The Science of the Total Environment, 105 (1991) 101-107 Elsevier Science Publishers B.V., Amsterdam


Selenium in human milk and dietary selenium intake by Greeks Michael S. Bratakos and Panayictis V. Ioannou* Department of Chemistry, University of Patras. Patras. Greece

(Received March 22nd, 1990: accepted June 7th, 1990)

ABSTRACT Fluorimetric determinaUon of selenium in colostrum, transitional and mature human milk gave the following concentrations (mean and standard deviation): 41 + 16, 23 +_ 6 and ! 7 _+ 3 ng Se ml-~, respectively. The ranges for each kind of milk, especially for mature milk, were narrow. For all cases studied, the Se concentralion in milk decreased with lactation time, reaching a plateau, at 17ng Se ml-~, after 20 days. it is estimated that breast-fed-only babies in Greece receive -,, 5-1 !/~g Se day- mup to 6 months of age. From consumed food data it was estimated that adult Greeks receive 100 + 6/~g Se day-~, in close agreement with our previously determined value of I I0/ig Se day-~ estimated from food disappearance data. INTRODUCTION Selenium is an essential trace element for man; low levels in the blood are associated with certain juvenile conditions, such as Keshan, Kaschin-Beck and kwashiorkor, and various non-malignant and malignant diseases Ill. Selenium has also been implicated in carcinogenesis prevention [1-3]. M a n receives Se from food, and, in general, the higher the proportion of Se in food the higher the concentrations in h u m a n tissues [l]. F r o m Se determinations in Greek foods [4-6] we calculated that breast-fed-only babies, up to 6 months old, are likely to receive -,- I I pg Se day-~, while, from food c o n s u m p t i o n (disappearance) data, we estimated that adult Greeks receive 110/~g Se day -~, based on uncooked food [4], or ~ 95/~g Se day -~, based on uncooked plus cooked food [6]. The purpose of this study was to determine the longitudinal changes in the Se content of mothers' milk, as there is a revival of breast-feeding in Greece, and also to provide an independent estimate o f the daily Se intake by adult Greeks based on eaten food data and the Se content of these foods. * Author to whom correspondence should be addressed.


© 1991 - - Elsevier Science Publishers B.V.




Subjects and sampling Eleven healthy lactating Greek mothers, living in Athens, d o n a t e d milk samples in 1988. Six of them, aged 30 + 5 years, donated milk on most of the sampling days indicated in Table 1. All mothers had uncomplicated pregnancies and deliveries, were well nourished and belonged to the middle income class [7]. The breasts were cleaned with double distilled water, and the milk samples (10-20ml) were collected m a n u a l l y from b o t h breasts, after discarding the first 1-2 ml, into sterilised polyethylene bottles before the first feeding of the baby [8]. The samples were analysed immediately for total Se. One middle income three-membered family, living in A t h e n s , participated in the dietary Se intake study. The kind and quantities of foods eaten freely by each member of the family during 30 consecutive days (April 1987) were recorded. F r o m each uncooked food item a sample was analysed for total Se. TABLE ! Changes in selenium concentration in milk of Greek mothers Kind of milk





Lactation time (days postpartum)

I 2 3 I-3 4 10 4-10 20 30 60 90 120 150 180 20-180 !-180


Se concentration in milk (ng ml -t , as received)


Age (yearr,) .~ _+ SD

.~ + SD


6 8 6 20 6 6 12 6 6 8 5 3 5 3 36 68

30 _+ 5 29 _+ 6 30 _+ 5 30 _+ 5 30 + 5 30 __. 5 30 + 5 30 _+ 5 30 + 5 31 + 5 30 + 5 29 + 5 29 + 6 28 + 7 30 :i: 5 30 + 5

59 +_ 12" 38 +_ 12b 27 _+ 9 41 +_ 16 24 + 7 21 + 5 23 + 6 19 + 3 19 + 3 18 + 3 16 + 2 17 + 1 15 + 2 13 + 3 17 + 3 25 + 14

46-79 27-63 20--44 20-79 18-J6 16-28 16-36 15-23 15-21 14-22 14-19 17-18 13-17 11-16 11-23 !1-79

~Mean differs significantly from other means (0t --- 0.05). hMean differs significantly from other means, except from that of Day 3 (at = 0.05).



