9 1992 by The Humana Press, Inc. All rights of any nature, whatsoever, reserved. 0163-4984/92/3201-3-0009 $02.00
Chromium Content of Foods and Diets J. T. KUMPULAINEN Central Laboratory, Agricultural R e s e a r c h Centre of Finland, SF-31600 Jokioinen, Finland Received January 31, 1991; Accepted February 16, 1991
ABSTRACT Comparatively few valid data are available on the chromium content of foods and on the dietary chromium (Cr) intake of various populations. This is chiefly because of the difficulties encountered in contamination control during sampling, sample pretreatment, and analysis. Moreover, there are several analytical problems involved that are mostly owing to the low concentration level of Cr in foods. However, with the recent establishment of food reference materials with certified low concentrations of Cr, the analytical validity of studies on Cr content of foods and on its dietary intake by various populations can be ascertained. With the exception of herbs and condiments, and certain other special food items with a relatively low average consumption rate, such as tea, coffee, and some candies, most foods contain Cr below 100 ~g/kg. Staple foods, particularly cereals and milk, are very low (-~10 F~g/kg) in Cr, showing little or no geographic variation. Food processing may increase food Cr content depending on the process. Processes, such as meat grinding and homogenization using stainless-steel equipment, very strongly increase the Cr content of foods. Also, acidic fruit juices in contact with steel cans are high in Cr, whereas cooking in aluminium vessels reduces the Cr content of foods. Average dietary Cr intake seems to fluctuate considerably among countries. In many developing countries, such as Brazil, the Sudan, and Iran, the dietary intake is high, from 50-100~g/d, whereas in certain developed countries, such as Finland, Sweden, Switzerland, and the US, the intake is 50 i~g/d or lower and, consequently, at or below the estimated safe and adequate daily dietary intake range of 50-200 I~g/d established by the US National Academy of Sciences. The average Cr content of human milk is below 0.5 t~g/L, thus resulting in a very low average intake of 0.3 I~g Cr/d by exclusively breast-fed infants in the US and Finland. Index F~ntries: Trace elements; chromium; intake; diets; foods. Biological Trace Element Research
9
Vol. 32, 1992
10
Kumpulainen
INTRODUCTION With an atomic number of 24 and an atomic weight of 52, chromium (Cr) belongs to Group VI a of the periodic table and is a transition elemenE Cr may exist in any of the oxidation states from - 2 to +6; however, oxidation states other than 0, + 2, +3, and + 6 are not common. Cr is ubiquitous in nature, and can be found in air, water, soil, and in biological materials at highly varying concentrations. Cr is of importance to humankind primarily owing to its wide industrial use, such as in the manufacture of stainless steel and other metal alloys. The second important aspect regarding Cr derives from the former, since the industrial use of Cr may result in concentrations in the work place and in industrial waste waters toxic to humans and other living organisms. Finally, Cr is also considered to be an essential element for humans and other mammals. In biological materials, the most stable oxidation state is + 3, because Cr compounds with oxidation states lower than + 3 are reducing, and those higher than + 3 are oxidizing. With the oxidation state of +6, chromium forms chromates and dichromates, which posess relatively high toxicity because of their strong oxidizing properties and easy penetration through biological membranes. In acidic solutions, Cr III is soluble, easily forming hexaedral coordinate complexes with suitable available ligands, such as oxalate and sulfate ions. In neutral or basic solutions, however, Cr III undergoes olation, i.e., forms polymerized high-mol wt compounds with hydroxyl ions that precipate. Thus, in biological materials, Cr III is always complexed with organic and inorganic ligands. During the 1970s, it was thought that foods contained a specific biologically active chromium complex called glucose tolerance factor (GTF), which it was claimed was necessary for the biologic activity of Cr (1); however, later studies do not support that hypothesis, and until now, such a compound has not been identified. Therefore, the content of total Cr in foods is presently believed to be responsible for the biological activity of Cr.
