)
Prostate cancer mortality; lung cancer; rectal cancer
CVD=cardiovascular disease. CAD=coronary artery disease. IGF-1=insulin-like growth factor 1. BAT=brown adipose tissue. SNP=single nucleotide polymorphism. ER=endoplasmic reticulum. CHD=coronary heart
disease. W:H=waist to hip. BMI=body-mass index. *Appendix pp 6–12 shows table with full list of references.
Table 1: A selection of selenoproteins with known functions relevant to health or with associated health eff ects
www.thelancet.com Vol 379 March 31, 2012
serum selenium (mean 73 μg/L) in a cohort of
1103 Chinese people of average age 57 years, who were
followed up for 15 years.35
Nonetheless, such studies are prone to confounding,
since plasma selenium concentrations are higher in fi t
and well nourished elderly people than in those who
are frail, poorly nourished, and unwell,36 possibly
indicating an increased concentration of infl ammatory
cytokines and lowering of selenium in the acute-phase
response.37,38 Selenium status could have fallen years
before death owing to suboptimal kidney function (the
kidney synthesises plasma GPx3) and subclinical
infl ammatory processes.8,38
Despite evidence from in-vitro and animal studies that
selenium is important to immunity,39–41 evidence in
human beings is scarce. Selenium supplementation,
even in apparently selenium-replete individuals, has
pronounced immunostimulant eff ects, including an
enhancement of proliferation of activated T cells,
increased cytotoxic lymphocyte-mediated tumour cyto-
toxicity, and natural killer cell activity.2,39,42–45 The immune
response is often compromised in elderly people and
during cancer treatment.
Selenium supplementation of elderly volunteers in
Arizona with 400 μg selenium per day (as selenium yeast)
signifi cantly increased total T-cell count by 27% more
than did placebo, largely because of an increase in subsets
of CD4+ T cells and increased cytotoxicity of natural killer
cells.42 Supplementation with 100 μg selenium per day
(as selenium yeast) for 6 months in elderly Belgian
residents in an institution signifi cantly increased the
proliferative response to antigen challenge to the upper
limit of the normal range for adults.2
Supplementation of patients with squamous-cell
carcinoma of the head and neck with 200 μg per day of
selenium (as sodium selenite) during surgery or radiation
led to signifi cantly enhanced cell-mediated immune
responsiveness both during and after therapy.43 By
contrast, immune responsiveness decreased in patients
Only one human study has shown a functional outcome
of selenium supplementation on the immune system.45
UK adults of fairly low selenium status supplemented
with selenium (50 μg and 100 μg per day as sodium
selenite) and challenged with an oral, live, attenuated
poliovirus cleared the virus more rapidly than did those
Panel: Food sources of selenium
Previous reviews have addressed food sources of selenium in detail.1,2,24,26 Figure 1 gives an
indication of the typical selenium content of common food sources.27 The selenium
content of cereals and grains, which are dietary staples, ranges from very low (mean
values of 0·025–0·033 mg/kg dry weight in the UK) to as much as 30 mg/kg in
high-selenium areas of the USA,26 accounting for much of the variation in dietary intake.
Although Brazil nuts are the richest selenium source, they are generally not a commonly
eaten food, and in any case, the content varies greatly, ranging from 0·03 mg/kg to
512 mg/kg fresh weight.26
The contribution of diff erent food groups to total dietary selenium intake in the UK is
known from the Total Diet Study—a continuous market-basket survey in which foods
representing the average UK diet are purchased, prepared, and combined into groups of
similar foods for elemental analysis (fi gure 2).28 There are no equivalent data from other
countries, but estimates of dietary sources of selenium in US adults suggest that the
contribution of bread and cereal sources is somewhat higher than in the UK (37% vs 26%).29
Forms of selenium in foods26
• Selenomethionine: selenium analogue of aminoacid, methionine; found in plant sources
(notably cereals), selenium yeast, and other selenium supplements. It is incorporated
non-specifi cally into body proteins in place of methionine (eg, selenomethionine in
albumin contributes to selenium measured in plasma); supplements containing
selenomethionine therefore seem to have more bioavailable selenium.
• Selenocysteine: selenium analogue of the aminoacid, cysteine; found in animal foods
(from their selenoproteins).
• Selenoneine (2-selenyl-Nα,Nα,Nα-trimethyl-L-histidine): newly discovered as the
major selenium compound in fi sh such as tuna and mackerel; lower concentrations
in squid, tilapia, pig, and chicken.30 It has strong radical-scavenging activity.
• Se-methylselenocysteine and γ-glutamyl-Se-methylselenocysteine: found in plant
sources such as selenium-enriched yeast, garlic, onions, and broccoli. It is metabolised
to methyl selenol, which is thought to have anticancer eff ects.
• Sodium selenite and selenate: components of dietary supplements; selenate
occasionally appears in water supplies. Some selenate is found in fi sh and plant
sources (eg, cabbage).
The way in which these diff erent species are metabolised is reviewed elsewhere,26 as is their bioavailability.31
Figure 1: Typical selenium content of food sources
0 0·25 0·500·751·001·251·50
Typical selenium content of foods (mg/kg)
Organ meats and seafood
Cereals and grains
Most agricultural crops
Milk and dairy products
Fruits and vegetables
Figure 2: Contribution of various food groups to UK selenium intake28
0 10 20 30
Contribution to UK selenium intake (%)
Vegetables and fruit
Milk and dairy
Bread and cereals
Meat and poultry
www.thelancet.com Vol 379 March 31, 2012 1259
That selenium supplementation seems to promote
diff erentiation of CD4+ T cells into T-helper-1 (Th1) rather
than T-helper-2 (Th2) eff ector cells is consistent with
early work showing an apparent benefi t of high selenium
status or selenium supplementation in patients with
allergic asthma.2,39 However, in the largest randomised,
double-blind, placebo-controlled trial done so far,46 no
clinical benefi t of supplementation with 100 μg selenium
per day (as high-selenium yeast) for 24 weeks was
recorded in 197 UK adults with asthma, although most
participants (75%) were taking inhaled steroids, which
may have reduced any potential benefi t.
In a case-control study of 259 HIV-1-infected drug users,47
those with plasma selenium less than 135 μg/L had a
signifi cantly (three-fold) higher risk of developing
mycobacterial disease, three-quarters of which was tuber-
culosis, than did those with higher plasma selenium.
Selenoproteins are essential for activated T-cell function.2
T cells are especially sensitive to oxidative stress, and
selenoprotein-defi cient T cells cannot proliferate in
response to T-cell-receptor stimulation because of their
inability to suppress production of reactive-oxygen species.41
In the few identifi ed individuals with heterozygous defects
in the selenocysteine (Sec) insertion sequence binding
protein 2 (SBP2), the ability to synthesise most seleno-
proteins is lowered.48 These individuals have reduced total
lymphocyte counts and the ability of their T cells to
proliferate after polyclonal stimulation is signifi cantly
reduced, emphasising the importance of selenoproteins in
establishment of an eff ective immune response.48
Interactions between interleukin 2 and its receptor
drive T-cell proliferation.39 Human studies have shown
a corre lation between selenium supplementation and
lymphocyte proliferation, preceded by enhanced expres-
sion of the high-affi nity interleukin-2 receptor.49 Mice
on a high-selenium diet showed increased expression
of both interleukin 2 and the high-affi nity interleukin-2
receptor chain accompanied by enhanced T-cell
signalling and in-vivo CD4+ T-cell responses.39 The high
selenium diet altered the Th1–Th2 balance towards Th1,
leading to increased expression of interferon γ and
CD40 ligand.39 Such an eff ect would benefi t antiviral
immune or antitumour responses that depend on
robust Th1 immunity.39,40,43,45
Eff ects on HIV and other viruses
Selenium defi ciency (serum or plasma selenium
≤85 μg/L) has been associated with decreased survival in
HIV-infected patients.2 However, associations between
low (not necessarily defi cient) serum or plasma selenium,
low CD4+ cell count, and high viral load could also
be attributable to the lowering of blood selenium
concentration by the acute-phase response in individuals
with more advanced HIV-1 infection.50
Two randomised controlled trials have shown
apparent benefi t from selenium supplementation in HIV
infection.51,52 In HIV-positive US adult drug users,
selenium supplementation (200 μg per day) signifi cantly
decreased hospital admissions and the percentage of
admissions due to infection.51 In another trial in
HIV-infected US adults, higher selenium concentration
in serum predicted decreased viral load even after
adjustment for antiretroviral-therapy regimen and
adherence, HIV-disease stage and duration, and hepatitis-C
virus co-infection,52 although some have criticised the
method by which the data were analysed and the relevance
of the diff erences recorded in CD4+ cell count and viral
load.53 By contrast, supplementation with 200 μg per day
selenium (selenomethionine) in a trial in 913 HIV-infected
Tanzanian pregnant women (selenium status unknown)
during the antenatal and post-partum periods in whom
use of antiretroviral therapy was uncommon had no eff ect
on HIV-1 viral load or CD4+ cell count, although it reduced
the risk of mortality in children older than 6 weeks.53
Selenium defi ciency has been linked to the inci dence,
virulence, or disease progression of other viral
infections.2,45 Beck and colleagues54 have shown that
selenium defi ciency in mice, with its associated low or
absent activity of protective GPx1, causes mutations in
RNA viruses that lead to the development of virulent
strains. This fi nding could explain the myocarditis-
inducing mutations in the coxsackie virus that result in
Eff ects on the brain
Selenium is crucial to the brain; during selenium
depletion, brain selenium is maintained at the expense
of other tissues whereas selenium defi ciency causes
irreversible brain injury.14 Selenoprotein P (SEPP1) has a
special role in delivery of selenium to the brain by binding
Figure 3: Adjusted hazard ratios for all-cause mortality by serum selenium
concentration in adult participants of the US Third National Health and
Nutrition Examination Survey followed up for up to 18 years until the end
Shaded area shows 95% CIs. The reference value (hazard ratio 1) was set at the
10th percentile of the serum selenium distribution (105·8 μg/L). The histogram
represents the frequency distribution of serum selenium concentration in the
study sample. Figure constructed by Eliseo Guallar and Yiyi Zhang with updated
data from the Third National Health and Nutrition Examination Survey (original
follow-up to end of 200032).
