Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
European Journal of Clinical Nutrition (1998) 52, 529±535 ß 1998 Stockton Press. All rights reserved 0954±3007/98 $12.00 http://www.stockton-press.co.uk/ejcn The serum selenium concentrations in children and young adults: a long-term study during the Finnish selenium fertilization programme W-C Wang1,5, A-L MaÈkelaÈ2, V NaÈntoÈ1,5, P MaÈkelaÈ3, and H LagstroÈm4, Departments of 1Clinical Chemistry and 2Pediatrics, 3Department of Nuclear Medicine, Turku University Central Hospital; 4the Cardiorespiratory Research Unit, University of Turku; and 5the Joint Clinical Biochemistry Laboratory of the University of Turku, Turku University Central Hospital and Wallac Ltd., FIN-20520 Turku, Finland Objective: To investigate the effects of the Finnish nationwide selenium (Se) fertilization programme on the Se status of the population. Design and subjects: Serum Se concentrations from 1985±1992 from 1568 healthy children and young adults in southwestern Finland were determined using direct electrothermal atomic absorption spectrometry. Results: The mean concentration in young adults increased from 1.04 mmol/L in 1985 to 1.59 mmol/L in 1990. Children younger than 15 y had lower concentrations than adults, with an increase from 0.87 mmol/L in 1985 to 1.31 mmol/L in 1990. The younger the children, the lower the Se concentrations tended to be. At the age of about seven months no signi®cant difference was noted between breast-fed and formula-fed infants. From 1991, when the amount of Se added to fertilizers was reduced and less foreign high-Se cereal was imported, the Se concentrations decreased in all age groups. Conclusions: The nationwide Se supplementation programme has succeeded in elevating the Se intake and the serum Se concentrations in the Finnish population. Sponsorship: Supported by the Juho Vainio Foundation and the Sigrid Juselius Foundation, Finland. Descriptors: serum selenium; age-dependency; selenium fertilization Introduction Subjects One thousand, ®ve hundred and sixty-eight apparently healthy people (716 males and 852 females, altogether 2143 serum samples) were investigated during this series of studies (MaÈkelaÈ et al, 1993, 1995; Wang et al, 1995a,b,c). Of them 1081 are children, with a mean age of 4.3 y (s.d. 4.0 y, range 0.5 ±14.7 y) and 487 young adults, with a mean age of 24.3 y (s.d. 3.4 y, range 15 ±30 y). Two hundred and fourteen were healthy siblings of children suffering from insulin-dependent diabetes mellitus or from juvenile rheumatoid arthritis (Wang et al, 1995b,c). In addition, we were given access to serum specimens, which had been collected in 1990 ±1992 for determination of serum lipids in a project on atherosclerosis risk factors in children at the Cardiorespiratory Research Unit (CRU), University of Turku (Lapinleimu et al, 1995). Of the 1290 subjects 867 were children under 15 y of age (from six months to 14.7 y) and 188 were their adult family members (age range from 19.7±30 y). There was no diet restriction for the subjects. Three groups of medical students were included in the study: 108 students (70 females and 38 males, ages from 18.6 ±29.4 y) from 1985±1990, and 35 students (31 females and 4 males, ages from 21.3±29.4 y) in 1991 (MaÈkelaÈ et al, 1993). In 1992, 65 students (46 females, 19 males, ages from 23.1±29.6 y) were investigated. Other adults (91 people) were members of the staff of the clinic. If more than one sample was taken from same person in the same year, the mean value of these samples was calculated and used when calculating the mean values presented in the tables. The study was approved by the Commission on Ethics of the University of Turku. Informed consent was obtained from all adults or all parents of the children. Correspondence: Dr W-C Wang, Department of Clinical Chemistry, University of Turku, FIN-20500 Turku, Finland. Received 6 February 1998; revised 10 March 1998; accepted 31 March 1998 Methods The blood samples for the serum Se analysis were collected into Vacutainer1 tubes (Becton Dickinson, Rutherford, NJ, Finland