Selenium determination The samples were analysed for total Se as described previously [4,5]. Each sample was analysed at least twice. Recoveries of added amounts of Se(IV) were 92-108%. Determination of Se content of the National Bureau of Standards reference material 1577a (Bovine Liver) gave 0.72 + 0.02/ag Se g-~ (certified value, 0.71 _ 0.07 #g Se g-~) for 10 determinations. The limit of detection, defined as 3SD of the means of 12 blanks in each category, was 0.005 ltg Se g-~ or #g Se ml-~ for foods and for milk.

Statistical analyses One-way analysis of variance and Tuckey's "honestly significant difference" test were used for intergroup comparisons; means were compared using a t-test assuming unknown but equal population variances [9]. RESULTS AND DISCUSSION

Selenium is regarded as a normal constituent of human milk [1,8,10], being mostly protein-bound [I !]. It is hypothesised that the role of the Se in milk is to depress the peroxidation of the long-chain polyunsaturated fatty acids in milk, which are important for the development of the brain and visual cells of newborn babies [12]. Table 1 shows that colostrum milk is richer in Se than transitional or mature milk, and that the latter is poorer in Se than transitional milk (p < 0.01). Our values for the Se levels in colostrum, transitional and mature milk fall between the, lower, corresponding values for Belgium [8] and Yugoslavia [13] and the higher values reported for Germany [14] and Japan [15]. The Se concentration in mature milk, 17 +_ 3 ng ml-~, obtained by us in 1988 differs significantly (p < 0.01) from the value, 20 + l n g Se ml -~ (n = 24), obtained for Athenian mothers in !973 [10]. The exact reason for this decline is not known, but it may be related to changing dietary habits; for example, in recent years Greeks have reduced their consumption of cereals and bread and increased that of meat [4]. The Se levels in colostrum decreased sharply from the first to the third day, as shown in Table I. Thereafter the reduction is smoother, reaching a plateau at 17 ng Se ml -t after 20 days of lactation, then declining very slowly with time. The same pattern has also been observed in Germany [ 14], India [ 17] and Japan [12,15], but not in Belgium [8] and Yugoslavia [13]. The reasons for the decline are not clear [16]. The plateau effect suggests the operation of a homeostatic mechanism [10,12]. Even though no correlation between Se in mature milk and Se intake during lactation [16], or serum [15] or plasma [16]



or erythrocyte [16] or hair [15] Se, has been found, the level of Se in mature milk can probably be affected by long-term dietary habits [17]. Thus, mothers residing in a high Se area of China had mature milk containing 283 ng Se ml[18]; those living in countries with moderate Se soils, such as Greece, parts of the USA, parts of China, and Germany, had mature milk containing 17 (this study), 20 [20,17], 21 [18] and 28 [14] ng Se ml -~, respectively; while those residing in a low Se area of China, Belgium or Finland had mature milk containing 3 [18], 10 [8] and 5-10 [16] ng Se ml -~, respectively. We found no correlation (p > 0.20) between the age of the mothers and the Se content of their milk for each sampling day, which agrees with the literature [17]. Table 1 also shows that the range of Se values is wide for colostrum, which agrees with the literature [14,15], but it is remarkably narrow for mature milk, in sharp contrast to the literature [14,15,17]. The narrowness of this range is probably due to the mothers having a constant, chronic intake of Se, since it is known [4] that Greeks from 1960 onwards have received an approximately constant amount of Se from food. In all cases, the Se content of the milk of each individual Greek mother decreased with lactation time. In Belgium, however, not all mothers showed such a trend and this has been attributed to food consumption patterns or to other factors [8]. We confirm the observation of Higashi et al. [15] that a mother with a higher (lower) Se content in colostrum maintains a higher (lower) Se content in the transitional and mature milk. From a limited number of cases (n = 2) we observed that mothers at the second delivery had milk (colostrum, transitional and mature) richer in Se than mothers at the first delivery. This interesting observation requires further investigation. Recommended values for Se intake for babies, infants and children ( < 7 years old) have not been published. Because the volume of breast milk consumed daily by a baby is not known with certainty [19], rough estimations, using the data of Table 1, show that Greek breast-led-only babies receive a minimum of ~ 5-7/tg Se day -~ during Days 4-30. Thereafter the intake increases slowly to 11 #g Se day -~ at 6 months of age. Because the Se intake of a population calculated from food disappearance data may be overestimated, we required an independent estimate based on eaten food data. Studies of this nature must take into account the question of how representative the subjects and the diets are. Instead of selecting many subjects, and measuring food consumption for a few days [20], we selected a middle income family which, we believe, is representative of the majority of Greek families regarding eating habits, and we measured food consumption for 30 days, thus avoiding day-to-day variabilities regarding Se intakes. The buying of foods and the eating habits of the family were not influenced by the study.