CHROMIUM CONTENT OF FOODS A comprehensive evaluation of the Cr content of foods and diets is hampered by the fact that Cr is one of the most difficult elements to determine accurately in biological materials (2). Difficulties encountered during the course of the history of chromium determination in foods are illustrated in data reported on the Cr content of NBS bovine liver (SRM 1577) by various investigators before the certification of this material for Cr content in 1978 (Table 1) (3). As can be seen, only one of the values reported falls within the certified range of 88 + 12 ~g/g established later Biological Trace Element Research
Vol. 32, 1992
C h r o m i u m in F o o d s a n d Diets
11 Table ]
Reported Values for Chromium in NBS Bovine Liver Reference Material (SRM 1577) Before Its Certification for Chromium Content in 1978 (3) Concentration found, ng Cr/g dry wt 3500-1500 1000 87O 490 210 150 130 130 130 120 80.6 50.1 44.9 5
Analytical method" INAA DNAA CL INAA DNAA NAA GFAAS NAA INAA NAA DNAA GFAAS INAA INAA
Reference Nadkami and Morrison, I978 Stella et al., I976 Li and Hercules, 1974 Brill et al., 1973 McClendon, 1978 Donev and Marichkova, I976 Mertz and Roginski, 1969 Tijoe et al., 1977 Plantin, 1979 Behne and Jtirgensen, 1978 Versieck et al., 1979 Christensen et al., 1976 (50) Pierce et al., 1976 Kasperek et al., 1973
"Abbreviations used: INAA = Instrumental Neutron Activation Analysis, DNAA D e s t r u c h v e N e u t r o n A c t i v a t i o n Analysis, N A A - N e u t r o n A c t i v a t i o n Analysis, G F A A S - G r a p h i t e Furnace A t o m i c A b s o r p t i o n S p e c t r o m e t r y , CL = C h e m i l u m i n e s c e n c e . ~'NBS certified value = 88 + 12 ng/g. on. In particular, not only contamination control during sampling, sample homogenization, and analysis, but also m a n y analytical difficulties owing to the low concentration level of Cr in foods as well as other analytical problems have detrimentally affected the accuracy of results reported for the c h r o m i u m content of foods and diets (2). For example, homogenization of fresh meat in a food processor equipped with standard stainless-steel blades almost doubles the chromium content of the meat, whereas a comparable homogenization of dried liver increases the Cr content 10-fold after 3 min of blending (4). Furthermore, a lack of reference materials similar in concentration level and matrix composition to foods and diets has prevented the evaluation of the validity of reported data, and it was not until quite recently that such materials have become available (2). One of the most comprehensive studies on the chromium content of foods was conducted in Finland during the latter part of the 1970s, including approx 450 different food items (5). Unfortunately, however, a subsequent evaluation of the data reported by that study proved it to be inaccurate (6). The concentration levels reported in that study for various foods, however, seem to be of the correct magnitude, and consequently, a rough idea, at least on foods of high Cr content, probably can be obtained on the basis of that study. Accordingly, canned and other processed foods are clearly higher in c h r o m i u m than fresh foods, with the exception of refined sugar, which is Biological TraceElementResearch
VoL32, 1992
12
Kumpulainen
very low in Cr, whereas brown sugar and molasses are high in chromium (5,18,50). Many candies, spices, condiments, tea, coffee, and cocoa, as well as some wines and beers seem to be high in chromium (5). In terms of spices, condiments, tea, and coffee, these results are in agreement with those of Gutherie (7), and for beer with those by Anderson and Bryden (8). Moreover, the results of Offenbacher and Pi-Sunyer (9) showing that Cr is strongly solubilized from stainless-steel cans into acidic foods support the above results. Recent results, obtained as part of the Trace Elements in Food Research Programme of the FAO European Research Network on Trace Elements (10) demonstrate that the Cr content of staple foods is very low. Thus, the Cr content of whole wheat and wheat flour ranges from 5-t0 ~g/g dry wt in nationally representative samples from some Western European countries (11). Similarly, the average Cr content of potatoes and cow's milk is approx 5 ~g/g dry wt (11). The above results on the Cr content of staple foods are lower than those reported earlier by other investigators (4,7,12,21,50).
