A Review of Dietary Selenium Intake and Selenium Status in Europe and the Middle East

This is a systematic review of existing data on dietary selenium (Se) intake and status for various population groups in Europe (including the United Kingdom (UK)) and the Middle East. It includes English language systematic reviews, meta-analyses, randomised controlled trials, cohort studies, cross-sectional and case-control studies obtained through PUBMED searches from January, 2002, to November, 2014, for European data and from 1990 to November 2014, for Middle Eastern data. Reports were selected if they included data on Se intake and status. The search identified 19 European/UK studies and 15 investigations in the Middle East that reported Se intake and Se concentration in water and/or food and 48 European/UK studies and 44 investigations in the Middle East reporting Se status. Suboptimal Se status was reported to be widespread throughout Europe, the UK and the Middle East, and these results agreed with previous reports highlighting the problem. Eastern European countries had lower Se intake than Western European countries. Middle Eastern studies provided varying results, possibly due to varying food habits and imports in different regions and within differing socioeconomic groups. In conclusion, Se intake and status is suboptimal in European and Middle Eastern countries, with less consistency in the Middle East.


Introduction
Interest in selenium (Se) has been growing over the past few decades in a number of areas of human health. The nutritional status of this metalloid has been difficult to assess via food intake data alone, because many factors influence its presence in the food chain. Although its distribution in soil is uneven [1] , a number of other factors affect its concentration in various foods, including varying uptake into plants due to soil pH, rainfall, land contour and microbial activity [2] and importation of food from higher Se areas. Therefore, estimates of Se intake from food nutrient databases alone can give inaccurate measures of Se status within a population or population groups.
Se is an essential non-metal trace element [3] that is required for selenocysteine synthesis and is essential for the production of selenoproteins [4]. Selenoproteins are primarily either structural or enzymatic [2], acting as catalysts for the activation of thyroid hormone and as antioxidants, such as glutathione peroxidases (GPxs) [5]. GPx activity is commonly used as a marker for Se sufficiency in the body [6], where serum or plasma Se concentrations are believed to achieve maximum GPx expression at 90-100 μg/L (90.01 μg/L as proposed by Duffield and colleagues [7] and 98.7 μg/L according to Alfthan et al. [8]). However, plasma selenoprotein P (SEPP1) concentration is a more suitable marker than plasma GPx activity [9]. Prospective studies provide some evidence that adequate Se status may reduce the risk of some cancers, while elevated risk of type 2 diabetes and some cancers occurs when the Se concentration exceeds 120 μg/L [10]. Higher Se status has been linked to enhanced immune competence with better outcomes for cancer, viral infections, including HIV progression to AIDS, male infertility, pregnancy, cardiovascular disease, mood disorders [2] and, possibly, bone health [11][12][13][14].
The two main dietary forms of Se are selenocysteine derived from animal-sourced foods and selenomethionine obtained from animal-sourced foods and cereal products grown on Se-rich soil in areas, such as in the United States (US) [6]. However, animal-derived foods tend to be a better dietary source of this nutrient for humans, because Se is required as an essential nutrient for animal life, and animal feed is sometimes supplemented with Se [15]. Figure 1 shows the biosynthetic pathway for these two amino acids in plants, marine algae and brewer's yeast.
Health conditions caused by severe Se deficiency include Keshan disease and Kashin-Beck disease. However, both of these conditions occur more often in conjunction with iodine deficiency or in the presence of certain environmental toxins. In the case of Keshan disease, viral infection has also been implicated [2].
The Food and Nutrition Board at the Institute of Medicine of the National Academies, US, has recommended, for 19-50-year-old men and women, 45 micrograms (μg) of Se/day as the estimated average requirement (EAR), 55 μg of Se/day as the recommended dietary allowance (RDA) and 400 μg of Se/day as the tolerable upper intake level (UL) [4,16]. A peer-reviewed article quoted a safety intake recommendation at around 800 μg of Se/day for no observed adverse effect level (NOAEL), 1540 to 1600 μg of Se/day for low observed adverse effect level (LOAEL) and 5000 μg of Se/day for a toxic level, where selenosis occurs [17]. LOAEL symptoms have been associated with hair, toe and fingernail loss and a garlicky odour on the breath, whilst toxicity can cause acute respiratory distress syndrome, myocardial infarction and renal failure [4].