is for geochemical reasons a low-Se area and, consequently, the daily dietary intake of Se in the Finnish population has been one of the lowest in the world. In 1984 an of®cial decision was made to increase the intake of Se in the population by adding Se to multimineral agricultural fertilizers (Ministry of Agriculture and Forestry, 1984). Fertilizers used for the production of hay and fodder were supplemented with 6 mg=kg of Se in the form of sodium selenate, and fertilizers for cereal were supplemented with 16 mg=kg. As a result, Se content increased in agricultural products, in dairy products and in meat (Varo et al, 1988; 1994; Eurola et al, 1989, 1991; Ekholm et al, 1991). An increase followed in the Se intake and resulted in higher serum Se concentration in the whole population. In 1991 the amount of Se added to the fertilizers was reduced to the level of 6 mg=kg for all fertilizers (Ministry of Agriculture and Forestry, 1990). We report here observations on the Se status in healthy children and young adults from the studies of the effects of the Se fertilization programme in southwestern Finland. Materials and Methods Effects of selenium fertilization in Finland W-C Wang et al 530 USA). After centrifugation, the serum was transferred into acid-washed polyethylene tubes, which were frozen immediately and stored at 720 C until analysis (Lakomaa et al, 1988). The serum samples from the risk factor project had been stored frozen at 720 C. The concentration of Se in serum was determined by direct electrothermal atomic absorption spectrometry, according to the method of Alfthan & Kumpulainen (1982), using a Zeeman=3030 spectrometer (PerkinÈ berlingen, Germany) with an auto sampler (ASElmer, U 60, Perkin-Elmer) and a printer (PR-100, Perkin Elmer). Internal quality control was carried out with Seronorm human serum reference controls (Nycomed Pharma AS, Oslo, Norway). The within-assay relative coef®cient of variation was 5.4% at a level of 0.82 mmol=L, 4.3% at 1.02 mmol=L, and 5.0% at 1.30 mmol=L. The inter-assay coef®cient of variation was 5.5% at a level of 0.82 mmol=L, 4.3% at 1.02 mmol=L, and 4.9% at 1.30 mmol=L. The recovery rates ranged from 102±95.1% at three Se levels of 0.50 ±2.00 mmol=L. Interlaboratory calibration using serum samples was carried out with National Public Health Institute, Helsinki, to con®rm Se analyses. In the risk factor project group food consumption and Se intake of children aged 7 and 13 months had been evaluated using 3-consecutive-day food records. The parents were carefully taught to record the child's food consumption and they were also given written instructions of how to ®ll the food consumption records with exact descriptions and amounts of all foods and drinks consumed. Amounts were estimated using household measures (glasses, cups, tablespoons, slices, deciliters, etc.). The parents or day-care personnel recorded the type, brand and preparation of all foods in detail. The records were reviewed, item by item, by the dietitians for completeness and accuracy during the follow-up visit. Missing portion sizes, food descriptions, etc., were added if needed. Daily energy and Se intakes were calculated as means of the recorded days for each child. Intake of nutrients from supplements was not included in these calculations. The Micro NUTRICA programme (Knuts et al, 1987), developed at the Research Centre of the Social Insurance Institution, Turku, Finland was used for the calculation. The programme uses the Food and Nutrient Data Base of the Social Insurance Institution (Rastas et al, 1993) and it calculates 63 nutrients of commonly used food stuffs and dishes in Finland. statistical signi®cance of the differences among the groups was tested by analysis of variance using SAS=STAT software (Statistical Analysis Systems, 1985). The Bonferroni method was used to test the signi®cance of changes within the groups (Wallenstein et al, 1980). The differences were considered signi®cant when the P values were less than or equal to 5%. Results Table 1 shows the number of subjects as well as