Hazard ratio (95% CI) for mortality
Serum selenium concentration (μg/L)
For the linked mortality fi le of
the Third National Health and
Nutrition Examination Survey
www.thelancet.com Vol 379 March 31, 2012
to a surface receptor, apoER2—a member of the
lipoprotein-receptor family.14 Mice that cannot synthesise
SEPP1 develop spasticity, abnormal movements, and
spontaneous seizures.14,15 Evidence from studies in
human beings suggests a role for selenium in seizures,
coordination, Parkinson’s disease, and cognitive decline.
Signifi cantly lower serum selenium was noted in
children and adults with epileptic seizures55,56 and in
children who had febrile seizures.57,58 In a few small
studies, selenium supplementation reduced intractable
In the InCHIANTI cohort study of 1012 participants
aged 65 years or older, 59 performance-based assessments
of coordination were signifi cantly worse in participants
with low plasma selenium than in those with higher
concentrations. Furthermore, investigators noted a
signifi cant trend towards increased prevalence of
Parkinson’s disease in the lower selenium quartiles.
SEPP1 has an important neuroprotective role, enhancing
neuronal survival and preventing apoptotic cell death in
response to amyloid-β-induced oxidative challenge.60 Data
from human studies link the risk of Alzheimer’s disease
and dementia to selenium status. In the French EVA
cohort of 1166 people aged 60–70 years,61 a signifi cantly
increased risk of cognitive decline was recorded over
4 years in participants with low plasma selenium at
baseline. Furthermore, cognitive decline was signifi cantly
associated with the magnitude of plasma selenium
decrease over 9 years.62 In a cross-sectional survey of
2000 rural Chinese adults aged 65 years or older, low nail
selenium concentration was signifi cantly associated with
low cognitive scores in four of fi ve tests, with a dose-
response eff ect across selenium quintiles.63
However, in the context of cognitive function in elderly
people, low plasma selenium could partly indicate a low
production of plasma GPx3 by an ineffi cient kidney8 or
low selenoprotein synthesis resulting from the action of
infl ammatory cytokines (in the acute-phase response).37
Failing kidneys also leak homocysteine—a known risk
factor for dementia—into the bloodstream.64 Whether
nail selenium would be similarly aff ected is unknown.
Despite earlier evidence for a benefi t of selenium
supplementation on mood,2 a large randomised, placebo-
controlled trial in individuals aged 60–74 years showed
no evidence that 6 months of selenium supplementation
(100 μg, 200 μg, or 300 μg selenium per day as high-
selenium yeast) benefi ted mood or quality of life.65
Fertility and reproduction
In men, GPx4 is found in the mitochondria that make up
the midpiece sheath of the sperm tail. In the early phase
of spermatogenesis, GPx4, as a peroxidase, protects
spermatozoa by its antioxidant function, whereas in the
later phase, it forms cross-links with midpiece proteins
to become a structural component of the mitochondrial
sheath surrounding the fl agellum, which is essential for
In a Japanese study, 10% of infertile men and 35% of
those with oligoasthenozoospermia whose spermatozoa
showed extensive lipid peroxidation had GPx4-defective
spermatozoa.11 In these men, spermatozoal GPx4 expres-
sion and sperm motility were lost, and spermatozoa in
the ejaculate decreased. The lower GPx4 expression was
not due to selenium defi ciency because selenium intake
in Japan is high1 and the level of expression in blood
leucocytes did not diff er from that of fertile men.11
Analysis of sperm samples in an Italian study showed
that GPx4 protein content was signifi cantly lower in
75 infertile men than in 37 controls (93 vs 188 units/mg
sperm protein) and correlated with viability (r=0·35),
morphological integrity (r=0·44), and forward motility
The testis has a special receptor (apoER2) to take up
SEPP1, the other seleonoprotein that it requires.14 The
selenium intake required for optimal activity and
concentration of GPx4 and SEPP1 is around 75 μg per day.66
In a randomised trial, selenium supplementation (100 μg
per day) of subfertile men with low selenium intake
signifi cantly increased sperm motility and enabled 11% of
the men to achieve paternity, compared with none in the
placebo group.2 However, high selenium intake (about
300 μg per day) was shown to decrease sperm motility.67
Signifi cantly lower selenium status has been noted
in women who had either fi rst-trimester or recurrent
miscarriages than in those who did not miscarry,
although this fi nding is not consistent.2,68 Selenium status
usually falls in pregnancy, partly because of plasma
volume expansion,68 but excessive infl ammation—a
probable feature of miscarriage—also
Both selenium intake and status have been linked to pre-
eclampsia.68,69 Signifi cantly lower concentrations of plasma
and toenail selenium, plasma, placental GPx, and placen-
tal thioredoxin reductase have been measured in pre-
eclamptic women than in matched healthy controls.69–71
In a UK study,69 median selenium concentration in
toenail clippings (which are largely laid down before preg-
nancy) from women with pre-eclampsia of mean gesta-
tional age 34 weeks was signifi cantly lower than in the
controls (p<0·001). Women in the lowest tertile of toenail
selenium were signifi cantly (4·4 times) more likely to have
pre-eclampsia than were those in the higher tertiles.69
Selenoproteins could counteract features of pre-
eclampsia by reducing oxidative stress, endoplasmic-
reticulum stress, and infl ammation, protecting the
endothelium, controlling eicosanoid production, and
regulating vascular tone.68 Since systemic infl ammation
is a major feature of pre-eclampsia, selenoprotein
S (SEPS1), which is involved in management of the stress
response in the endoplasmic reticulum and in infl am-
mation control,72 may be crucial. A retrospective study in
a large Norwegian case-control cohort72 showed that
women with pre-eclampsia (n=1139) were signifi cantly
more likely than were controls (n=2269) to carry the
www.thelancet.com Vol 379 March 31, 2012 1261
A allele of the SEPS1 g.-105G>A polymorphism,
implicating SEPS1 in the risk of pre-eclampsia.