TABLE 2 Selenium intake from uncooked foods b)' the members of a Greek familyduring 30 consecutive days (April 1987) Food eaten

Meats (excluding kidney and liver) Fish (excluding crustaceans) Milk, whole Cheese Eggs Bread Cereals and their products Pulses Vegetables Fruits Sugar and sweets Other foods Water Total

Selenium intake (pg month-~ + SD) Boy I 1 years old

Woman 33 years old

Man 37 years old

227 _+ 33

326 _+ 47

370 _+ 54

179 _+ 24

431 ± 57

508 + 67

267 41 135 940 147

± 74 +_ 15 -t_ 17 + 2Ot~ + ll

89 + 112 + 68 + 29 + 14 + 7+

19 45 30 18 2 I

2255 + 227

180 Ill 86 1517 156 60 166 55 7 13 9

_ 50 _+ "9 _-4 II + 323 + II ± + ± ± ± ±

13 63 25 5 2 1

3117 + 345

81 ± 23 151 -± 53 145 + 19 1714 + 36,1. 297 ± 23 158 126 75 2 17 9

+ + + ± ± +

34 48 42 I 3 1

3655 + 386

The total Se content of foods analysed in April 1987 does not differ significantly (p > 0.05) from the values determined in 1983-5 [4]. The calculated daily Se intakes for the family (boy, 75 _ 8; woman, 104 _+ 12; and man, 122 _ 13pg Se day-~; Table 2) depend upon the kind and quantities of foods eaten; the mean value was 100 +_ 611g Se a d u l t - t day-~. A value of I l0 pg adult -I day -~ has been arrived at [4] from food disappearance data, but the new mean value, 100pg Se adult ~day " ~, would appear to be more realistic. When Se losses [6] are taken into account, the daily Se intakes are reduced by ~ 12%. O u r values fall into the daily Se intake ranges of 60-120 [21] and 50-150 [22] pg recommended for adults ( > 7 years old). The major Se c o n t r i b u t o r to the Greek diet is bread ( ~ 50%) (Table 2 and ref. 4). Since wheat products contain highly available Se [1], Greeks have a satisfactory nutritional Se status [23], despite the fact that Greek foods are relatively poor in Se [4]. CONCLUSIONS The Se concentration of h u m a n milk decreases in the order: colos-