the number of collected serum samples and the mean serum Se concentrations and ranges for each year. The number of samples varies from year to year because the samples were taken from different projects in different years. The high number of samples in 1990 ±1991 is because a lot of samples were taken from the large material of the project of atheriosclerosis risk factors. Serum Se concentrations for subjects of different ages during the study are shown in Figures 1 and 2. The most notable changes in Se values were seen during the initial phase of the Se fertilization programme, between 1985 and 1986. The increase was highly signi®cant (P < 0.001). In 1985 the lowest measured Se value was 0.54 mmol=L, observed in a nine-year old girl, in the following years such low values were not found any more. At the start of the Se supplementation, in 1985, children younger than 15 y had signi®cantly (P < 0.001) lower mean serum Se than young adults (Table 2). The full effect of the supplementation was observed during the following year, in 1986, when the mean value was 1.26 mmol=L in children and 1.30 mmol=L in young adults. The serum Se concentrations continued to increase, reaching the highest value in 1990 (mean 1.59 mmol=L, range 1.08±2.19 mmol=L) in adults. Some increase was also documented in children younger than 15 y until a mean value of 1.31 mmol=L (range 0.82±1.99 mmol=L) was reached in 1990. The increase was signi®cantly smaller in children than in the older age group (P < 0.001), and the difference between the two groups remained. In 1991±1992 a slight decrease in the serum Se values was documented in both age groups, which obviously was due to the reduction in the amount of Se added to fertilizers in 1991. Another factor which in addition to the supplementation affected the Se intake was the import of foreign cereal with a high Se content (Varo et al, 1988). Figure 3 presents the annual import of foreign cereal together with the serum Se values in children and adults. In 1988 and 1989 the amount of foreign cereal was high and may also have contributed to the increase in the serum Se in the adults up to 1990. Statistical analyses The Student's t-test was used for testing the statistical signi®cance of the differences between two groups. The Table 1 The number of subjects and serum concentrations (mmol=L, mean and range) in 1985 ±1992 Males Females Total Year na nb na nb na nb 1985 1986 1987 1988 1989 1990 1991 1992 119 56 16 29 39 281 260 46 129 60 20 44 65 284 260 48 179 104 19 46 77 367 250 86 193 115 23 67 116 378 250 91 298 160 35 75 116 648 510 132 322 175 43 111 181 662 510 139 Total a Number of subjects. Number of samples. b 910 1233 2143 Se 0.97 1.29 1.26 1.34 1.46 1.43 1.32 1.35 (0.54±1.44) (0.87±1.72) (1.01±1.60) (1.04±1.80) (0.99±2.10) (0.82±2.19) (0.72±2.13) (0.79±1.88) Effects of selenium fertilization in Finland W-C Wang et al 531 Figure 1 Serum selenium concentrations of healthy Finnish children and young adults in 1985±1988. Figure 2 Serum selenium concentrations of healthy Finnish children and young adults in 1989±1992. Table 2 Serum selenium values (mmol=L, mean and range) in healthy children (<15 y) and young adults (>15 y) in different years Year na nb 1985 1986 1987 1988 1989 1990 1991 1992 119 37 29 41 36 372 473 48 140 52 37 72 72 372 473 48 a Number of subjects. Number of samples. c Not signi®cant. b Children 0.87 1.26 1.26 1.29 1.32 1.31 1.30 1.22 (0.54±1.44) (0.96±1.57) (1.01±1.60) (1.04±1.70) (1.09±1.75) (0.82±1.99) (0.72±1.77) (0.79±1.52) na nb 179 123 6 34 80 276 37 84 182 123 6 39 109 290 37 91 1.04 1.30 1.29 1.40 1.52 1.59 1.57 1.42 Adults P value (0.62±1.35) (0.87±1.72) (1.16±1.41) (1.20±1.80) (0.99±2.10) (1.08±2.19) (1.13±2.13) (1.06±1.88) <0.001 NSc NSc <0.001 <0.001 <0.001 <0.001 <0.001 Effects of selenium fertilization in Finland W-C Wang et al 532 However, in children serum Se concentrations remained at the level reached in 1988. In Figure 3 the serum values are also presented separately in males and females. A more detailed analysis shows that the differences are only marginal between