A cross-sectional study in the Netherlands of
1129 pregnant women showed that those who delivered
preterm had signifi cantly lower serum selenium at
12 weeks’ gestation than did those who delivered at term.73
Even after adjustment for the occurrence of pre-eclampsia,
which is associated both with selenium status69 and with
preterm birth, women in the lowest quartile of serum
selenium had a signifi cantly (two-fold) greater risk of
preterm birth did than the rest. However, whether low
selenium status was a cause or an outcome (eg, of the
associated increased infl ammation) is unknown.73
Thyroid function and autoimmune thyroid disease
The thyroid gland has the highest selenium concentra-
tion of all tissues.13 Selenium has various roles in
the thyroid: the selenium-dependent iodothyronine
deiodinases produce active thyroid hormone, tri-
iodothyronine (T3), from its inactive precursor, thyroxine
(T4).13 Nonetheless, no evidence of an eff ect on thyroid
function or on the ratio of free or total T4 to T3 was
shown in a randomised controlled trial of selenium
supplementation in 368 apparently euthyroid, UK elderly
adults with low-to-moderate selenium status.74
Selenium, in the form of GPx3, protects thyroid cells
from the hydrogen peroxide that is generated there to be
used by thyroid peroxidase in the synthesis of T3 and
T4 from iodide and thyroglobulin.13 This function is
consistent with the inverse association detected between
selenium status and thyroid volume, thyroid tissue
damage, and goitre in French women,75 and the positive
association between the incidence of thyroid cancer and
low prediagnostic serum-selenium concentration in
Norway.76 Moreover, several studies have shown that
selenium supplementation (80 μg or 200 μg per day as
sodium selenite or selenomethionine) is eff ective against
Hashimoto’s thyroiditis, the most common form of
autoimmune thyroid disease, characterised by the presence
of complement-fi xing autoantibodies to thyroid peroxi-
dase.77,78 A systematic review and meta-analysis showed
that selenium supplementation signifi cantly lowered
thyroid peroxidase autoantibody titre at 3 months.77
Pregnant women with autoimmune thyroiditis
(thyroid-peroxidase-antibody positive) are prone to
develop post-partum thyroid dysfunction and permanent
hypothyroidism. When women were given 200 μg per
day selenomethionine, thyroid infl ammatory activity
fell, and post-partum thyroid disease and permanent
hypothyroidism were signifi cantly reduced.79
Selenium is also eff ective in autoimmune hyper-
thyroidism—ie, Graves’ disease. In a randomised
controlled trial of 100 μg sodium selenite twice daily, or
pentoxifylline (600 mg), or placebo for 6 months, selenium
treatment alone was signifi cantly associated with an
improved quality of life, less eye involvement, and slower
progression of Graves’ orbitopathy.80
Selenium administration has been associated with
benefi t in systemic infl ammatory response syndrome
and sepsis.81 Patients with systemic infl ammatory
response syndrome or septic shock had a 40% decrease
in plasma selenium and a 70% decrease in plasma
SEPP1 (important because SEPP1 is thought to provide
endothelial protection).82,83 Two meta-analyses have
shown that mortality tended to decrease when such
patients were infused with high-dose sodium selenite.82
However, treatment needs to be initiated by bolus (1 mg),
since studies in which continuous administration was
used have shown no eff ect on mortality.82
Potential cardiovascular benefi ts of selenium are
supported by evidence that selenoproteins prevent
oxidative modifi cation of lipids, inhibit platelet
aggregation, and reduce infl ammation2,68,84 in addition to
the many cardiometabolic eff ects that have been linked
to polymorphisms in GPx1, GPx3, Dio2, and SEPS1
(table 1). However, randomised trials of selenium-
containing supplements have not shown a signifi cant
protective eff ect on cardiovascular disease or mortality
endpoints,85–87 although a meta-analysis of 25 observa-
tional studies showed a signifi cant inverse association
between selenium status and risk of coronary heart
disease, particularly in populations with low selenium
intake or status.85 By contrast, no associations were
recorded between toenail-selenium concentration and
measures of subclinical
intimamedia thickness and coronary-artery calcium
score—in a study of 3112 young American adults.88
Several cross-sectional studies have shown an
association between high selenium status and raised
plasma cholesterol.84 In the UK PRECISE Pilot ran-
domised trial of 501 elderly people with low selenium
status,84 total serum cholesterol
chol esterol were signifi cantly lowered after 6 months
supple mentation with 100 μg and 200 μg selenium per
day (as high-selenium yeast) but not with 300 μg selenium
per day, even though this dose raised HDL cholesterol
signifi cantly. With increasing selenium dose, the ratio
of total cholesterol to HDL cholesterol signifi cantly
decreased, suggesting a potentially benefi cial eff ect of
supplementation on cardiovascular risk, at least in that
population. In two other small trials, no signifi cant
diff erence between treatment groups was reported.89,90
Selenium status of the populations studied might
account for diff erences in results. Beyond a specifi c
plasma selenium concentration (at which relevant
selenoproteins might be optimised), there may be no
further advantage of high selenium status in reduction of
cardiovascular mortality, and there is some evidence of a
U-shaped association.32 For example, at the baseline
selenium status of participants in the UK PRECISE trial,
GPx1 activity would only have been optimised in half the
www.thelancet.com Vol 379 March 31, 2012
volunteers, so that potential benefi ts of selenium
supplementation could be detected.84 By contrast, the
negative results reported in American trials86,87 were in
populations in whom most selenoproteins would already
have been optimised at baseline66,91,92 so that if the benefi cial
eff ect of selenium is dependent on raising selenoprotein
concentrations, no eff ect of supple mentation would have
been apparent. In support of a potentially benefi cial eff ect
of GPx1 activity on cardiovascular risk, baseline erythrocyte
GPx1 activity was a strong predictor of the risk of a
subsequent cardiovascular event during 4·7 years of
follow-up in a cohort of 636 patients with suspected
coronary artery disease whose selenium status was low
(mean plasma selenium 74 μg/L).93
Prospective studies have provided some evidence for a
benefi cial eff ect of selenium on the risk of lung,94 bladder,95
colorectal,96 liver,97 oesophageal,35 gastric-cardia,35 thyroid,76
and prostate cancers.21,96,98–100 Table 2
of meta-analyses of selenium supplementation for lung,
bladder, and prostate cancers.
Two subsequent studies did not show signifi cant
associations between selenium intake101 or status102 and
lung cancer risk, despite a signifi cant inverse trend in
men in an earlier large case-control study based on
dietary-selenium intake.103 No subsequent study has
shown a signifi cant benefi cial or detrimental eff ect on
prostate cancer risk,104–107 except for African-American
men of the US Multiethnic Cohort whose risk was 41%
Table 2 summarises results
lower in the top rather than the bottom tertile of serum
selenium.108 However, in the EPIC-Heidelberg cohort,
a signifi cantly decreased risk, especially of high-grade
disease, was noted in the third (although not the fourth,
≥95·0 μg/L) quartile of serum selenium concentration
compared with the fi rst quartile.107
A review96 of all but the three most recent prostate
cancer studies106–108 shows that more signifi cant protective
associations are consistently detected between selenium
and risk of advanced, rather than localised or low-grade,
prostate cancer, and that the strongest associations are
Interventions with selenium as a single nutrient
are scarce. A systematic review and meta-analysis109 of
antioxidant supplements for the prevention of gastro-
intestinal cancers pooled data from three trials in
Qidong, China, where 15% of the population is
hepatitis B positive; although supplementation signifi -
cantly reduced the incidence of hepatocellular carci-
noma by 50%, the quality of reporting of the trials has
The Nutritional Prevention of Cancer (NPC) trial
recruited 1312 volunteers with a previous history of
non-melanoma skin cancer from southeastern USA.110
Treatment with 200 μg selenium per day (as selenium
yeast) for a mean of 4·5 years had no eff ect on the primary
outcome of non-melanoma skin cancer but led to a
signifi cant reduction in cancer mortality (50%) and in
the incidence of total (37%), prostate (67%), colorectal
(58%), and lung (46%) cancers after a follow-up of
6·4 years. However, in later analyses, only the reduction
in incidence of total (25%) and prostate (52%) cancers
remained signifi cant,111 except for those in the bottom
tertile of plasma selenium (<106 μg/L) at baseline.112
Those in the highest tertile (>122–123 μg/L) had
non-signifi cant increases in the risk of total (20%) and
prostate cancers (14%). Furthermore, a signifi cantly
increased risk of squamous-cell carcinoma was recorded
in participants with baseline plasma selenium in the top
two tertiles, although the risk was non-signifi cantly
decreased in the bottom tertile (plasma selenium
>106 μg/L).111,113,114 Clearly there is no protective eff ect of
supplementation, and indeed the possibility of increased
risk, in participants with baseline plasma selenium
greater than 122 μg/L.