CHROMIUM CONTENT OF DIETS The lack of knowledge, proper instrumentation, and suitable reference materials resulted in poor estimates of the Cr intake by various populations during the early years of dietary Cr research. In general, Cr intake estimates seem to be overly high as shown by Table 2, which lists results published before 1978 when the National Bureau of Standards certified their Bovine Liver reference material for Cr content. This was the first material with a certified Cr content at the same concentration level as diets and most foods, and has been of great importance in helping scientists in this field to refine their analytical and contamination control methods. Most estimates for dietary Cr intake listed in Table 2 range from 100-200 ~g/d. The judgment that most of these values are probably too high is supported by the results listed in Table I showing that most values reported during that time were overly high for Cr content in biological materials. The dietary Cr intake estimates reported before 1977 (Table 2) can be compared with those published after that time as shown in Table 3. Most of the average intake estimates representing various populations living in 14 different countries fall between 30--60 ~g/d (Table 3), whereas the earlier estimates (Table 2) are approximately three times higher, obviously reflecting extraneous contaminations during sampling, sample homogenization, and analysis. Furthermore, the sample handling and analytical quality control measures taken in many of the more recent studies are very strict, indeed (4,6,25,26,29,33,34). Approximately half of the mean intake estimates fall at or below the lower limit of the present estimated safe and adequate daily dietary intake (ESADDI) as established Biological Trace Element Research
Vol. 32, 1992
Chromium in Foods and Diets
.~
13
~
~, ~
.
,_4
~
~
~
~,
.~
o
o
9
~
9
~ ~ -
-~
t~
;> o
c~ r~
b
r-
Biological Trace Element Research
Vol. 32, 1992
14
Kumpulainen
~r3 v
o
v
~
b., p,.
,a,...i
0
m
>
ee3
"a 9 cl 0
o
> o
9
r.-
o
==
=m=~
~
=-~~'~
m
L)
5
Biological Trace Element Research
Vot 32, 1992
Chromium in Foods and Diets
15
Table 4 Estimated Safe and Adequate Daily Dietary Intake (ESADDI) of Chromium as Proposed by the Food and Nutrition Board of the US National Academy of Sciences in 1989 (35) Category Infants Infants Children and Children and Children and Children and Adults
adolescents adolescents adolescents adolescents
Age, yr 0-0.5 0.5-1 1-3 4-6 7-10 11 +
ESADDI, I~g Cr/d 1040 20-60 20-80 30-120 50-200 50-200 50-200
by the Food and Nutrition Board of the US National Academy of Sciences (Table 4) (35). The reasons for the observed differences in Cr intakes of various populations are not known, however, it seems that certain populations living in developing countries, such as the Sudan, Brazil, and Iran, do have higher dietary Cr intakes than those in developed countries, such as Sweden, Finland, England, or the US.
CHROMIUM CONTENT OF HUMAN MILK AND INTAKE OF INFANTS Table 5 lists the values reported for the Cr content of mature human milk showing that a difference of over two orders of magnitude exists between the highest and lowest values reported. A similar decline with time as that reported for Cr concentration values in human milk has occurred for other body fluids, too (48). However, the three lowest values reported, one from Finland (37) and two others from the US (36,,t7), several years later are surprisingly similar. This is in line with the fact that they all meet very high analytical quality and contamination control standards. Also, the above studies (36,37,47) showed that the milk Cr concentration did not decline with advancing stages of lactation (Table 6). Furthermore, the average total dietary Cr intake of exclusively breast-fed Finnish infants was determined to be only 0.27 _+ 