Selenium Levels of Estonians

ORIGINAL ARTICLE Selenium levels of Estonians

P Rauhamaa1, M Kantola1, A Viitak2, T Kaasik3 and H Mussalo-Rauhamaa4

1Finnish Forest Research Institute, Vantaa, Finland; 2Department of Chemistry, Tallinn Technical University, Estonia; 3Department of Public Health, University of Tartu, Estonia and 4Department of Public Health, University of Helsinki, Finland

Background: The objective of this study is to sum up the dietary selenium intake of the Estonian people according to the serum selenium concentrations. Subjects/methods: This research compiles the information published in the literature about the levels of selenium in the serum of the Estonian people. We compare these results with the findings obtained from the analyses of serum samples gathered by us in the 1990s in Estonia. Results: The selenium concentration in sera of 404 Estonians ranged from 26 to 116 mg/l; the mean was 75 mg/l. The selenium contents of Estonians were at a similar level as reported for Finns before the selenium supplementation of fertilizers carried out in Finland in 1984. Conclusion: In view of the selenium concentration in the sera, we suggest that the dietary selenium intake among the Estonian people might be scarce, and Estonian authorities should consider in their nutrition policies the possible low intake of selenium of Estonians. European Journal of Clinical Nutrition (2008) 62, 1075–1078; doi:10.1038/sj.ejcn.1602818; published online 30 May 2007

Keywords: selenium; intake; serum; Estonia

Introduction Both Finland and Estonia belong to the Scandinavian area surrounding the Baltic Sea, which is regarded to be selenium- Selenium is known to be essential to activate various key deficient (Reimann et al., 2003). In contrast to Finland, enzymes, such as the antioxidant glutathione peroxidase, where selenium supplementation took place in 1984, the metabolic enzyme thioredoxin reductase and the thyroid Estonia, our nearest neighbour in south, is still ‘untouched’. hormone-activating enzyme iodothyronine deiodinase. For The objective of this study is to sum up the dietary humans, the range between poisonous and necessary selenium intake of the Estonian people according to the amounts of selenium is rather narrow. serum selenium concentrations. It has been shown that there is a connection between scarce intake of selenium, and heart diseases and cancer. First, it was proved that there is a connection between selenium and fatal Methods and subjects cardiomyopathy called Keshan disease (Keshan Disease Research Group, 1979). Later, some studies have provided As far as we know, no other studies have been carried out some evidence for the role of selenium in the etiology of concerning the amounts of selenium in Estonians apart from coronary diseases and cancer (Salonen et al., 1982; 1984; our study group’s research. In the 1990s, our research team Huttunen 1997). As in Finland, morbidity and mortality from collected several series of serum specimens from Estonians of cardiovascular diseases are also one of the central challenges different age and analyzed selenium concentrations in them. of national public health in Estonia (Polluste et al., 2005). The findings are collected in this study and compared with Selenium may also reduce cancer rates (Ip et al., 2000). results found in the literature about amounts of selenium in other countries, especially Finland. The serum samples were transported in dry ice, and Correspondence: Dr M Kantola, Finnish Forest Research Institute, P.O.Box 18, selenium (Se) concentrations were determined, after storing Vantaa, Fin-01301, Finland. for short time in the freezer (À181C), in two laboratories: at E-mail: [email protected] Received 8 November 2006; revised 26 March 2007; accepted 23 April 2007; the Department of Chemistry, Kuopio University and published online 30 May 2007 Department of Public Health, Helsinki University. Both Selenium levels of Estonians P Rauhamaa et al 1076 laboratories used the same method: flameless atomic menopause, and living in Haapsalu and Narva, Estonia. absorption spectrometry (Perkin-Elmer 5000; HGA-400, Sampling locations are shown in Figure 1. Norwalk, CT, USA) with nickel nitrate as the matrix modifier (Alfthan and Kumpulainen, 1982, Kantola et al., 1988). Accuracy was checked with the external commercial stan- Statistics dard reference material Moni-Trol (Dade, Miami, FL, USA) and Seronorm Trace Element Serum (Batch 311089). The The results have been calculated using the SPSS for Windows, inter-assay variation in both the laboratories was within the release 9.0 statistical package (Chicago, IL, USA). The results confidence limit of international reference materials, show- are represented as means, and significance of the differences ing that the results of these laboratories were both liable. The between the means was tested by the non-parametric Mann– quantification limit for selenium was 15 mg/l. Both labora- Whitney test. tories have participated successfully in national and international external quality assessments for selenium determinations: The Trace Element Quality Assessment Scheme, University of Surrey, UK and inter-laboratory trial Ethical aspects organized by the Faculty of Agriculture and Forest Sciences in the University of Helsinki. In addition, the validity of Written consent was obtained from all the pregnant women. selenium determinations was checked annually during the Ethical permission for the studies was obtained from the study time by participating in quality assessment organized ethics committees of the National Public Health Institute by the Labquality Company (Kliinisten Laboratoriotutki- and Helsinki University Department of Public Health (The musten Laaduntarkkailu Oy). National Public Health Institute, Kuopio University and Kuopio University Hospital (14/1995, 16/1996, 3/1996)). In