The Selenium and Vitamin E Cancer Trial (SELECT),
which investigated the eff ect of selenium and vitamin E
on prostate cancer risk in 35 533 American men, showed
that administration of 200 μg selenium per day (as
selenomethionine) to men of median baseline serum
selenium 136 μg/L did not reduce the risk of localised
prostate cancer after a median follow-up of 5·5 years.87
However, the trial included almost no participants within
the range of selenium status (<106 μg/L in plasma)
that had previously shown benefi t from selenium
supplementation on prostate cancer risk.111,113 In men in
SELECT, the IQR of baseline serum selenium was
RR/OR (95% CI) Type of study
16 studies 0·74 (0·57 to 0·97)* Cohort,
Higher vs lower selenium
exposure (serum, toenails,
Zhuo et al (2004)94
7 studies 0·61 (0·42 to 0·87)*Cohort, nested
Highest vs lowest selenium
status (serum, toenails)
Amaral et al (2010)95
5 studies0·74 (0·61 to 1·39) Case-control
Any vs lowest selenium
Moderate vs lowest
Any vs lowest selenium
Moderate vs lowest
Overall pooled standardised
mean diff erence (serum,
plasma, toenail selenium)
Etminan et al (2005)98
5 studies 0·74 (0·39 to 1·39)Case-control Etminan et al (2005)98
11 studies 0·72 (0·61 to 0·84)*CohortEtminan et al (2005)98
11 studies 0·74 (0·61 to 0·90)*Cohort Etminan et al (2005)98
20 studies –0·23 (–0·40 to –0·05)* Cohort, nested
Brinkman et al (2006)99
RR=relative risk. OR=odds ratio. *Signifi cant fi nding. †Any intake: average between the fi rst and fourth quintile (plasma,
serum, toenail concentrations, or intake) or the fi rst and third quartile. Moderate intake: average between the second
and fourth quintile or the second and third quartile. The lowest level (fi rst quartile) was used as the reference group.
Table 2: Meta-analyses of prospective studies of selenium and cancer risk, by tissue type
www.thelancet.com Vol 379 March 31, 2012 1263
122·4–151·8 μg/L, which, according to the NPC trial,
put them into the category of non-signifi cant increased
risk from selenium supple mentation. Indeed, the
investi gators recorded some potential indications of
toxic eff ects in terms of alopecia and dermatitis in the
selenium group.87 Additionally, participants in SELECT
were given selenium as seleno methionine rather than
selenium yeast, in which 30–40% of the selenium is not
Results of SELECT do not explain the eff ect of selenium
on: (1) risk of advanced disease, for which various studies
have suggested a greater eff ect,96,100,115 since only 1% of
cases were non-localised; (2) prostate cancer mortality,
since only one participant died from prostate cancer;87
(3) current smokers, for whom the strongest protective
eff ects have been detected,96 since they represented only
7·5% of the SELECT population; or (4) men of low
selenium status, since very few were included in the
study.116 Clearly, whereas at least a third of men in the
NPC trial did not have optimal SEPP1 or even GPx
concentration or activity before supplementation, this is
unlikely to be true of men in SELECT, since most would
have had maximal selenoprotein activities or con-
centrations at baseline.116
Various factors could explain the disparate fi ndings
between studies and especially between the trials. First,
selenium might be more important in prevention of
prostate cancer progression, hence its stronger eff ect
against advanced than primary disease.23,96 Second,
selenium might show an eff ect on risk only over a
particular range of status, neither too low nor too high—
eg, when the selenium concentrations in the group under
investigation range from below, to above, the
concentration needed to optimise the activity of
selenoenzymes such as GPx and SEPP1. This scenario
would not have been the case in SELECT (too high) nor
probably in the European Prospective Investigation into
Cancer and Nutrition (EPIC) cohort (too low).21,87,106,117
Indeed for selenium, as for many nutrients, several
human studies have provided evidence of a U-shaped
relationship between intake or status and protection
from cancer.1,96,111 Third, the interaction between selenium
intake or status and genetic background could be
important. SNPs in selenoprotein genes can aff ect the
effi ciency with which a selenoprotein is synthesised, its
activity and concentration in plasma, and risk of disease.21
Examples include a GPx1 SNP that aff ects the risk of
prostate, lung, breast, and bladder cancers; SNPs in
SEPP1, GPx4, and SEPS1 that aff ect the risk of colorectal
cancer; a SNP in GPx3 linked with thyroid cancer;
variants in SEP15 that aff ect the risk of lung and rectal
cancers and survival from prostate cancer; and a GPx4
SNP that aff ects breast cancer survival (table 1). Moreover,
these selenoprotein SNPs (and SNPs in related pathways)
can interact with selenium status such that individuals
with specifi c genetic variants might benefi t more from
additional selenium than might others.115,118
Evidence linking selenium to glucose metabolism is
confl icting.119 High selenium status was associated with
reduced diabetes prevalence in three case-control
studies,120–122 while in the prospective EVA study,123 high
plasma selenium correlated with a decreased risk of
onset of hyperglycaemia during a 9-year follow-up period
in male participants.
By contrast, high serum selenium concentration was
associated with an increased prevalence of diabetes in
the large US National Health and Nutrition Examination
Surveys.124,125 Similarly, in the French SUVIMAX trial
population,126 investigators recorded positive correlations
between plasma selenium and fasting plasma glucose
both at baseline and follow-up.
Results from randomised trials in which type-2 diabetes
was a secondary outcome also vary. In SELECT,
supplementation of 35 533 American men with 200 μg
selenium per day as selenomethionine had no eff ect on
risk of type-2 diabetes after a median follow-up of
5·5 years.87 By contrast, a post-hoc analysis of the NPC
trial in 1312 participants in southeastern USA showed a
signifi cantly increased risk of type-2 diabetes in those
supplemented with selenium (200 μg per day as selenium
yeast) and followed up for a mean of 7·7 years.127 An
exposure-response gradient was noted across tertiles of
baseline plasma selenium concentration, and the
increased risk was driven by those in the highest tertile
(>121·6 μg/L) whose risk was signifi cantly increased with
How might these discrepant fi ndings be interpreted?
The lower selenium status measured in men with
type-2 diabetes in the case-control studies might be an
eff ect of the disease, or its associated infl ammation, on
selenium status. For example, a systemic infl ammatory
response produces cytokines that inhibit the expression
of SEPP1 and will reduce plasma selenium.37,38,83 Similar
eff ects associated with preclinical disease are unlikely
to explain fully the decreased risk of onset of
hyperglycaemia in male participants of the EVA study.123
Insulin resistance can be triggered by oxidative stress
and ameliorated by antioxidant treatment.128 The plasma
selenium of men in the EVA cohort ranged from below
to above the concentration needed for optimum
antioxidant GPx activity, suggesting that participants in
the top tertile were better protected from oxidative
stress, and hence the development of insulin resistance,
than were those in the bottom tertile (median plasma
selenium 104 μg/L vs 71 μg/L).123 Why the same eff ect
was not noted in women is unclear.
The increased risk of type-2 diabetes with high
selenium intake or supplementation might be explained
by the eff ect of high selenium on insulin signalling.
Binding of insulin to its receptor initiates the insulin-
signalling cascade, which is accompanied by a burst of
hydrogen peroxide that acts as a second messenger.119
High activity of GPx1, which removes hydrogen
www.thelancet.com Vol 379 March 31, 2012
peroxide, might thus interfere with insulin signalling.
For example, transgenic mice overexpressing GPx1
developed insulin resistance, hyperglycaemia, hyper-
insulinaemia, and obesity,129 and a strong correlation
was noted between increased erythrocyte GPx1 activity
and mild insulin resistance in pregnant women.130 By
contrast, knockout of GPx1 improved insulin-induced
glucose uptake and insulin resistance in mice.119
However, GPx1 cannot be the only relevant selenoprotein
because plasma GPx activity is maximised well below
the selenium doses associated with increased risk of
type-2 diabetes, while a polymorphism in Dio2 is known
to aff ect the risk of insulin resistance and type-2 diabetes
(table 1). Indeed, individuals with a reduced ability to
synthesise many selenoproteins (owing to heterozygous
defects in SBP2) had enhanced systemic and cellular
Another selenoprotein implicated in diabetes risk
is SEPP1, which requires a higher selenium intake
than does GPx1 to achieve maximum plasma concen-
tration.66 SEPP1 functions as a negative insulin modulator:
it inhibits the insulin-induced burst of reactive oxygen
species in vitro and further contributes to insulin
resistance by inactivating
kinase—a positive regulator of insulin synthesis and
secretion in pancreatic islet β cells.17 Clinical studies in
Japanese adults have shown that serum SEPP1
concentration is signifi cantly correlated with glycated
haemoglobin A1c and fasting plasma glucose, and is raised
in people with type-2 diabetes.17 Furthermore, serum
SEPP1 concentrations were signifi cantly higher in Korean
patients with type-2 diabetes or pre-diabetes than in those
with normal glucose tolerance and decreased in a stepwise
manner.18 SEPP1 was higher in overweight and obese
participants than in lean participants,18 indicating that
dysregulated carbohydrate metabolism could be driving
the increase in SEPP1 through the action of PGC1α—a
transcription factor co-activator that is a key regulator of
both hepatic gluconeogenesis and SEPP1 biosynthesis.119
Why did selenium supplementation cause an increased
risk of type-2 diabetes in participants in the NPC trial
but not in SELECT? The small NPC trial was at much
greater risk of chance fi ndings than was the much larger
SELECT. Alternatively, the higher baseline selenium
status of men in SELECT (median serum selenium
136 μg/L vs mean plasma selenium 114 μg/L87,127) had
already caused selenoprotein expression or activity to
plateau or pass a threshold of risk before supplementation.