11 IJ,g/d (Table 7), whereas the total daily Cr output via milk in US women was still lower, only 0.15 IJ,g/d (Table 7). The above values can be compared with the ESADDI values for infants 0-6 mos of age, from 10-40 IJ,g/d (Table 4) (35). Thus, the average dietary Cr intake of exclusively breast-fed Finnish or American infants seems to be approximately two orders of magnitude lower than that proposed by the present ESADDI value for such infants (36,37,,t7). However, the Finnish infants studied were healthy and grew according to Finnish norms (37). Thus, although the bioavailability of Cr from human milk may be better than that from cow's milk or other infant foods, it is very likely that the present ESADD1 values for Cr intake of infants are Biological Trace Element Research
Vol. 32, 1992
16
Kumpulainen
Table 5 Reported Chromium Concentrations in Mature Human Milk Mean ng Cr/mL
Reference Casey et al., 1985 (47) Casey and Hambidge, 1984 (36) Kumpulainen and Vuori, 1980 (37) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Gunshin et al., 1984 (39) Casey, 1976 (40) Hambidge, 1971 (41) Cigna et al., 1977 (42) Varo et al., 1980 (43) Grimanis et al., 1979 (44) Medvedeva, 1966 (45) Carter et al., 1968 (46)
Range ng Cr/mL
0.25 0.30 0.39 1.3 1.50 1.87 3.65 5.1 7 10
0.22-0.34 0.06-1.56 0.19-0.69 (Guatemala) (Zaire) (Sweden) (Philippines) (Nigeria) 1-21
12
6-19 3-37 5-40 18-40
20 27 29 60
43-80
Table 6 Concentration of Chromium from Healthy American and Finnish Mothers with Advancing Stages of Lactation Stage of lactation, d or mo post partum 0-14 d 15-28 d 1-3 mo 4--6 mo 7+ mo
ng Cr/mL American mothers, Finnish mothers, mean" mean `',/' 0.25' 0.25' 0.28'~
0.26 J 0.46'~
"Overall mean + SD. 0.30 + 0.17 t'From Kumpulainen and Vuori, 1980 (see ref. 37) 'From Casey et al., 1985 (see ref. 47) 'tFrom Casey and Hambidge, 1984 (see ref. 36)
0.43 0.39 0.34 0.39 _+ 0.15
approximately two orders of m a g n i t u d e too high. Obviously outdated, erroneously high values reported for h u m a n milk Cr content were employed as the basis for decision making in establishing the ESADDI values and, consequently, o u g h t to be r e m o v e d for the time being. The view that the dietary Cr requirement of infants is very likely < 0.5 ~g/d is supported by similar levels as those for h u m a n milk Cr content, recently found in the liver and spleen tissues of infants w h o accidentally died (49). Furthermore, normal whole blood, serum, and urine Cr concentration levels (48) are similar to or lower than that f o u n d in h u m a n milk. Thus, the present author's view that the dietary Cr r e q u i r e m e n t of infants from 0-6 mo of age is < 0.5 p~g/d is well-justified. Biological Trace Element Research
VOl.32, 1992
C h r o m i u m in F o o d s a n d Diets
17
Table 7 Average Daily Chromium Intake of Healthy, Exclusively Breast-Fed Finnish and American Infants Infant age, d or mo
Finnish infants, mean p,g/d'''b
American infants, mean ~g/d ''~"
0.27 0.26
0.15
8--12 d 1-2 mo
"Mean -+ SD. 0.27 + 0.11 ~'From Kumpulainen and Vuori, 1980 (see ref. 37) 'From Casey et al., 1985 (see ref. 47)
0.15
H o w e v e r , revised ESADDI values s h o u l d n o t be established until e n o u g h reliable data o n the bioavailability of Cr f r o m h u m a n milk are available.