Collection of Samples

The study collection contained five substudies. For study Finland group 1, serum samples were collected in 1993 and 1994 Narva from 60 students of the University of Tartu. The students are from different parts of Estonia. For study group 2, serum Rakvere specimens were collected from 1992 to 1994 from pregnant Baltic Sea Tallinn Estonian women. Part of the results has been published Russia before (Kantola et al., 1997). The specimens were collected in Estonia Estonia from two different regions: Tallinn and Rakvere. For Haapsalu study group 3, 34 serum samples were collected in 1998 from Tartu men and postmenopausal women with no severe diseases in the hospital of Rakvere (Seppa¨nen et al., 2000). For study group 4, serum samples were drawn from 80 young sportsmen and 36 controls from different regions of Estonia in the Latvia beginning of the 1990s. The sampling took place in Tartu. For study group 5, serum samples were collected during 2000 and 2001 from 126 women who were mostly near the age of Figure 1 Location of the places where the samples were collected.

Table 1 Concentrations of selenium in sera of Estonians

Tartu Tallinn Rakvere Haapsalu Narva Other places in Estonia

Object Young male Young female Women Men Women Pregnant Women Women Men students students women Years of sampling 1993–1994 1994–1994 1992–1997 1997 1997 1992–1997 2000–2001 2000–2001 1991 N 10 50 11 20 14 57 44 82 116 Age (years) Mean 19.5 19 ND ND ND ND 45.8 52 21.9 Range 18–33 19–23 ND 20–69 20–69 ND 26–56 28–69 16–31 Se concentration (mg/l) Mean 77 83 48.2 64.7 66 63 64 81 59.3 Range 69–98 57–111 38–58 48–83 57–87 31–104 42–104 40.5–131 26–116

Abbreviations: N, number of Estonians analyzed in the study; ND, not declared.

European Journal of Clinical Nutrition Selenium levels of Estonians P Rauhamaa et al 1077 Estonia, the permissions were obtained from the hospitals may need to be greater than that of World Health Organiza- and the students. tion, because full expression of selenoprotein P requires a greater selenium intake than does full expression of plasma glutathione peroxidase (Xia et al., 2005). Results Before the selenium supplementation (in 1984), the Se levels in sera of the Finns ranged from 65 to 70 mg/l (Varo The concentrations of the selenium in the sera of Estonians et al., 1988; Mussalo-Rauhamaa and Lehto, 1989). According are compiled in Table 1. The mean selenium concentration to Alfthan (2005), the mean of whole blood Se content then in sera of all 404 Estonians analyzed in this study was 75 mg/l was 91 mg/l, and is currently 178 mg/l. After the supplementa- and the range was from 26 to 116 mg/l. tion, Se levels increased quickly almost 50% in Finland. The mean Se concentrations in serum of young students Whole blood Se reached its peak value approximately 1 year studying in Tartu (study group 1) were 77 and 83 mg/l for later than serum Se. male and female students, respectively. No statistical differ- In this study, we found that in Estonia, the serum Se ences were found between sexes. In addition, either donor’s concentrations seem to mostly depend on the geographical height, weight or body mass index did not correlate with location of the study site. In Tallinn and Haapsalu in the selenium content. northwestern Estonia, the Se levels in pregnant women were In Rakvere, the mean concentration of Se in the serum lower than in Narva and Rakvere in the east and middle of samples of 57 mothers (study group 2) was 63 mg/l. In Estonia. The difference was statistically significant among Tallinn, the mean concentration in the samples of 11 the study groups, whose serum samples were collected mothers (study group 2) was 48.2 mg/l (Kantola et al., 1997). between 2000 and 2001 in Haapsalu and Narva, Se Se contents in Tallinn were statistically, significantly lower concentrations being higher in Narva. In Tallinn, Se than in Rakvere (P ¼ 0.007). concentrations were also statistically significantly lower The study population in Rakvere consisted of 20 men and than in Rakvere (Kantola et al., 1997). In that study, the 14 women (study group 3). No statistical differences in Se mean Se concentration for the 118 Finnish pregnant women contents were found between men and women. The mean Se living in Kuopio during the same period (1992–1994) was concentrations were 65 and 66 mg/l for these men and 91 mg/l with the range of 71–163 mg/l. The effect of the Se women, respectively. The serum concentrations in active supplementation is already notable in the values above. and recreational sportsmen, and inactive controls (study In the follow-up study of the project Cardiovascular Risk in group 4) sampled in Tartu did not differ. The mean Young Finns, the mean serum Se concentrations of 288 concentration was 59.3 mg/l. Finnish adolescents aged 18 years was 78 mg/l before the Se The mean concentrations of selenium in samples of supplementation and 108 mg/l (105 adolescents) after the women aged 26–69 years (study group 5) living in Haapsalu addition of Se to fertilizers (Mussalo-Rauhamaa et al., 1993). and Narva were 64 and 81 mg/l, respectively. A significant In Estonian adolescents, Se levels were similar to those of difference in Se levels was noticed; the Se levels in Narva Finnish adolescents before the supplementation. were about 27% higher than in Haapsalu (P ¼ 0.001).