t r u m > transitional > mature. The range o f Se values for m a t u r e milk is quite n a r r o w . The Se c o n t e n t o f m a t u r e breast milk of G r e e k m o t h e r s is similar to t h a t found in other countries considered as " m e d i u m Se areas". Breast-fedonly babies in Greece, up to 6 m o n t h s old, receive ~ 5-11 p g Se day -~. There is a g o o d agreement for the Se intake o f G r e e k s calculated on food disappearance a n d food eaten bases. REFERENCES I G.F. Combs Jr and S.B. Combs, The Role ef Selenium in Nutrition, Academic Press, Orlando, FL, 1986. 2 L.N. Vernie, Selenium in carcinogenesis, Biochim. Biophys. Acta, 738 (1984) 203-217. 3 G. Hockman, Chemoprevention of cancer: selenium, Int. J. Biochem., 20 (1988) 123-132. 4 M.S. Bratakos, T.F. Zafiropoulos, P.A. Siskos and P.V. Ioannou, Selenium in foods produced and consumed in Greece, J. Food Sci., 52 (1987) 817-822. 5 M.S. Bratakos, T.F. Zarifopoulos, P.A. Siskos and P.V. loannou, Total selenium concentration in tap and bottled drinking water and coastal waters of Greece, Sci. Total EnviJ'on., 76 (1988) 49-54. 6 M.S. Bratakos, T.F. Zarifopoulos, P.A. Siskos and P.V. Ioannou, Selenium losses on cooking Greek foods, Int. J. Food Sci. Technol., 23 (1988) 585-590. 7 H.S. Dang, H.B. Desai, S.R. Kayasth, D.D. Jaiswal, C.N. Wadhwani and S. Somasundaram, Daily requirements of Fe, Co and Se during infancy, J. Radioanal. Nucl. Chem., 84 (1984) 177-183. 8 E. Roekens, H. Deelstra and H. Robberecht, Trace elements in human milk, selenium a case study, Sci. Total Environ., 42 (1985) 91-108. 9 W.W. Daniel, Biostatistics: A Foundation for Analysis in the Health Sciences, Wiley, Singapore, 4th edn, 1987. 10 D.M. Hadjimarkos and T.R. Shearer, Selenium in mature human milk, Am. J. Clin. Nutr., 26 (1973) 583-585. ! ! J.A. Milner, L. Sherman and M.F. Picciano, Distribution ofselenium in human milk, Am. J. Clin. Nutr., 45 (1987) 617-624. 12 Y. Hojo, Sequential study on glutatbione peroxidase and selenium contents of human milk, Sci. Total Environ., 52 (1986) 83-91. 13 L. Kosta, A.R. Byrne and M. Dermelj, Trace elements in some human milk samples by radiochemical neutron activation analysis, Sci. Total Environ.. 29 (1983) 261-268. 14 I. Lombeck, K. Kasperek, B. Bonnermann, L.E. Feinendegen and H.J. Bremer, Selenium content of human milk, cow's milk and cow's milk infant formulas, Eur. J. Pediatr., 129 (1978) 139-145. 15 A. Higashi, H. Tamari, Y. Kuroki and 1. Matsuda, Longitudinal changes in selenium content of breast milk, Acta Paediatr. Scand., 72 (1983) 433-436. 16 O.A. Levander, P.B. Moser and V.C. Morris, Dietary selenium intake and selenium concentrations of plasma, erythrocytes, and breast milk in pregnant and postpartum lactating and nonlactating women, Am, J. Clin. Nutr., 46 (1987) 694-698. 17 T.R. Shearer and D.M. Hadjimarkos, Geographic distribution of selenium in human milk, Arch. Environ. Health, 30 (1975)230-233. 18 G.-Q. Yang, L.-Z. Zhu, S.-J. Liu, L.-Z. Gu, P.-C. Qian, J.-H. Huang and M.-D. Lu, Human selenium requirements in China, in G.F. Combs Jr, J.E. Spallholz, O.A. Levander


19 20 21

22 23


and J.E. Oldfield (Eds), Selenium in Biology and Medicine, Part B, AVI Publishers, Westport, CT, 1987, Chapt. 62, p. 589. R.G. Whitehead, A.A. Paul and T.J. Cole, How much breast milk do babies need?, Acta Paediatr. Seand., Suppl., 299 (1982) 43-50. S.O. Welsh, J.M. Holden, W.R. Wolf and O.A. Levander, Selenium in self-selected diets of Maryland residents, J. Am. Diet. Assoc., 79 (1981) 277-285. National Research Council, Board of Agriculture, Committee on Animal Nutrition, Subcommittee on Selenium in Nutrition, National Academy Press, Washington, DC, 1983. H. Sakurai and K. Tsuchiya, A tentative recommendation for the maximum daily intake of selenium, Environ. Physiol. Biochem., 5 (1975) 107-118. M.S. Bratakos, H.C. Kanaki, A. Vasiliou-Waite and P.V. loannou, The nutritional selenium status of healthy Greeks, Sci. Total Environ., 91 (1990) 161-176.

Selenium in human milk and dietary selenium intake by Greeks.

Fluorimetric determination of selenium in colostrum, transitional and mature human milk gave the following concentrations (mean and standard deviation...
488KB Sizes 0 Downloads 0 Views