REFERENCES 1. E. W. Toepfer, W. Mertz, E. E. Roginski, and M. M. Polansky, 1. Agric. Food Chem. 21, 69 (1973). 2. ]. Kumpulainen, Quantitative Trace Analysis of Biological Materials, H. A. McKenzie and L. E. Smythe, eds., Elsevier Science Publishers, Amsterdam, 1988, p. 451. 3. J. Kumpulainen, Composition and Physiological Properties of Human Milk, J. Schaub, ed., Elsevier Science Publishers, Amsterdam, 1985, p. 105. 4. J. Kumpulainen, E. Vuori, S. M/ikinen, and R. Kara, Br. J. Nutr. 44, 257 (1980). 5. P. Koivistoinen, Acta Agr. Scand. Suppl. 22 (1980). 6. J. Kumpulainen, M. Mutanen, M. Paakki, and J. Lehto, V~r Ffida suppl. 1/87, 39, 75 (1987). 7. B. A. Guthrie, N. Z. Med. J. 82, 418 (1975). 8. R. A. Anderson and N. Bryden, ]. A,~ric. Food Chem. 31, 308 (1983). 9. R. Offenbacher and X. Pi-Sunyer, ]. Agric. Food Chem. 31, 89 (1983). 10. Anon. Report of the 1989 Consultation of the FAO European Cooperative Research Network on Trace Eleme, ts, Lausanne, September 5-9, 1989. FAO, Rome (1989). 11. J. Kumpulainen, unpublished results (1991). 12. E. E. Ca W and O. E. Olsen, J. Assoc. Off. Anal. Chem. 58, 433 (1975). 13. H. A. Schroeder, H. Balassa, and I. H. Tipton, J. Chronic. Dis. 15, 941 (1962). 14. M. L. Deutsch, D. Duffy, H. D. Pillsbury, and H. W. Lay, J. Assoc. Offic. Anal. Chem. 46, 759-762 (1963). 15. Y. Murakami, T. Suzuki, T. Yamagata, and N. Yamagata, ]. Ra&at. Res. 6, 105 (1965). 16. R. A. Levine, D. H. Streeten, and R. J. Doisy, Metabolism 17, 114 (1968). 17. I. H. Tipton, P. L. Steward, and ]. Dickson, Health Phys. 16, 455-462 (1969). 18. W. R. Wolf, W. Mertz, and R. Masironi, J. Agric. Food Chem. 22, 1037 (1974). 19. S. D. Soman, V. K. Panday, K. T. Joseph, and S. J. Raut, Health Phys. 17, 36 (1969). 20. E. W. Murphy, L. Page, and B. K. Watt, J. Amer. Diet. Assoc. 58, 115 (1971). 21. H. A. Schroeder, Am. J. Clin. Nutr. 24, 562 (1971). 22. ]. C. Meranger and D. C. Smith, Ca,. ]. Publ. Health 63, 53 (1972). Biological Trace Element Research
Vol. 32, 1992
18
t(umpulainen
23. B. Gutherie, Proc. Univ. Otago Med. Sch. 51, 47 (1973). 24. G. F. Clemente, L. C. Rossi, and G. P. Santaroni, J. Radioanal. Chem. 37, 549 (1977). 25. R. Schelenz, J. Radioanal. Chem. 37, 539 (1977). 26. J. T. Kumpulainen, W. R. Wolf, C. Veillon, and W. Mertz, J. Agric. Food Chem. 27, 490 (1979). 27. R. S. Gibson and C. A. Scythes, J. Biol. Trace Element. Res. 6, 105 (1984). 28. W. Bunker, M. S. Lawson, H. T. Delves, and B. E. Clayton, Am. J. Clin. Nutr. 39, 797 (1984). 29. R. A. Anderson and A. S. Kozlowsky, Am. J. Clin. Nutr. 41, 1177 (1985). 30. N. K. Aras, Middle East Technical University, Ankara, Turkey, unpublished results (1987). 31. S. Wyttenbach, S. Bajo, and L. Tobler, Trace Element--Analytical Chemistry in Medicine and Biology, vol. 4, P. Bratter and P. Schramel, eds., Walter de Gruyter & Co., Berlin, 1987, p. 169. 32. Anon. Food Monitoring in Denmark--Nutrients and Contaminants 1983-1987, The National Food Agency of Denmark, ed., 1990, p. 176. 33. W. Becker and J. Kumpulainen, Br. ]. Nutr. 66 (in press) (1991). 34. R. M. Parr, M. Abdulla, N. K. Aras, A. R. Byrne, C. Camara-Rica, S. Finnie, A. G. Gharib, G. Ingrao, G. V. Iyengar, F. A. Khangi, S. S. Krishnan, J. Kumpulainen, S. Liu, R. Schelenz, S. Srianujata, J. T. Tanner and W. Wolf. Proceedings of the TEMA-7 symposium in Dubrovnik, May 20-25, 1990. B. Mom~ilovi4, ed. IMI, Zagreb, 1991, p. 13-3. 35. Anon., Food and Nutrition Board. Recommended Dietary Allowances, 10th ed., National Academy of Sciences, Washington D.C. (1989). 36. C. E. Casey and K. M. Hambidge, Br. J. Nutr. 52, 73 (1984). 37. J. Kumpulainen and E. Vuori, Am. J. Clin. Nutr. 33, 2299 (1980). 38. G. V. Iyengar, Composition and Physiological Properties of Human Milk, J. Schaub, ed., Elsevier Science Publishers, Amsterdam, 1985, p. 17. 39. H. Gunshin, M. Yoshikowa, T. Doudou, and N. Kato, Agric. Biol. Chem. 49, 21 (1985). 40. C. Casey, Proc. Univ. Otago Med. School 54, 7 (1976). 41. K. M. Hambidge, W. Mertz, and W. E. Cornatzer, eds., Newer Trace Elements in Nutrition, Marcel Dekker, New York, 1971, p. 169. 42. C. R. Cigna, G. F. Clemente, G. Mastimi, F. Petri, and G. P. Santaroni, Proc. 3rd Int. Conf. Nucl. Meth. in Environ. and Energy Res., Univ. Missouri, Columbia, Oct. 10-13, 1977, p. 619. 43. P. Varo, M. Nuurtamo, E. Saari, and P. Koivistoinen, Acta Agric. Scand. Suppl. 22, 115 (1980). 