the two genders even if the results are presented separately in the groups of adults and of children. An analysis of serum Se for children in the different age groups showed that the younger the children, the lower the serum Se concentrations are for example, in 1990 children younger than 3 y (n 259 subjects) had a mean serum Se value of 1.25 mmol=L (range 0.82±1.76 mmol=L) and children, ages 3±15 y (n 113 subjects), had a mean serum Se concentration of 1.46 mmol=L (range 1.11±1.99 mmol=L). The difference between the groups was highly signi®cant (P < 0.001). Moreover, the difference has existed during 1985±1992. In 1990, at the age of about seven months, 66 children were breast-fed and 83 were formula-fed. Of the formulafed infants 14.3% were given `Piltti' (Van den Bergh, Lapinlahti, Finland), 5.4% received `Bona' (Chymos- Figure 3 Serum selenium concentrations in 1985±1992 as compared with the yearly amount of imported Se-rich cereals (information from the Finnish Grain Board, Helsinki). Effects of selenium fertilization in Finland W-C Wang et al Table 3 533 Estimated selenium intake of infants Age Year (months) na 1990 1990 1991 1992 7 13 13 13 56 78 422 44 Energy intake (kcal=d) Mean (s.d., range) 807 952 981 998 (150, (169, (172, (225, 282±1128) 688±1428) 321±1553) 469±1635) Se intake (mg=d) Mean (s.d., range) PRI of Sea (mg=d) 6.4 (4.9, 0.1±23.8) 16.4 (7.7, 1.2±41.8) 15.8 (8.5, 1.2±46.9) 16.0 (9.3, 2.3±48.8) 8 10 10 10 Serum Se (mmol=L) Mean (s.d., range) 1.17 1.33 1.29 1.22 (0.15, (0.14, (0.15, (0.15, 0.91±1.49) 1.03±1.63) 0.72±1.77) 0.79±1.52) a Number of infants. PRI the population reference intake, recommended by the Scienti®c Committee for Food of the European Community. b Nestle, Lappeenranta, Finland) and 80.3% were given `Tutteli' (Valio, Turenki, Finland). In this study no signi®cant difference was found between the serum Se concentrations of breast-fed and formula-fed infants. The mean serum Se value in the breast-fed infants was 1.18 mmol=L (s.d. 0.19 mmol=L, range 0.82±1.76 mmol=L) and that in the formula-fed was 1.16 mmol=L (s.d. 0.14 mmol=L, range 0.89±1.57 mmol=L). From the formula-fed infants at the age of seven months, complete food records were available from 56 children (Table 3). According to food records, the mean calculated daily intake of Se in this group of children was 6.4 mg (s.d. 4.9 mg). The range extended from 0.1±23.8 mg of Se per day in individual daily food records, indicating large differences in the intake. Se serum concentrations of these 56 children, however, were on an adequate level, with the mean value being 1.17 mmol=L (Table 3). In the age group of 13 months, food records for children and the calculated average daily Se intake were shown in Table 3. The range of intake for individual children was very wide extending from 1.2 mg± 48.8 mg of Se per day. On the other hand, in these groups of children the Se serum concentrations were also on adequate levels, the mean values being in 1990 1.33 mmol=L, in 1991 1.29 mmol=L and in 1992 1.22 mmol=L (Table 3). Discussion Serum Se concentrations between 0.51 and 1.79 mmol=L have been reported for European populations (for example, Kostakopoulos et al, 1990; Ringstad et al, 1991). In the 1970s the serum Se concentrations in the Finnish population ranged from 0.63±0.76 mmol=L (Alfthan, 1988; Knekt et al, 1990) being among the lowest ones in Europe. In the mid-1970s, the dietary intake of Se in an average Finnish diet was about 25 mg per 10 MJ (or 2400 kcal) (Mutanen & Koivistoinen, 1983). In the early 1980s, when high-Se grain (200 ±500 mg=kg) was imported from North America, the daily Se intake was 40 ±50 mg (Varo et al, 1988). The value was at an adequate level compared with the average requirement of 40 mg recommended by the Scienti®c Committee for Food of the European Community (1993). About two years after the start of the Se-supplemented fertilization programme, in 1986, the mean daily Se intake of the Finnish population was estimated to be about 80±90 mg=10 MJ (Varo et al, 1988). A level of 110±120 mg=10 MJ was reached in 1987 