Reasons for low serum Se concentrations in Estonia Discussion Low median Se concentrations (0.11 mg/kg) have been reported in the ploughing layer (0–25 cm) in Estonia, while The serum Se concentrations have been widely studied in Finland, the median varies from 0.15 mg/kg to more than around the world (Alfthan and Neve, 1996). In 1990s, the 0.37 mg/kg (Reimann et al., 2003). Kevvai (1994) measured lowest serum Se values in Europe were found in Serbia and the Se content in 13 different Estonian soils and found a Bulgaria, and the highest in Norway. Intake of Se in foodstuff mean value of 0172 mg/kg (range 0010–0443 mg/kg). Malbe has been exceptionally scanty in Finland and New Zealand. et al. (2003) showed that there was a connection between the In view of the plasma glutathione peroxidase activities, the Se deficiency and the plant–animal chain of obtaining average normative requirement of Se for most adult males is selenium. The knowledge of the Se status of Estonian food estimated to be 40 mg/day. The upper limit of the safe range chains is incomplete. In 1997, Pehrson et al. analyzed a few of adult population mean intake of Se has been defined as samples of blood and milk from Estonian cows and found 400 mg/day for both men and women. For pregnant women, that they were low in Se. Vikerpuur (1995) also reported the basal population requirement for Se has been set to similar results. In 1993, Suoranta et al. also recorded very low 18 mg/day. The normative individual and population require- Se concentration in feedstuffs, milk and tissues from one ments for Se are based on the results from the United States, Estonian dairy farm. and are 27 and 39 mg/day, respectively (WHO, 1996). It has There are very few results about the Se contents of food in also been suggested that selenoprotein P is a better indicator Estonia. Malbe et al. (1995) showed that the Se levels of of selenium nutritional status than glutathione peroxidase, the cows were extremely low by studying the dairy cows and thus the recommended dietary allowance of selenium of Estonian Agricultural University. The levels of Se were