The recommended reference nutrient intake (RNI) of Se for adults in the United Kingdom (UK) is 60 μg/day for adult women and 75 μg/day for lactating women and adult men [18]. The US daily recommended intake (DRI) is 55 μg/day for both men and women [16] and the European Food Safety Authority (EFSA), in the European Union (EU), has recently set a daily adequate intake for Se at 70 μg [9]. To date, the optimal intake to achieve additional health benefits over general health maintenance and the best biological marker(s) to assess Se sufficiency have not been conclusively established [3].
It is generally recognised that Se intakes across Europe are low, reflecting inadequate soil levels, particularly in Eastern Europe. Se intake in the UK has declined since the 1970s, and previous government surveys [19] indicate low Se intake across a wide age range of the UK population (Caucasian and Asian) [20].
A number of studies investigating Se levels in blood (serum and/or plasma), breast milk and umbilical cord blood indicate generally low Se status in women from the Kingdom of Saudi Arabia (KSA), Kuwait and Turkey [21][22][23], but to date, to the author's knowledge, no systematic reviews have been completed on the data originating from the Middle East. In 2002, Rayman et al. [2] reported similarly low serum or plasma Se status in numerous European countries. In this publication, we review and summarize Se intake and status data collected and reported since 2002 in European countries and since 1990 for Middle Eastern countries. A comparison of the overall Se status in the UK and Middle Eastern countries was of interest to the authors, since a manuscript pertaining to Se status in Saudi Arabian women and UK age-matched controls and its influence on calcium homeostasis, bone density and skeletal metabolism [24] was in preparation.

Experimental Section
A review of current literature was undertaken to obtain data on Se intake and status in European and Middle Eastern countries. A review article containing relevant data pertaining to European countries had been previously published in 2002 [2]. Therefore, 2002 was used as the starting point for studies selected from the literature search for inclusion in this systematic review. However, there have been no previously published review articles on the subject for Middle Eastern countries. Rather than completing an extensive historical search, the year 1990 was selected as a starting point for Middle Eastern studies to include in this systematic review.
A systematic search was performed on PUBMED for English-language articles published from January, 2002, through to November, 2014, to obtain European data and from January, 1990, through to November, 2014, to obtain Middle Eastern data using the following search words: bone health and Middle East/KSA/UK, dietary surveys and Middle East/KSA/UK, selenium and bone, antioxidants and bone, selenium gene expression and bone, selenium status and Middle East/KSA/UK, selenium levels in food and water, vitamin D/vitamin D status and bone/Middle East/KSA/UK, vitamin D and selenium dependent enzymes, vitamin D and osteoporosis and vitamin D deficiency and osteoporosis. Additional studies were identified through a Google search using the search criterion, selenium deficiency. Titles and abstracts were reviewed and reports were selected for inclusion in the review if they were systematic reviews, meta-analyses, randomised controlled trials, cohort studies, cross-sectional or case-control studies and if they included data on Se intake and status.

Selenium Intake Studies
The literature search identified 19 European/UK studies and 15 investigations in the Middle East that reported Se intake and Se concentration in water and/or food. A summary of the findings are presented in Tables 1 and 2. 3.1.1. Selenium Intake Studies: European Countries/UK Eastern European countries tended to have lower Se intake than their western counterparts, as indicated by a Polish study [25], where Se content of the consumed foods was four-times lower than in Spain, which tended to have intakes that exceeded the DRI and RDA [26][27][28][29][30]. Studies in France [31] and Belgium [32] reported intakes equivalent to the RDA, while those from Slovenia [33] and Italy [34,35] found intakes below the RDA.  Case-control study of Se in people exposed to Se concentration in drinking water greater than the maximum recommended limit (10 μg/L) using an FFQ 40 exposed subjects and 40 non-exposed controls Exposed subjects intake 64 ± 14 μg/day; non-exposed 52 ± 14 μg/day Belgium [32] To determine the Se status of the population    Primary sources of Se in the diet were: meat and meat products (31%), egg (20.4%), cereals and cereal products (16%), legumes (8.7%), fruits (6.8%), milk and dairy products (2.0%), beverages (2%), sweets (1.8%), pickles (0.2%) and oil (0.02%).