In support of this assertion, the number of cases in the
placebo group per 1000 person-years in SELECT was
higher than in the NPC trial (14·1 vs 8·4).
In animal models, low levels of expression of GPx1 and
other stress-related (regulated by the amount of selenium
in the diet) selenoproteins are as damaging as high levels
of expression with respect to insulin resistance and
hyperglycaemia.131 Hence a U-shaped association between
selenoproteins and type-2 diabetes risk might explain
some of the apparently contradictory fi ndings.
Selenium status in relation to health eff ects
Selenium status varies widely in diff erent parts of the
world, in line with selenium intakes.1,2 The distribution
of plasma selenium in the UK132 and that in the USA133
(fi gure 4) shows the diff erence in status between Europe
(represented by UK data) and North America (repre-
sented by US data). The dotted vertical line on fi gure 4
(at 122 μg/L) represents the concentration of baseline
plasma selenium that marked a change from negative
to positive in the risk of cancer, non-melanoma skin
cancer, and type-2 diabetes with selenium supple-
mentation in the NPC trial.110,111,114,127 The impli cations are
clear: people whose serum or plasma selenium
concentration is already 122 μg/L or higher—a large
proportion of the US population—should not supple-
ment with selenium.
The converse, however, is also true: there are various
health benefi ts, and more importantly, no extra risk, for
people with serum or plasma selenium concentrations
less than 122 μg/L associated with raising their selenium
status, perhaps to 130–150 μg/L—a concentration asso-
ciated with minimal mortality (fi gure 3).32
Conclusions and suggestions for future research
The eff ects of selenium on human health are multiple
and complex,1 necessitating further research to optimise
the benefi ts and reduce the risks of this potent trace
mineral. Trials should be undertaken only in populations
Figure 4: Distribution of serum or plasma selenium in adults aged
40–64 years in the UK 2001 NDNS population132 (representing status in
Europe), and the US 2003–04 NHANES population133 (representing status in
The height of the histogram bars represents the weighted percentage of the
population having the corresponding range of serum or plasma selenium. The
dotted black vertical line at 122 μg/L represents the concentration of baseline
plasma selenium that delineates a change in risk of cancer, non-melanoma skin
cancer, and type-2 diabetes from lower to higher on supplementation with
200 μg selenium per day in the Nutritional Prevention of Cancer trial.110,111,114,127 The
blue dotted lines show the range of serum selenium concentration associated
with minimum mortality in the third NHANES population.32 Histograms
constructed by Eliseo Guallar and Yiyi Zhang. NDNS=National Diet and Nutrition
Survey. NHANES=National Health and Nutrition Examination Survey.
0 40 6080 100120140160 180200
Serum or plasma selenium (μg/L)
UK NDNSUS NHANES
www.thelancet.com Vol 379 March 31, 2012 1265
of low or relatively low selenium status. Furthermore,
since polymorphisms in selenoproteins aff ect both
selenium status and disease risk or prognosis (table 1),
future studies must genotype participants. Further work
aimed at understanding the potential links between
selenoproteins and the highly prevalent condition of
type-2 diabetes should also be a priority.
The crucial factor that needs to be emphasised is the
inextricable U-shaped link with selenium status:
additional selenium intake (eg, from food fortifi cation
or supplements) may well benefi t people with low
status. However, people of adequate or high status
could be aff ected adversely and should not take
selenium supplements. This observation is a particular
case of the general principle recognised by Paracelsus
as long ago as 1567.
Confl icts of interest
I have received a contribution towards funding of a PhD studentship
from Wassen International and funding from Pharma Nord for the
measurement of plasma selenium in a cohort of 1200 pregnant women.
I thank Holger Steinbrenner for helpful discussions on selenium
and type-2 diabetes. Special thanks are due to Eliseo Guallar and
Yiyi Zhang for constructing fi gure 3 and the histograms in fi gure 4.
I also acknowledge the work of the many scientists whose data could
only be cited within review articles owing to restrictions on number
1 Rayman MP. Food-chain selenium and human health: emphasis on
intake. Br J Nutr 2008; 100: 254–68.
2 Rayman MP. The importance of selenium to human health. Lancet
2000; 356: 233–41.
3 Kryukov GV, Castellano S, Novoselov SV, et al. Characterization of
mammalian selenoproteomes. Science 2003; 300: 1439–43.
4 Reeves MA, Hoff mann PR. The human selenoproteome: recent
insights into functions and regulation. Cell Mol Life Sci 2009;
5 Beck MA, Levander OA, Handy J. Selenium defi ciency and viral
infection. J Nutr 2003; 133 (suppl 1): 1463S–67S.
6 Lei C, Niu X, Wei J, Zhu J, Zhu Y. Interaction of glutathione
peroxidase-1 and selenium in endemic dilated cardiomyopathy.
Clin Chim Acta 2009; 399: 102–08.
7 Florian S, Krehl S, Loewinger M, et al. Loss of GPx2 increases
apoptosis, mitosis, and GPx1 expression in the intestine of mice.
Free Radic Biol Med 2010; 49: 1694–702.
8 Schiavon R, Guidi GC, Biasioli S, De Fanti E, Targa L. Plasma
glutathione peroxidase activity as an index of renal function.
Eur J Clin Chem Clin Biochem 1994; 32: 759–65.
9 Schmutzler C, Mentrup B, Schomburg L, Hoang-Vu C, Herzog V,
Köhrle J. Selenoproteins of the thyroid gland: expression,
localization and possible function of glutathione peroxidase 3.
Biol Chem 2007; 388: 1053–59.
10 Ursini F, Heim S, Kiess M, et al. Dual function of the selenoprotein
PHGPx during sperm maturation. Science 1999; 285: 1393–96.
11 Imai H, Suzuki K, Ishizaka K, et al. Failure of the expression
of phospholipid hydroperoxide glutathione peroxidase in the
spermatozoa of human infertile males. Biol Reprod 2001;
12 Foresta C, Flohé L, Garolla A, Roveri A, Ursini F, Maiorino M.
Male fertility is linked to the selenoprotein phospholipid
hydroperoxide glutathione peroxidase. Biol Reprod 2002;
13 Schomburg L, Köhrle J. On the importance of selenium and iodine
metabolism for thyroid hormone biosynthesis and human health.
Mol Nutr Food Res 2008; 52: 1235–46.
14 Burk RF, Hill KE. Selenoprotein P-expression, functions, and roles
in mammals. Biochim Biophys Acta 2009; 1790: 1441–47.
15 Schweizer U, Bräuer AU, Köhrle J, Nitsch R, Savaskan NE.
Selenium and brain function: a poorly recognized liaison.
Brain Res Brain Res Rev 2004; 45: 164–78.
16 Olson GE, Winfrey VP, Nagdas SK, Hill KE, Burk RF.
Selenoprotein P is required for mouse sperm development.
Biol Reprod 2005; 73: 201–11.
Misu H, Takamura T, Takayama H, et al. A liver-derived secretory
protein, selenoprotein P, causes insulin resistance. Cell Metab 2010;
18 Yang SJ, Hwang SY, Choi HY, et al. Serum selenoprotein P levels in
patients with type 2 diabetes and prediabetes: implications for
insulin resistance, infl ammation, and atherosclerosis.
J Clin Endocrinol Metab 2011; 96: E1325–29.