and remained constant until 1990 (Varo et al, 1994). The increase in the serum Se values in the Finnish subjects re¯ected a good response to the use of Se-enriched agricultural fertilizers, but as mentioned above, it also re¯ected the increased amount of imported high-Se grain in 1988±1989 (MaÈkelaÈ et al, 1993). In our study, the nutrient composition of the commonly used baby formulas and commercial infant foods were added to the database of Micro NUTRICA programme for use in these analyses. The Se content of Finnish foods has increased considerably since the addition of Se into the fertilizers was started in 1985. If the Se content for a certain food was analysed after the Se supplementation of the fertilizers, this value has been used in the programme (Ekholm et al, 1991; Eurola et al, 1989, 1991; Varo et al, 1994). In other cases earlier values are given dating back to the middle of the 1970s and many of these values, for example, for ®sh products, mushrooms, wild berries and imported vegetables and fruits, remain valid today (Varo, 1981). From 1991 on the amount of Se added to the agricultural fertilizers was reduced to the level of 6 mg=kg for all fertilizers (Ministry of Agriculture and Forestry, 1990), and the import of foreign high-Se grain was considerably reduced at the same time (MaÈkelaÈ et al, 1993). In 1992 the Se intake was estimated to be 80±90 mg=10 MJ (Varo et al, 1994). According to the results of our study, in 1991±1992, the mean serum Se concentrations decreased in all age groups as a response to the decreased dietary intake of Se. The results of this study show that the changes in the serum Se values in the population, as well in adults as in children of all ages, re¯ect alterations in the mean daily dietary intake of Se. Children younger than three years had the lowest concentrations of Se during the whole eight-year observation period. Lombeck et al, (1977) had earlier reported that the serum Se concentrations during childhood were age-dependent. In their study, the healthy children under the age of 3 y had signi®cantly lower serum Se concentrations than the children in older age groups. During the Se fertilization programme no Se has been added to any infant formulas in Finland. According to the manufacturers, the Se content of `Piltti' in 1990 ±1991 was 10 mg=L, that of `Bona' 8.8 mg=L in 1990 ±1991 and the Se content of `Tutteli' in 1991±1992 was 10 ±11 mg=L. The Se content of breast milk in Finland in 1990 has ranged from 14.6 ±17.2 mg=L (0.18 ± 0.22 mmol=L) as determined from pooled samples received from breast-milk centers (Kantola & Vartiainen, 1991). As calculated by food consumption records, however, the mean dietary daily intake of Se in infants aged 13 months was higher than the population reference intake (PRI) recommended by the Scienti®c Committee for Food of the European Community in those years, when the Se intake was at the highest level in general, namely in the years 1990 ±1991. Although the mean dietary daily intake of Se in infants aged seven months in 1990 was lower than the PRI, the Se serum concentrations of these infants were still at an adequate level. Effects of selenium fertilization in Finland W-C Wang et al 534 The mean breast milk concentration of Se (when infants were at the age of 16 weeks) in Germany (Dorner et al, 1990) was reported to be 17.6 mg=L, being similar compared with that in this study. Robberecht et al (1985) reported that the Se content of breast milk (two months post partum) in Belgium was 10.2 mg=L and the estimated daily intake of Se for Belgian infants of three months of age was 8.1 mg=L (boys) and 7.1 mg=L (girls). New Zealand is also recognized as a Se-poor country. Dolamore et al (1992) reported that a mean Se concentration of breast milk in New Zealand was 13.4 mg=L and a mean plasma Se concentration of infants less than 12 months old was 33 mg=L (0.42 mmol=L). Both of the Se concentrations were lower than those in this study. Interesting is also, that still in 1990, the Se content of the breast milk of Finnish women did not reach those