Daily intake of Se was estimated to be 93 μg/day Iran [59] Cross-sectional analysis to determine Se intake using a 3-day food record in postmenopausal women 30 postmenopausal women Daily intake of Se was estimated to be 40 μg/day * These values are substantially higher than most others provided in Table 2. The corresponding author of the reference was contacted to confirm the units stated in the reference. No response was obtained prior to this manuscript submission. Therefore, this reference was excluded from our interpretation of the data.
A 2008 longitudinal study in healthy British adults reported low Se intake [37]. Subsequently, the National Diet and Nutrition Survey (NDNS) 2008/2009 found that roughly 50% of adult women and older girls and 20% of men and older boys had Se intakes below the lower reference nutrient intake (LRNI) level [36]. A 2009 longitudinal analysis of Se intakes of Caucasian and South Asian women living in the UK who were enrolled in the vitamin D, Food Intake, Nutrition and Exposure to Sunlight in Southern England (D-FINES) study [60] showed that the RNI of Se was not achieved by 80%-90% of Caucasians and by 83%-95% of the South Asians. Likewise, 60% of the former and 60%-70% of the latter failed to meet the LRNI. The number of women that had Se intakes below the LRNI or RNI was not significantly different in terms of ethnicity. Only in the South Asian women was a seasonal variation noted, with significantly lower Se intake in the autumn [20].

Selenium Intake Studies: Middle Eastern Countries
Breast milk analysis in Turkey shows that Se content is below the international reference range set at 18.5 μg/L throughout the lactation period [61]. Analysis of various dairy products in Turkey also indicates that the Se content varies depending on the type of food. Butter and different types of cheeses had quite high concentrations, whereas other dairy products, such as milk and ice cream, contained negligible amounts.
The concentration of Se in water is extremely variable and, in some places, excessive, as reported in parts of Jordan [47]. However, a recent case-control study found Se intakes equivalent to the RDA in control subjects, while patients with colorectal cancers consumed significantly less [53].
Recent studies within Iran have reported adequate Se intakes within the general population [54,55], but extremely low intakes in oesophageal squamous cell carcinoma patients [57].
The concentration of Se in wheat grain grown in the KSA is extremely variable, with one study reporting a range from 8 to 293 μg/kg (median 78.4 μg/kg) [48]. The lowest Se content (average 50.6 μg/kg) was found in wheat from the Eastern parts of KSA, in the Wadi Al-Dowasir area, while the highest levels (average 285.5 μg/kg) was found in the northwest, around the Al-Jouf region. A further investigation of Se in soil, water and alfalfa from Al-Kharj in the centre of the country found that some samples had Se content as low as the low-Se zone in China [48].
Overall, Se intake varied throughout the Middle East. A recent study from Riyadh reported intakes nearly double the RDA [58], while a study from King Abdulaziz University in KSA reported intakes between 75 and 121.65 μg/person per day in a population sample in Jeddah [49]. In this study, the major food groups providing Se were cereals/cereal products, legumes and meats. Another study, including infant formulas, mostly imported from Europe, contained adequate levels of this trace element for infants up to six months [50,51]. However, some breast-fed children may still be at risk of low Se intakes, according to a study by Al Saleh et al., where intakes between 0.9 and 15 μg/day were reported [51,52].