19 Curran JE, Jowett JB, Elliott KS, et al. Genetic variation in
selenoprotein S infl uences infl ammatory response. Nat Genet 2005;
20 Gao Y, Walder K, Sunderland T, et al. Elevation in Tanis expression
alters glucose metabolism and insulin sensitivity in H4IIE cells.
Diabetes 2003; 52: 929–34.
21 Rayman MP. Selenoproteins and human health: insights from
epidemiological data. Biochim Biophys Acta 2009; 1790: 1533–40.
22 Johnson CC, Fordyce FM, Rayman MP. Symposium on
‘Geographical and geological infl uences on nutrition’: factors
controlling the distribution of selenium in the environment and
their impact on health and nutrition. Proc Nutr Soc 2010; 69: 119–32.
23 Rayman MP. Selenium in cancer prevention: a review of the
evidence and mechanism of action. Proc Nutr Soc 2005; 64: 527–42.
24 Fairweather-Tait SJ, Bao Y, Broadley MR, et al. Selenium in human
health and disease. Antioxid Redox Signal 2011; 14: 1337–83.
25 Rayman MP. The use of high-selenium yeast to raise selenium
status: how does it measure up? Br J Nutr 2004; 92: 557–73.
26 Rayman MP, Infante HG, Sargent M. Food-chain selenium and
human health: spotlight on speciation. Br J Nutr 2008; 100: 238–53.
27 WHO. Selenium. A report of the International Programme on
Chemical Safety. Environmental Health Criteria number 58.
Geneva: World Health Organization, 1987.
28 Food Standards Agency. Survey on measurement of the
concentrations of metals and other elements from the 2006 UK
total diet study. 2009. http://www.food.gov.uk/multimedia/pdfs/
fsis0109metals.pdf (accessed Feb 14, 2011).
29 Cotton PA, Subar AF, Friday JE, Cook A. Dietary sources of
nutrients among US adults, 1994 to 1996. J Am Diet Assoc 2004;
30 Yamashita Y, Yamashita M. Identifi cation of a novel selenium-
containing compound, selenoneine, as the predominant chemical
form of organic selenium in the blood of bluefi n tuna. J Biol Chem
2010; 285: 18134–38.
31 Fairweather-Tait SJ, Collings R, Hurst R. Selenium bioavailability:
current knowledge and future research requirements.
Am J Clin Nutr 2010; 91: 1484S–91S.
32 Bleys J, Navas-Acien A, Guallar E. Serum selenium levels and
all-cause, cancer, and cardiovascular mortality among US adults.
Arch Intern Med 2008; 168: 404–10.
33 Akbaraly NT, Arnaud J, Hininger-Favier I, Gourlet V, Roussel AM,
Berr C. Selenium and mortality in the elderly: results from the EVA
study. Clin Chem 2005; 51: 2117–23.
34 Ray AL, Semba RD, Walston J, et al. Low serum selenium and total
carotenoids predict mortality among older women living in the
community: the women’s health and aging studies. J Nutr 2006;
35 Wei WQ, Abnet CC, Qiao YL, et al. Prospective study of serum
selenium concentrations and esophageal and gastric cardia cancer,
heart disease, stroke, and total death. Am J Clin Nutr 2004; 79: 80–85.
36 Bates CJ, Thane CW, Prentice A, Delves HT. Selenium status and
its correlates in a British national diet and nutrition survey: people
aged 65 years and over. J Trace Elem Med Biol 2002; 6: 1–8.
37 Hesse-Bahr K, Dreher I, Kohrle J. The infl uence of the cytokines
Il-1beta and INFgamma on the expression of selenoproteins in the
human hepatocarcinoma cell line HepG2. Biofactors 2000; 11: 83–85.
38 Nichol C, Herdman J, Sattar N, et al. Changes in the concentrations
of plasma selenium and selenoproteins after minor elective surgery:
further evidence for a negative acute phase response? Clin Chem
1998; 44: 1764–66.
www.thelancet.com Vol 379 March 31, 2012
39 Hoff mann FW, Hashimoto AC, Shafer LA, Dow S, Berry MJ,
Hoff mann PR. Dietary selenium modulates activation and
diff erentiation of CD4+ T cells in mice through a mechanism
involving cellular free thiols. J Nutr 2010; 140: 1155–61.
40 Hoff mann PR. Mechanisms by which selenium infl uences immune
responses. Arch Immunol Ther Exp (Warsz) 2007; 55: 289–97.
41 Carlson BA, Yoo MH, Shrimali RK, et al. Role of selenium-
containing proteins in T-cell and macrophage function.
Proc Nutr Soc 2010; 69: 300–10.
42 Wood SM, Beckham C, Yosioka A, Darban H, Watson RR.
β-Carotene and selenium supplementation enhances immune
response in aged humans. Integr Med 2000; 2: 85–92.
43 Kiremidjian-Schumacher L, Roy M, Glickman R, et al. Selenium
and immunocompetence in patients with head and neck cancer.
Biol Trace Elem Res 2000; 73: 97–111.
44 Hawkes WC, Kelley DS, Taylor PC. The eff ects of dietary selenium
on the immune system in healthy men. Biol Trace Elem Res 2001;
45 Broome CS, McArdle F, Kyle JA, et al. An increase in selenium
intake improves immune function and poliovirus handling in
adults with marginal selenium status. Am J Clin Nutr 2004;
46 Shaheen SO, Newson RB, Rayman MP, et al. Randomised, double
blind, placebo-controlled trial of selenium supplementation in adult
asthma. Thorax 2007; 62: 483–90.
47 Shor-Posner G, Miguez MJ, Pineda LM, et al. Impact of selenium
status on the pathogenesis of mycobacterial disease in
HIV-1-infected drug users during the era of highly active
antiretroviral therapy. J Acquir Immune Defi c Syndr 2002; 29: 169–73.
48 Schoenmakers E, Agostini M, Mitchell C, et al. Mutations in the
selenocysteine insertion sequence-binding protein 2 gene lead to a
multisystem selenoprotein defi ciency disorder in humans.
J Clin Invest 2010; 120: 4220–35.
49 Roy M, Kiremidjian-Schumacher L, Wishe HI, Cohen MW,
Stotzky G. Supplementation with selenium and human immune
cell functions. I. Eff ect on lymphocyte proliferation and interleukin
2 receptor expression. Biol Trace Elem Res 1994; 41: 103–14.
50 Drain PK, Baeten JM, Overbaugh J, et al. Low serum albumin and
the acute phase response predict low serum selenium in HIV-1
infected women. BMC Infect Dis 2006; 6: 85.
51 Burbano X, Miguez-Burbano MJ, McCollister K, et al. Impact of a
selenium chemoprevention clinical trial on hospital admissions of
HIV-infected participants. HIV Clin Trials 2002; 3: 483–91.
52 Hurwitz BE, Klaus JR, Llabre MM, et al. Suppression of human
immunodefi ciency virus type 1 viral load with selenium
supplementation: a randomized controlled trial. Arch Intern Med
2007; 167: 148–54.
53 Kupka R, Mugusi F, Aboud S, et al. Randomized, double-blind,
placebo-controlled trial of selenium supplements among
HIV-infected pregnant women in Tanzania: eff ects on maternal
and child outcomes. Am J Clin Nutr 2008; 87: 1802–08.
54 Beck MA, Handy J, Levander OA. Host nutritional status:
the neglected virulence factor. Trends Microbiol 2004; 12: 417–23.
55 Ashrafi MR, Shabanian R, Abbaskhanian A, et al. Selenium and
intractable epilepsy: is there any correlation? Pediatr Neurol 2007;
56 Ashrafi MR, Shams S, Nouri M, et al. A probable causative factor
for an old problem: selenium and glutathione peroxidase appear to
play important roles in epilepsy pathogenesis. Epilepsia 2007;
57 Amiri M, Farzin L, Moassesi ME, Sajadi F. Serum trace element
levels in febrile convulsion. Biol Trace Elem Res 2010; 135: 38–44.
58 Mahyar A, Ayazi P, Fallahi M, Javadi A. Correlation between
serum selenium level and febrile seizures. Pediatr Neurol 2010;
59 Shahar A, Patel KV, Semba RD, et al. Plasma selenium is positively
related to performance in neurological tasks assessing coordination
and motor speed. Mov Disord 2010; 25: 1909–15.
60 Takemoto AS, Berry MJ, Bellinger FP. Role of selenoprotein P in
Alzheimer’s disease. Ethn Dis 2010; 20 (suppl 1): S1-92–95.