values reported from USA (Litov et al, 1989) or from Japan (Hatano et al, 1985). The Se content of mature breast milk in USA from the Se-rich area of Salt Lake City is reported to be 25 mg=L (USA (Litov et al, 1989). In Japan the Se content of breast milk is reported to be 20 mg=L (Hatano et al, 1985). When comparing the Se serum concentrations of young infants in our study to those published, for example, from USA (Litov et al, 1989), it appears, that they do not differ markedly from each other. In USA the plasma Se concentration of infants at two months of age is reported to be 1.2 mmol=L (s.d., 0.15 mmol=L) (Litov et al, 1989). Conclusions The nationwide Se supplementation programme has succeeded in elevating the Se intake and the serum Se concentrations in the Finnish population. However, well planned long-term follow-up studies connected with regular measurements of Se serum concentrations are needed in order to reveal the possible health effects of increased Se intake as well as the optimal amount of Se, which should be added to the fertilizers in Finland to keep the Se intake of the population at an adequate level. AcknowledgementsÐThe analyses were carried out at the Joint Clinical Biochemistry Research Laboratory of the University of Turku, Turku University Central Hospital and Wallac Ltd., Turku. We thank Dr M HaÈmaÈlaÈinen, Ph.D., for the technical assistance during this work and Dr Else-Maj Suolinna, Ph.D., who revised the English language. This study was supported by the Juho Vainio Foundation and the Sigrid Juselius Foundation, Finland. References Alfthan G (1988): Longitudinal study on the selenium status of healthy adults in Finland during 1975 ± 1984. Nutr. Res. 8, 467 ± 476. Alfthan G & Kumpulainen J (1982): Determination of selenium in small volumes of blood plasma and serum by electrothermal atomic absorption spectrometry. Anal. Chim. Acta 140, 221 ± 227. Dolamore BA, Brown J, Darlow BA, George PM, Sluis KB & Winterbourn CC (1982): Selenium status of Christchurch infants and the effect of diet. N. Z. Med. J. 105, 139 ± 142. Dorner K, Schneider K, Sievers E, Schulz-Lell G, Oldigs HD & Schaub J (1990): Selenium balances in young infants fed on breast milk and adapted cow's milk formula. J. Trace Elem. Electrolytes Health Dis. 4, 37 ± 40. Ekholm P, Varo P, Aspila P, Koivistoinen P & SyrjaÈlaÈ-Qvist L (1991): Transport of feed selenium to different tissues of bulls. Br. J. Nutr. 66, 49 ± 55. Eurola M, Ekholm P, Ylinen M, Koivistoinen P & Varo P (1989): Effects of selenium fertilization on the selenium content of selected Finnish fruits and vegetables. Acta Agric. Scand. 39, 345 ± 350. Eurola MH, Ekholm PI, Ylinen ME, Koivistoinen PE & Varo PT (1991): Selenium in Finnish foods after beginning the use of selenate supplemented fertilizers. J. Sci. Food Agric. 56, 57 ± 70. Hatano S, Aihara K, Nishi Y & Usui T (1985): Trace elements (copper, zinc, manganese, and selenium) in plasma and erythrocytes in relation to dietary intake during infancy. J. Pediatr. Gastroenterol. Nutr. 4, 87 ± 92. Kantola M & Vartiainen T (1991): Selenium content of breast milk in Finland after fertilization of soil with selenium. J. Trace Elem. Electrolytes Health Dis. 5, 283 ± 284. Knekt P, Aromaa A, Maatela J, Alfthan G, Aaran R-K, Hakama M, Hakulinen T, Petro R & Teppo L (1990): Serum selenium and subsequent risk of cancer among Finnish men and women. J. Natl. Cancer Inst. 82, 864 ± 868. Knuts LR, Rastas M, Haapala P, SeppaÈnen R, Karvetti RL & Aro A (1987): Nutrica Ð a computer programme for calculation of food and nutrient intake. In Abstract Book of the 3rd Eurofoods Meeting. Warsaw, Poland. Kostakopoulos A, Kostalos A, Alexopoulos J, Sofras F, Deliveliotis C & Kallistratos G (1990): Serum selenium levels in healthy adults and its changes in chronic renal failure. Inter. Urol. Nephrol. 