Selenium Status Studies
The literature search located 48 European/UK studies and 44 investigations in the Middle East reporting Se status. A summary of the findings are presented in Tables 3 and 4.    Germany [72] Cross-sectional analysis of participants in the Lipid Analytic Cologne cohort 792 participants who never smoked, who did not use antihypertensives and who did not have diabetes or known atherosclerotic disease Mean serum Se concentration was 68 ± 32 μg/L     Prospective cross-sectional analysis during early gestation to determine the impact of Se status on the risk of preterm births 1197 white Dutch women with a singleton pregnancy followed from 12 weeks' gestation Serum Se at 12 weeks' gestation was significantly lower among women who had a preterm birth than among those who delivered at term mean 0.96 ± 0.14 μmol/L vs.   Mean ± SD of milk Se (μg/L) 0-6-month lactation period were 20 ± 0.8 and 16 ± 0.4 in Kuwaiti and non-Kuwaiti women, respectively; mean ± SD of milk Se (μg/L) 6-12-month lactation period were 18 ± 0.4 and 15 ± 0.2 in Kuwaiti and non-Kuwaiti women, respectively; in both Kuwaiti and non-Kuwaiti women, the milk Se was significantly lower at the 6-12-month lactation period compared to the 0-6-month one (p < 0.05); in addition, Kuwaiti women had significantly higher milk Se levels during both the 0-6-month and 0-12-month lactation periods Turkey [23] Cross-sectional analysis of Se in mothers and their neonates using hair, breast milk, meconium and maternal and umbilical cord blood

pairs of mothers and their new-born babies
Average Se concentrations in blood plasma: 68.5 ± 3.6 ng/g (~70 μg/L using density of blood plasma 1.025 g/mL) Group A (control, n = 101) mean serum Se 50 ± 9.8 μg/L; Group B (abnormal 50-g and abnormal 100-g OGTT, n = 30) mean serum Se 34.7 ± 8.7 μg/L; Group C (glucose intolerant (GIT), n = 49) mean serum Se level 39.9 ± 6.5 μg/L Iran [105] Cross-sectional analysis of serum Se in healthy individuals living in Tehran 184 healthy inhabitants, male and female; 24 women (over 16 years) Mean ± SD serum Se level was 93.9 ± 13.6 μg/L for women and 102.2 ± 12 μg/L for men. The intake in men was statistically significantly higher than in women p < 0.0005    Retrospective cohort, including previously reported indices of bone health and dietary Se intake and Se analysis (during 2010) of previously collected plasma and serum in Jeddah;Saudi Arabian Bone Health Study (SABHS) study [120].   Case-control study to evaluate the possible associations between serum Se and idiopathic intractable epilepsy Case-control study to evaluate the effects of pica and iron-deficiency anaemia on oxidative stress and antioxidant capacity 47 children with iron-deficiency anaemia plus pica, 22 children with iron-deficiency anaemia only and 21 non-anaemic children as controls Mean serum Se was 57.5 + 11.2 μg/L in those with pica, 62 + 11 μg/L in those with iron-deficient anaemia only and 80.9 + 13.1 μg/L in controls Turkey [133] Case-control study to evaluate serum Se in relation with hyperandrogenism and insulin resistance in women with polycystic ovary syndrome (PCOS) 36 cases with a diagnosis of PCOS and 33 age-and BMI-matched healthy women Mean serum Se was 41.8 ± 1.8 μg/L in the PCOS women and 49.7 ± 1.6 μg/L in controls Turkey [134] Case-control study to investigate the plasma and erythrocyte Se in H1N1-infected children 11 infected children (4 girls, 7 boys; mean age 9.3 ± 4.5 years) and 12 controls (6 girls, 6 boys; mean age 10.   Mean serum Se was 86.08 μg/L in the obese group and 101.14 μg/L in the control group * Se μg/L calculated using an atomic weight of 78.96; ** these values are substantially higher than most others provided in Table 4. The corresponding author of the reference was contacted to confirm the units stated in the reference. No response was obtained prior to this manuscript submission. Therefore, this reference was excluded from our interpretation of the data. Overall, the results indicate that if 98.7 μg/L of Se in plasma or serum are required to optimize GPx activity [8], then suboptimal Se status is found in both regions, but is less consistent in the Middle East. Figure 2 shows the serum and/or plasma Se concentrations of study subjects in various European countries (including the UK) in relation to the suggested levels required for full GPx expression, as well as the level proposed by the WHO to achieve two thirds of the GPx expression. Figure 3 gives the same information in relation to the Se concentrations found for the study subjects in various Middle Eastern countries.