61 Berr C, Balansard B, Arnaud J, Roussel AM, Alperovitch A.
Cognitive decline is associated with systemic oxidative stress: the
EVA study. Etude du Vieillissement Arteriel. J Am Geriatr Soc 2000;
62 Akbaraly NT, Hininger-Favier I, Carriere I, et al. Plasma selenium
over time and cognitive decline in the elderly. Epidemiology 2007;
63 Gao S, Jin Y, Hall KS, et al. Selenium level and cognitive function in
rural elderly Chinese. Am J Epidemiol 2007; 165: 955–65.
64 Ravaglia G, Forti P, Maioli F, et al. Homocysteine and cognitive
function in healthy elderly community dwellers in Italy.
Am J Clin Nutr 2003; 77: 668–73.
65 Rayman M, Thompson A, Warren-Perry M, et al. Impact of
selenium on mood and quality of life: a randomized, controlled
trial. Biol Psychiatry 2006; 59: 147–54.
66 Xia Y, Hill KE, Li P, et al. 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-defi cient Chinese
subjects. Am J Clin Nutr 2010; 92: 525–31.
67 Hawkes WC, Turek PJ. Eff ects of dietary selenium on sperm
motility in healthy men. J Androl 2001; 22: 764–72.
68 Rayman MP. Selenium and adverse health conditions of human
pregnancy. In: Hatfi eld DL, Berry MJ, Gladyshev VN, eds.
Selenium: its molecular biology and role in human health, 3rd edn.
New York: Springer Science and Business Media, 2011: 531–46.
69 Rayman MP, Bode P, Redman CW. Low selenium status is
associated with the occurrence of the pregnancy disease
preeclampsia in women from the United Kingdom.
Am J Obstet Gynecol 2003; 189: 1343–49.
70 Vanderlelie J, Venardos K, Clifton VL, Gude NM, Clarke FM,
Perkins AV. Increased biological oxidation and reduced anti-oxidant
enzyme activity in pre-eclamptic placentae. Placenta 2005;
71 Mistry HD, Wilson V, Ramsay MM, Symonds ME,
Broughton Pipkin F. Reduced selenium concentrations and
glutathione peeroxidase activity in preeclamptic pregnancies.
Hypertension 2008; 52: 881–88.
72 Moses EK, Johnson MP, Tømmerdal L, et al. Genetic association of
preeclampsia to the infl ammatory response gene SEPS1.
Am J Obstet Gynecol 2008; 198: 336 e1–5.
73 Rayman MP, Wijnen H, Vader H, Kooistra L, Pop V. Maternal
selenium levels during early gestation and risk for preterm birth.
CMAJ 2011; 183: 549–55.
74 Rayman MP, Thompson AJ, Bekaert B, et al. Randomized
controlled trial of the eff ect of selenium supplementation on
thyroid function in the elderly in the United Kingdom.
Am J Clin Nutr 2008; 87: 370–78.
75 Derumeaux H, Valeix P, Castetbon K, et al. Association of selenium
with thyroid volume and echostructure in 35- to 60-year-old French
adults. Eur J Endocrinol 2003; 148: 309–15.
76 Glattre E, Thomassen Y, Thoresen SO, et al. Prediagnostic serum
selenium in a case-control study of thyroid cancer. Int J Epidemiol
1989; 18: 45–49.
77 Toulis KA, Anastasilakis AD, Tzellos TG, Goulis DG, Kouvelas D.
Selenium supplementation in the treatment of Hashimoto’s
thyroiditis: a systematic review and a meta-analysis. Thyroid 2010;
78 Nacamulli D, Mian C, Petricca D, et al. Infl uence of physiological
dietary selenium supplementation on the natural course of
autoimmune thyroiditis. Clin Endocrinol (Oxf) 2010; 73: 535–39.
79 Negro R, Greco G, Mangieri T, Pezzarossa A, Dazzi D, Hassan H.
The infl uence of selenium supplementation on postpartum thyroid
status in pregnant women with thyroid peroxidase autoantibodies.
J Clin Endocrinol Metab 2007; 92: 1263–68.
80 Marcocci C, Kahaly GJ, Krassas GE, et al, European Group on
Graves’ Orbitopathy. Selenium and the course of mild Graves’
orbitopathy. N Engl J Med 2011; 364: 1920–31.
81 Rayman MP. The argument for increasing selenium intake.
Proc Nutr Soc 2002; 61: 203–15.
82 Forceville X. Eff ects of high doses of selenium, as sodium selenite,
in septic shock patients a placebo-controlled, randomized,
double-blind, multi-center phase II study – selenium and sepsis.
J Trace Elem Med Biol 2007; 21 (suppl 1): 62–65.
83 Renko K, Hofmann PJ, Stoedter M, et al. Down-regulation of the
hepatic selenoprotein biosynthesis machinery impairs selenium
metabolism during the acute phase response in mice. FASEB J
2009; 23: 1758–65.
www.thelancet.com Vol 379 March 31, 2012 1267
84 Rayman MP, Stranges S, Griffi n BA, Pastor-Barriuso R, Guallar E.
Eff ect of supplementation with high-selenium yeast on plasma lipids:
a randomized, controlled trial. Ann Intern Med 2011; 154: 656–65.
85 Flores-Mateo G, Navas-Acien A, Pastor-Barriuso R, Guallar E.
Selenium and coronary heart disease: a meta-analysis.
Am J Clin Nutr 2006; 84: 762–73.
86 Stranges S, Marshall JR, Trevisan M, et al. Eff ects of selenium
supplementation on cardiovascular disease incidence and mortality:
secondary analyses in a randomized clinical trial. Am J Epidemiol
2006; 163: 694–99.
87 Lippman SM, Klein EA, Goodman PJ, et al. Eff ect of selenium
and vitamin E on risk of prostate cancer and other cancers:
the Selenium and Vitamin E Cancer Prevention Trial (SELECT).
JAMA 2009; 301: 39–51.
88 Xun P, Liu K, Morris JS, Daviglus ML, He K. Longitudinal
association between toenail selenium levels and measures of
subclinical atherosclerosis: the CARDIA trace element study.
Atherosclerosis 2010; 210: 662–67.
89 Luoma PV, Sotaniemi EA, Korpela H, Kumpulainen J. Serum
selenium, glutathione peroxidase activity and high-density
lipoprotein cholesterol–eff ect of selenium supplementation.
Res Commun Chem Pathol Pharmacol 1984; 46: 469–72.
90 Yu SY, Mao BL, Xiao P, et al. Intervention trial with selenium for
the prevention of lung cancer among tin miners in Yunnan, China.
A pilot study. Biol Trace Elem Res 1990; 24: 105–08.
91 Duffi eld AJ, Thomson CD, Hill KE, Williams S. An estimation of
selenium requirements for New Zealanders. Am J Clin Nutr 1999;
92 Hurst R, Armah CN, Dainty JR, et al. Establishing optimal
selenium status: results of a randomized, double-blind,
placebo-controlled trial. Am J Clin Nutr 2010; 91: 923–31.
93 Blankenberg S, Rupprecht HJ, Bickel C, et al. Glutathione
peroxidase 1 activity and cardiovascular events in patients with
coronary artery disease. N Engl J Med 2003; 349: 1605–13.
94 Zhuo H, Smith AH, Steinmaus C. Selenium and lung cancer:
a quantitative analysis of heterogeneity in the current
epidemiological literature. Cancer Epidemiol Biomarkers Prev 2004;
95 Amaral AF, Cantor KP, Silverman DT, Malats N. Selenium and
bladder cancer risk: a meta-analysis.
Cancer Epidemiol Biomarkers Prev 2010; 19: 2407–15.
96 Peters U, Takata Y. Selenium and the prevention of prostate and
colorectal cancer. Mol Nutr Food Res 2008; 52: 1261–72.
97 Yu MW, Horng IS, Hsu KH, Chiang YC, Liaw YF, Chen CJ. Plasma
selenium levels and risk of hepatocellular carcinoma among men
with chronic hepatitis virus infection. Am J Epidemiol 1999;
98 Etminan M, FitzGerald JM, Gleave M, Chambers K. Intake of
selenium in the prevention of prostate cancer: a systematic review
and meta-analysis. Cancer Causes Control 2005; 16: 1125–31.
99 Brinkman M, Reulen RC, Kellen E, Buntinx F, Zeegers MP. Are
men with low selenium levels at increased risk of prostate cancer?