European Journal of Clinical Nutrition Selenium levels of Estonians P Rauhamaa et al 1078 5.6 mg/l in whole blood and 3.2 mg/l in milk. It was shown Huttunen JK (1997). Selenium and cardiovascular disease – an that there is a connection between selenium intake and update. Biomed Environ Sci 10, 220–226. In vitro in vivo content of selenium in human sera (Aro et al., 1995). This as Ip C, Thompson HJ, Zhu Z, Ganther HE (2000). and studies of methylseleninic acid: evidence that a monomethylated well as the Se levels of domestic foodstuff supports our selenium metabolite is critical for cancer chemoprevention. Cancer findings that the intake of Se in Estonia may not be optimal. Res 60, 2882–2886. Theoretically, the somewhat better intake of Se of the Kantola M, Saaranen M, Vanha-Perttula T (1988). Selenium and glutathione peroxidase in seminal plasma of men and bulls. Estonians compared with that of the Finns before the J Reprod Fertil 83, 785–794. fertilization with selenium may be explained by two facts: Kantola M, Mand E, Viitak A, Juraskaja J, Purkunen R, Vartiainen T the bedrock of northern Estonia consists of younger et al. (1997). Selenium content of serum and human milk from sedimentary rock type and its pH is higher than in Finland, Finland and neighbouring countries. J Trace Elem Exp Med 10, 225–232. where the bedrock consists mainly of acidic rock types. This Keshan Disease Research Group (1979). Epidemiologic studies on the as well as the climate and high iron content of the soil in etiologic relationship of selenium and Keshan disease. Chin Med J Finland contributes to the low bioavailability of Se for plants. 92, 477–482. The findings of this research also show that the Se levels of Kevvai T (1994). Seleeni u¨ldsisaldusest mo˜nedes Eesti po˜llumuldade profiilides. In: Eesti po˜llumajandusu¨likool: Magistrandide ja Doctor- Estonians are noticeably lower than that of the Europeans in antide teaduslike to¨o¨de, Kogumik II. Tarto: Estonia. pp. 72–78. general. The Se levels found are also somewhat lower than in Malbe M, Klaassen E, Kaartinen L, Attila M, Atroshi F (2003). Effects the neighbouring countries, such as Lithuania (mean Se level of oral selenium supplementation on mastitis markers and in serum, 90 mg/l) (Golubkina et al., 1992) or the western pathogens in Estonian cows. Vet Ther 4, 145–154. Malbe M, Klaassen M, Fang W, Myllys V, Vikerpuur M, Nyholm K regions of Russia (Pskov 0.80 mmol/l (63 mg/l), Novgorod et al. (1995). Comparisons of selenite and selenium yeast feed 1.04 mmol/l (82 mg/l)) (Golubkina and Alfthan, 1999). supplements on Se-incorporation, mastitis and leucocyte function in Se-deficient dairy cows. Zentralblatt Veterinarmedizin- Reihe A 42, 111–121. Mussalo-Rauhamaa H, Lehto J (1989). Selenium addition to fertilizers Conclusion effectively increased the serum levels of this element in the Finnish population. J Am Coll Nutr 8, 588–590. Mussalo-Rauhamaa H, Vuori E, Lehto JJ, A˚ kerblom H, Ra¨sa¨nen L This study suggests that the selenium intake of Estonians has (1993). Increase in serum selenium levels in Finnish children and been quite scarce (both in adolescents and adults), near the young adults during 1980–1986: a correlation between the serum levels that the Finns had before the selenium supplementa- levels and the estimated intake. Eur J Clin Nutr 47, 711–717. tion of fertilizers. There seems to be differences within the Pehrson B, Ling K, Ortman K (1997). The selenium status of dairy cattle in Estonia. Acta Vet Scand 38, 353–356. country, highest Se concentrations being in the eastern part Polluste K, Mannik G, Axelsson R (2005). Public health reforms in of Estonia. Concentrations of Se in the sera of Estonians are Estonia: impact on the health of the population. Br Med J 23, somewhat lower or at a similar level as in the populations of 210–213. the neighbouring countries in the Baltic area. Estonian Reimann C, Siewers U, Tarvainen T, Bityukova L, Eriksson J, Gilucis A et al. (2003): E. Schweitzerbart’sche Verlagsbuchhandlung: Stutt- authorities should consider in their nutrition policies the gart, Germany 279 pages. possible low intake of selenium of Estonians. Salonen JT, Alfthan G, Huttunen JK, Puska P (1982). Association between cardiovascular death and myocardiol infarction and serum selenium in a matched-pair longitudinal study. Lancet 2, 175–179. References Salonen JT, Alfthan G, Huttunen JK, Pikkarainen J, Puska P (1984). Association between serum selenium and the risk of cancer. Am J Alfthan G (2005). Trends in blood and tissue selenium levels in Epidemiol 120, 342–349. Finland 1984–2004. In: M Eurola (ed). Proceedings Twenty Years of Seppa¨nen K, Kantola M, Laatikainen R, Nyysso¨nen K, Valkonen V-P, Selenium Fertilization, 8–9 September, 2005. Helsinki: Finland, pp. Kaarlo¨pp V et al. (2000). Effect of supplementation with organic 71–74. Agrifood Research Reports 69. selenium on mercury status as measured by mercury in pubic hair. Alfthan G, Kumpulainen J (1982). Determination of selenium in J Trace Elem Med Biol 14, 84–87. small volumes of blood plasma and serum by electrothermal Suoranta K, Sinda E, Pihlak R (1993). Selenium of the selenium yeast atomic absorption spectrometry. Anal Chim Acta 140, 221–227. enters the cow’s milk. Norw J Agric Sci 1443, 215–216. Alfthan G, Neve J (1996). Reference values for serum selenium in Varo P, Afthan G, Ekholm P, Aro A, Koivistoinen P (1988). Selenium various areas evaluated according to the TRACY protocol. J Trace intake and serum selenium in Finland: effects of soil fertilization Elem Med Biol 10, 77–87. with selenium. Am J Clin Nutr 48, 324–329. Aro A, Alfthan G, Varo P (1995). Effects of supplementation of Vikerpuur M (1995). Effect of selenium supplementation on Estonian fertilizers on human selenium status in Finland. Analyst 120, dairy cows with special reference to udder health and host- 841–843. microbe balance in mastitis. Thesis. Saarija¨rvi: Finland. Golubkina NA, Alfthan G (1999). The human selenium status in 27 WHO (1996). Trace elements in human nutrition and health. World regions of Russia. J Trace Elem Med Biol 13, 15–20. Health Organisation: Geneva. Golubkina NA, Shagova MV, Spirichev VB, Alfhan J, Laaksonen P, Xia Y, Hill K, Byrne D, Xu J, Burk R (2005). Effectiveness of selenium Kumpulainen I et al. (1992). Selenium intake by the population of supplements in a low-selenium area of China. Am J Clin Nutr 81, Lithuania. Voprosy Pitaniia 1, 35–37 (in Russian). 829–834.

European Journal of Clinical Nutrition