Selenium Status Studies: European Countries/UK
Results of Se status studies in Europe suggest suboptimal blood Se concentrations for populations in most of this region if 90.01 μg of Se/L of plasma or serum is the assumed requirement for full plasma GPx expression [7]. Exceptions were Austria [89], Hungary [98], healthy subjects in Denmark [75,95], healthy subjects, but not obese [86] or those with Type 1 diabetes [87] in Poland and some of the participants in the Dietary Habit Profile in European Communities with Different Risk of Myocardial Infarction: the Impact of Migration as a Model of Gene-Environment Interaction "IMMIDIET" study [84]. The lowest reported serum Se status of all European studies investigated was in Albanian adults living in Greece, with 37.4 μg/L [62]. Schulpis et al. [62] speculated these low values could be related to their poor animal protein intake, which could be the consequence of their low socioeconomic status. The highest Se status of all studies in the European region was in Poland, where boys had a mean serum Se concentration of 111.1 μg/L [86]. Finland was an exception, where Se levels were amongst the lowest in the world until the early 1980s, before the implementation of a nationwide Se fertilisation programme. People in this region are now Se sufficient [83,90].
There were unusual findings noted in some studies from European countries. For example, there was a larger than expected gender effect in plasma Se concentrations reported from Granada, Spain, where males had 87 μg/L and women had 67 μg/L [96]. There was no explanation for this gender difference in the reference, and lifestyle factors, including smoking, alcohol consumption, etc., were reported to show no significant association with plasma Se levels. However, there were significantly higher plasma Se levels in individuals with higher energy intake who performed more physical activity, although this was not reported separately by gender [96]. Therefore, it is difficult to speculate the reason for the gender discrepancy.
UK studies, including a broad range of population groups, identified a fairly consistent overall suboptimal Se status in this country based on the range of 90-100 μg/L to achieve maximum GPx expression [20,37,60,70,71,[77][78][79][80][81]. Pre-term infants in the Hampshire area who had received parenteral nutrition and blood transfusions had significantly lower mean plasma Se levels that normalised by six months of gestation-corrected age. However, term infants had normal Se status at birth and six months of age [70]. In 14-18-year-old adolescents where plasma Se was determined between 28 and 32 weeks of gestation, plasma Se concentration was lower in those who gave birth to small-for gestational age (SGA) infants compared to those having appropriate-for-gestational age (AGA) infants [79]. Smoking mothers had a lower Se concentration compared with non-smokers, and Afro-Caribbean women had higher Se concentrations compared with white Europeans [79]. The overall low plasma Se concentration in these adolescent mothers was speculated to contribute to the risk of delivering an SGA infant, possibly through lowering placental antioxidant defence, thus directly affecting foetal growth [79]. Differences in plasma Se between ethnicities was thought to be related to nutritional intake variation [79]. Longitudinal analysis of healthy British middle-aged adults found suboptimal Se status in both males and females [37]. Analysis of plasma/serum Se concentration in a number of D-FINES study participants showed that full plasma GPx expression (serum or plasma Se level of 98.7 μg/L) was not achieved in the majority of all population groups included in the study, and postmenopausal women had higher Se status across all ethnicities [24]. Se deficiency in independent and institutionalised adults over age 65 years was also consistent and found to be lowest in the autumn and highest in spring [71]. However, a study investigating the effect of Se supplementation on thyroid function in the elderly found a mean baseline plasma Se concentration sufficient to achieve maximum GPx expression [5].