Eur J Cancer 2006; 42: 2463–71.
100 Rayman MP. Selenium. In: Milner JA, Romagnolo DF, eds.
Bioactive compounds and cancer. New York: Humana Press/
Springer, 2010: 411–48.
101 Mahabir S, Forman MR, Dong YQ, Park Y, Hollenbeck A,
Schatzkin A. Mineral intake and lung cancer risk in the
NIH-American Association of Retired Persons Diet and Health
study. Cancer Epidemiol Biomarkers Prev 2010; 19: 1976–83.
102 Epplein M, Franke AA, Cooney RV, et al. Association of plasma
micronutrient levels and urinary isoprostane with risk of lung
cancer: the multiethnic cohort study.
Cancer Epidemiol Biomarkers Prev 2009; 18: 1962–70.
103 Mahabir S, Spitz MR, Barrera SL, Beaver SH, Etzel C, Forman MR.
Dietary zinc, copper and selenium, and risk of lung cancer.
Int J Cancer 2007; 120: 1108–15.
104 Peters U, Foster CB, Chatterjee N, et al. Serum selenium and risk of
prostate cancer-a nested case-control study. Am J Clin Nutr 2007;
105 Peters U, Littman AJ, Kristal AR, Patterson RE, Potter JD, White E.
Vitamin E and selenium supplementation and risk of prostate
cancer in the Vitamins and lifestyle (VITAL) study cohort.
Cancer Causes Control 2008; 19: 75–87.
106 Allen NE, Appleby PN, Roddam AW, et al. European Prospective
Investigation into Cancer and Nutrition. Plasma selenium
concentration and prostate cancer risk: results from the European
Prospective Investigation into Cancer and Nutrition (EPIC).
Am J Clin Nutr 2008; 88: 1567–75.
107 Steinbrecher A, Méplan C, Hesketh J, et al. Eff ects of selenium
status and polymorphisms in selenoprotein genes on prostate
cancer risk in a prospective study of European men.
Cancer Epidemiol Biomarkers Prev 2010; 19: 2958–68.
108 Gill JK, Franke AA, Steven Morris J, et al. Association of selenium,
tocopherols, carotenoids, retinol, and 15-isoprostane F(2t) in serum
or urine with prostate cancer risk: the multiethnic cohort.
Cancer Causes Control 2009; 20: 1161–71.
109 Bjelakovic G, Nikolova D, Simonetti RG, Gluud C. Antioxidant
supplements for prevention of gastrointestinal cancers: a systematic
review and meta-analysis. Lancet 2004; 364: 1219–28.
110 Clark LC, Combs GF Jr, Turnbull BW, et al. Eff ects of selenium
supplementation for cancer prevention in patients with carcinoma
of the skin. A randomized controlled trial. Nutritional Prevention of
Cancer Study Group. JAMA 1996; 276: 1957–63.
111 Duffi eld-Lillico AJ, Reid ME, Turnbull BW, et al. Baseline
characteristics and the eff ect of selenium supplementation on
cancer incidence in a randomized clinical trial: a summary report of
the Nutritional Prevention of Cancer Trial.
Cancer Epidemiol Biomarkers Prev 2002; 11: 630–39.
112 Reid ME, Duffi eld-Lillico AJ, Garland L, Turnbull BW, Clark LC,
Marshall JR. Selenium supplementation and lung cancer incidence:
an update of the nutritional prevention of cancer trial.
Cancer Epidemiol Biomarkers Prev 2002; 11: 1285–91.
113 Duffi eld-Lillico AJ, Dalkin BL, Reid ME, et al. Nutritional
Prevention of Cancer Study Group. Selenium supplementation,
baseline plasma selenium status and incidence of prostate cancer:
an analysis of the complete treatment period of the Nutritional
Prevention of Cancer Trial. BJU Int 2003; 91: 608–12.
114 Duffi eld-Lillico AJ, Slate EH, Reid ME, et al. Nutritional Prevention
of Cancer Study Group. Selenium supplementation and secondary
prevention of nonmelanoma skin cancer in a randomized trial.
J Natl Cancer Inst 2003; 95: 1477–81.
115 Cooper ML, Adami H-O, Grönberg H, Wiklund F, Green FR,
Rayman MP. Interaction between SNPs in selenoprotein P and
mitochondrial superoxide dismutase determines prostate cancer
risk. Cancer Res 2008; 68: 10171–77.
116 Rayman MP, Combs GF Jr, Waters DJ. Selenium and vitamin E
supplementation for cancer prevention. JAMA 2009; 301: 1876.
117 Bekaert B, Rayman MP. Plasma selenium concentration and
prostate cancer risk. Am J Clin Nutr 2009; 89: 1276–77.
118 Penney KL, Schumacher FR, Li H, et al. A large prospective study of
SEP15 genetic variation, interaction with plasma selenium levels,
and prostate cancer risk and survival. Cancer Prev Res (Phila) 2010;
119 Steinbrenner H, Speckmann B, Pinto A, Sies H. High selenium
intake and increased diabetes risk: experimental evidence for
interplay between selenium and carbohydrate metabolism.
J Clin Biochem Nutr 2011; 48: 40–45.
120 Navarro-Alarcon M, López-G de la Serrana H, Perez-Valero V,
Lopez-Martinez C. Serum and urine selenium concentrations as
indicators of body status in patients with diabetes mellitus.
Sci Total Environ 1999; 228: 79–85.
121 Kljai K, Runje R. Selenium and glycogen levels in diabetic patients.
Biol Trace Elem Res 2001; 83: 223–29.
122 Rajpathak S, Rimm E, Morris JS, Hu F. Toenail selenium and
cardiovascular disease in men with diabetes. J Am Coll Nutr 2005;
123 Akbaraly TN, Arnaud J, Rayman MP, et al. Plasma selenium and
risk of dysglycemia in an elderly French population: results from
the prospective Epidemiology of Vascular Ageing Study.
Nutr Metab (Lond) 2010; 7: 21–27.
124 Bleys J, Navas-Acien A, Guallar E. Serum selenium and diabetes
in U.S. adults. Diabetes Care 2007; 30: 829–34.
125 Laclaustra M, Navas-Acien A, Stranges S, Ordovas JM, Guallar E.
Serum selenium concentrations and diabetes in U.S. adults:
National Health and Nutrition Examination Survey (NHANES)
2003–2004. Environ Health Perspect 2009; 117: 1409–13.
www.thelancet.com Vol 379 March 31, 2012
126 Czernichow S, Couthouis A, Bertrais S, et al. Antioxidant
supplementation does not aff ect fasting plasma glucose in the
Supplementation with Antioxidant Vitamins and Minerals (SU.
VI.MAX) study in France: association with dietary intake and
plasma concentrations. Am J Clin Nutr 2006; 84: 395–99.
127 Stranges S, Marshall JR, Natarajan R, et al. Eff ects of long-term
selenium supplementation on the incidence of type 2 diabetes:
a randomized trial. Ann Intern Med 2007; 147: 217–23.
128 Houstis N, Rosen ED, Lander ES. Reactive oxygen species have
a causal role in multiple forms of insulin resistance. Nature 2006;
129 McClung JP, Roneker CA, Mu W, et al. Development of insulin
resistance and obesity in mice overexpressing cellular glutathione
peroxidase. Proc Natl Acad Sci USA 2004; 101: 8852–57.
130 Chen X, Scholl TO, Leskiw MJ, Donaldson MR, Stein TP.
Association of glutathione peroxidase activity with insulin
resistance and dietary fat intake during normal pregnancy.
J Clin Endocrinol Metab 2003; 88: 5963–68.
131 Labunskyy VM, Lee BC, Handy DE, Loscalzo J, Hatfi eld DL,
Gladyshev VN. Both maximal expression of selenoproteins and
selenoprotein defi ciency can promote development of type 2
diabetes-like phenotype in mice. Antioxid Redox Signal 2011;
132 Ruston D, Hoare S, Henderson L, et al. The National Diet and
Nutrition Survey: adults aged 19 to 64 years. Nutritional status
(anthropometry and blood analytes), blood pressure and physical
activity (vol 4). London: Stationery Offi ce, 2004.
133 National Center for Health Statistics. National Health and Nutrition
Examination Survey. 2003–04. US Department of Health and
Human Services, Centers for Disease Control. http://www.cdc.gov/
nchs/nhanes.htm (accessed March 13, 2011).