22, 397 ± 401. Lakomaa EL, Mussalo-Rauhamaa H & Salmela S (1988): Avoidance of contamination in element analysis of serum samples. J. Trace Elem. Electrolytes Health Dis. 2, 37 ± 41. Lapinleimu H, Viikari J, Jokinen E, Salo P, Routi T, Leino A, RoÈnnemaa T, SeppaÈnen R, VaÈlimaÈki I & Simell O (1995): Prospective randomised trial in 1062 infants of diet low in saturated fat and cholesterol. Lancet 345, 471 ± 476. Litov RE, Sickles VS, Chan GM, Hargett IR & Cordano A (1989): Selenium status in term infants fed human milk or infant formula with or without added selenium. Nutr. Res. 9, 585 ± 596. Lombeck I, Kasperek K, Harbisch HD, Feinendegen LE & Bremer HJ (1977): The selenium state of healthy children. I. Serum selenium concentration at different ages; Activity of gluthione peroxidase of erythrocytes at different ages; Selenium content of food of infants. Eur. J. Pediatr. 125, 81 ± 88. Ministry of Agriculture and Forestry (1984): Proposal for amounts of selenium to be added into fertilizers. Working Group Report No. 7. (SeleenityoÈryhmaÈn ehdotus lannoitteisiin lisaÈttaÈvistaÈ seleenimaÈaÈristaÈ. Maa- Ja metsaÈtalousministerioÈ. TyoÈryhmaÈmuistio MMM 1984:7). Helsinki (in Finnish). Ministry of Agriculture and Forestry (1990): Annual report of Selenium Working Group. Working Group Report No. 28 (SeleenityoÈryhmaÈn raportti. IV vuosiraportti. Maa-ja metsaÈtalousministerioÈ. TyoÈryhmaÈmuistio MMM 1989:28). Helsinki (in Finnish). Mutanen M & Koivistoinen J (1983): The role of imported grain in the selenium intake of Finnish population in 1941 ± 1981. Inter. J. Vitamin Nutr. Res. 53, 102 ± 108. MaÈkelaÈ A-L, NaÈntoÈ V, MaÈkelaÈ P & Wang, W (1993): The effect of nationwide selenium enrichment of fertilizers on selenium status of healthy Finnish medical students living in south western Finland. Biol. Trace Elem. Res. 36, 151 ± 157. MaÈkelaÈ A-L, Wang W-C, HaÈmaÈlaÈinen M, NaÈntoÈ V, Laihonen P, Kotilainen H, Meg L-X & MaÈkelaÈ P (1995): Environmental effects of nationwide selenium fertilization in Finland. Biol. Trace Elem. Res. 47, 289 ± 298. Rastas M, SeppaÈnen R, Knuts L-R, Karvetti R-L & Varo P (1993): Nutrient composition of foods (Ruoka aineiden ravintoainesisaÈltoÈ). 4th edn. Helsinki: Publications of the Social Insurance Institution XXV (in Finnish). Ringstad J, Aaseth J, Johnson K, Utsi E & Thomassen Y (1991): High serum selenium concentrations in reindeer breeding Lappish men. Artic Med. Res. 50, 103 ± 106. Robberecht H, Roekens E, van Caillie-Bertrand M, Deelstra H & Clara R (1985): Longitudinal study of the selenium content in human breast milk in Belgium. Acta Pediatr. Scand. 74, 254 ± 258. Scienti®c Committee for Food of the European Community (1993): Proposed nutrient and energy intakes for the European Community. A report of the Scienti®c Committee for Food of the European Community. Nutr. Rew. 51, 209 ± 212. Statistical Analysis Systems (1985): SAS User's Guide, Statistics. Cary, NC: SAS Institute Inc. Varo P (1981): Table of the Nutrient Minerals (KivennaÈisainetaulukko). 2nd edn. Helsinki: Otava (in Finnish). Varo P, Alfthan G, Ekholm P, Aro A & Koivistoinen P (1988): Selenium intake and serum selenium in Finland: effects of soil fertilization with selenium. Am. J. Clin. Nutr. 48, 342 ± 349. Varo P, Alfthan G, Huttunen JK & Aro A (1994): Nationwide selenium supplementation in FinlandÐeffects on diet, blood and tissue levels, and health. In: Burk RF (ed). Selenium in Biology and Human Health, Springer-Verlag: New York, pp. 197 ± 218. Effects of selenium fertilization in Finland W-C Wang et al Wallenstein S, Zucker CL & Fleiss JL (1980): Some statistical methods useful in circulation research. Am. J. Clin. Nutr. 47, 1 ± 9. Wang W-C, Heinonen O, MaÈkelaÈ A-L, MaÈkelaÈ P, NaÈntoÈ V & Branth S (1995a): Serum selenium, zinc and copper in Swedish and Finnish orienteers. A comparative study. Analyst 120, 837 ± 840. Wang W-C, MaÈkelaÈ A-L, NaÈntoÈ V & MaÈkelaÈ P (1995b): Serum selenium levels in diabetic children. A follow up study during selenium-enriched agricultural fertilization in Finland. Biol. Trace Elem. Res. 47, 355 ± 364. Wang W-C, NaÈntoÈ V, MaÈkelaÈ A-L & MaÈkelaÈP (1995c): Effect of nationwide selenium supplementation in Finland on selenium status in children with juvenile rheumatoid arthritis. A ten-year follow-up study. Analyst 120, 955 ± 958. 535