Selenium Status Studies: Middle Eastern Countries
Se status studies throughout the Middle East provided varying results. Even within the same country, this variability was apparent. For example, in Iran, mostly case-control studies reported serum Se concentrations below 90 μg/L [106,109,110,123,[128][129][130], almost as many reporting serum Se content between 90 and 100 μg/L [105,107,108,127], and some found above 120 μg/L [111,124,125]. The highest serum Se in normal/control subjects was found in Lebanon (151.2 μg/L) [112], and the lowest was reported in Turkey in school children from an endemic goitre area (31 ± 23 μg/L) [45].
The majority of studies from Turkey indicated that the population had suboptimal Se status [23,45,103,104,132,133,135] if 90.01 μg of Se/L of plasma or serum is the assumed requirement for full plasma GPx expression [7]. However, two studies reported plasma [134] and serum [136] Se concentrations slightly above 120 μg/L.
All of the case-control studies from Egypt reported serum Se content of less than 90 μg/L in healthy control subjects [137][138][139].
Analysis of serum/plasma samples from the Saudi Arabian Bone Health Study, funded by King Abdulaziz University, showed that a large proportion of the KSA women under investigation had low Se status [24]. High blood Se concentrations were recorded in some KSA studies, including patients with cardiovascular disease [1,118], in a Lebanon study investigating metabolic syndrome [112] and one study in Jordan investigating Se status in smokers and non-smokers [114].
Toenail analysis, which is a measure of intake over a longer term, indicates low consumption in the Al-Kharj district [21,102]. Although none of the participants had serum Se levels below 45 μg/L, which would be considered as inadequate, 41% had toenail Se below 56 μg/g, indicative of a low Se status. A 2006 study that reported serum and toenail Se status indicated a similar outcome in children from the same district. In that study, serum and toenail Se were significantly related, a finding that could reflect dietary Se intake [116].
Infant blood Se status, according to more than one study in the Middle East, is within the normal range [103,115], while one study in Yemeni infants found Se blood status to be very low [113]. However, studies that date back to the 1990s indicate low Se content in breast milk [52]. One study of Kuwaiti and non-Kuwaiti women residents in Kuwait indicates that the latter have Se breast milk levels below the 18.5 μg/L (international reference range) at term and from 6 to 12 months during the lactation period [61]. Kuwaiti women had sufficient levels at term, but dropped just below the recommended level from 6 to 12 months. These results agree with low maternal-foetal Se status found in gestational diabetic pregnancies, where serum Se concentrations for the obese control and diabetic groups were below the level required to optimise GPx activity [117].

Discussion
This review of recently published studies on Se intake and status throughout Europe, the UK and the Middle East shows that Se deficiency is widespread among these populations and agrees with previous reports highlighting the problem. An earlier report of low serum or plasma Se throughout the UK population included data from the UK PRECISE study [78], where the mean plasma Se at baseline was 90.8 μg/L, below the Se concentration required for optimal plasma GPx activity according to Alfthan et al. [8]. Another publication reported daily Se intake less than the RDA of 55 μg for numerous countries throughout Europe (including the UK) and the Middle East [3].
The low Se intakes in the UK population, as reported in the NDNS 2008/2009 [36] and mirrored in the D-FINES study [20,60], agree with previous government surveys, including that of the Ministry of Agriculture, Fisheries and Foods and an analysis of Se in food completed during the mid to late 1990s [142], where UK Se intakes ranged from 29 to 39 μg/day [19]. Given the geographical location of Northern Ireland relative to the UK, it is not surprising that the Se intake is less than the RDA in that country, as well [38].
The Se blood concentrations reported in the various studies outlined in this review indicate low to borderline Se status in the participants if 90.01 μg of Se/L of plasma or serum is the assumed requirement for full plasma GPx expression [7]. However, if the higher Se concentration of 98.7 μg/L were required [8], then the Se status of many study subjects would be considered inadequate. As plasma and serum Se concentration only reflects Se intake over the short term, it is unclear what similar long-term Se measurements would show. Bearing in mind that a high proportion of the UK population fail to obtain the RNI for Se and with Se soil levels low in the UK, it is highly likely that the Se status of the UK population is less than adequate. It appears that in spite of acknowledging the risk of low Se intakes in the UK population over a decade ago, little impact has been made to improve this situation [36]. A switch from importing wheat grown within the United States and Canada to that grown in Europe is believed partly to blame for the low dietary Se intake in recent years. However, even if optimal plasma Se levels (e.g., 105-115 μg/L, compared with a current estimated mean of approximately 87 μg/L) were attained within the UK population, there is no guarantee that population disease levels would improve in the medium term. Although, some individuals, such as heavy smokers or people with specific mutations within genes coding for mitochondrial antioxidant enzymes [10], would likely benefit with increased Se intake.
Se status/intake studies within the overall Middle East provide varying results. Toenail analysis indicates low consumption in the Al-Kharj district. Serum Se levels are regarded to reflect short-term Se intakes, whereas the toenail provides a better measure for long-term Se intakes [6]. This finding concurs with the low Se intake levels in infants and babies of between 0.9 and 15 μg/day [52]. It is interesting to note that in certain areas of Jordan, the Se content of aquifers is very high, which might have had an influence on the very high blood Se levels found in smokers and non-smokers in that country. In addition, Se content from tobacco products might contribute to the even higher blood Se concentrations found in smokers. In Turkey and Iran, where many studies have been carried out to assess Se status in different population and patient groups, some studies reported very low blood Se levels, similar to those in Eastern Europe; a few were high, but overall, the values were comparable to Central European status. If a serum or plasma Se concentration of 90.01 μg/L is the assumed level for sufficiency, then more of the reviewed Middle Eastern studies would meet this reference value than not.
The KSA studies indicated that the soil Se levels and food grown in those areas within KSA [48] are low to deficient. Similar variation in Se content has been reported in Hungary [98], which is considered a low to deficient Se area in Europe according to current Se intake recommendations [15]. There is currently an effort by the KSA government to ensure food security in the Kingdom by encouraging home-grown produce. However, food imports form a significant part of the food supply, with a large part of this sourced from the United States, where soil Se levels tend to be high to adequate [15,143]. This could be contributing to higher Se intake in some people in KSA and may be reflected in the King Abdulaziz University study in Jeddah [49]. In this study, the major food groups providing Se were cereals/cereal products, legumes and meats. Infant formulas imported from Europe may ensure adequate intake in infants consuming those nutrition sources [50,51]. However, some breast-fed children may not be obtaining sufficient Se [51,52]. Dietary habits may differ between rural and urban areas and different parts of the country, where the availability of certain imported food items might differ. This, in addition to food security influences and social status, could have a considerable effect on Se status and explain the varied Se concentrations that were found in the KSA population.

Conclusions
Overall, the results of this systematic review indicate that Se intake and status is suboptimal in European and Middle Eastern countries with less consistency in the Middle East according to current measures of sufficiency. These results combined with growing knowledge of the importance of Se to overall health warrant more work initially to establish an agreed range of blood Se concentration or other markers to determine optimal Se intake and, subsequently, to ensure adequate Se supplementation in populations at risk of low Se intake. The Se fertilisation programme in Finland would appear to have successfully raised Se status in that population.

Conflicts of Interest
Wassen International Ltd. Cedar Court Office Park, Denby Dale Road, Wakefield WF4 3DB, UK, is an international distributor of nutritional supplements, including some that contain Se. Rita Stoffaneller was employed by Wassen International Ltd. at the time of completing her M.Sc. and was engaged in the capacity of consultant by Wassen International Ltd. during the preparation of the draft manuscript. This work was supported by a financial contribution from Wassen International Ltd., and they were involved in the decision to publish the results. The funder enabled Rita Stoffaneller to complete her M.Sc. requirements while under their employ, and only in this capacity did they contribute to the study design, conduct of the study, analysis of the data and interpretation of the findings. Nancy Morse was hired on contract by Wassen International Ltd. to complete a literature search and to write and submit the final paper.