44. A. P. Grimanis, M. Vassilaki-Grimani, D. Alexiou, C. Papadatos, Proc. Nucl. Activ. Tech. Life Sci. 1978, IAEA, Viemta, 1979, p. 241. 45. V. I. Medvedeva, Dobl. Acad. BSSR, 10, 98 (1966). (Ref. Chem. Abstr. 64, 20316). 46. J. P. Carter, A. Kattab, L. Abid, K. Hadi, and J. T. Davis, Am. ]. Clin. Nutr. 21, 195 (1968). 47. C. E. Casey, M. Hambidge, and M. Neville, Am J. Clin. Nutr. 41, 1193 (1985). 48. J. Versieck, CRC Crit. Rev. Clin. Lab. Sci. 22, 97 (1985). 49. E. Vuori and J. Kumpulainen, Trace-Elem. Med. 4, 88 (1987). 50. J. J. Christensen, P. A. Hearty, and R. M. Izatt, ]. Agric. Food Chem. 24, 811 (1976).
Biological TraceElementResearch
Vol.32, 1992
Chromium Content of Foods and Diets J. T. KUMPULAINEN Central Laboratory, Agricultural R e s e a r c h Centre of Finland, SF-31600 Jokioinen, Finland Received January 31, 1991; Accepted February 16, 1991
ABSTRACT Comparatively few valid data are available on the chromium content of foods and on the dietary chromium (Cr) intake of various populations. This is chiefly because of the difficulties encountered in contamination control during sampling, sample pretreatment, and analysis. Moreover, there are several analytical problems involved that are mostly owing to the low concentration level of Cr in foods. However, with the recent establishment of food reference materials with certified low concentrations of Cr, the analytical validity of studies on Cr content of foods and on its dietary intake by various populations can be ascertained. With the exception of herbs and condiments, and certain other special food items with a relatively low average consumption rate, such as tea, coffee, and some candies, most foods contain Cr below 100 ~g/kg. Staple foods, particularly cereals and milk, are very low (-~10 F~g/kg) in Cr, showing little or no geographic variation. Food processing may increase food Cr content depending on the process. Processes, such as meat grinding and homogenization using stainless-steel equipment, very strongly increase the Cr content of foods. Also, acidic fruit juices in contact with steel cans are high in Cr, whereas cooking in aluminium vessels reduces the Cr content of foods. Average dietary Cr intake seems to fluctuate considerably among countries. In many developing countries, such as Brazil, the Sudan, and Iran, the dietary intake is high, from 50-100~g/d, whereas in certain developed countries, such as Finland, Sweden, Switzerland, and the US, the intake is 50 i~g/d or lower and, consequently, at or below the estimated safe and adequate daily dietary intake range of 50-200 I~g/d established by the US National Academy of Sciences. The average Cr content of human milk is below 0.5 t~g/L, thus resulting in a very low average intake of 0.3 I~g Cr/d by exclusively breast-fed infants in the US and Finland. Index F~ntries: Trace elements; chromium; intake; diets; foods. Biological Trace Element Research
9
Vol. 32, 1992
10
Kumpulainen
INTRODUCTION With an atomic number of 24 and an atomic weight of 52, chromium (Cr) belongs to Group VI a of the periodic table and is a transition elemenE Cr may exist in any of the oxidation states from - 2 to +6; however, oxidation states other than 0, + 2, +3, and + 6 are not common. Cr is ubiquitous in nature, and can be found in air, water, soil, and in biological materials at highly varying concentrations. Cr is of importance to humankind primarily owing to its wide industrial use, such as in the manufacture of stainless steel and other metal alloys. The second important aspect regarding Cr derives from the former, since the industrial use of Cr may result in concentrations in the work place and in industrial waste waters toxic to humans and other living organisms. Finally, Cr is also considered to be an essential element for humans and other mammals. In biological materials, the most stable oxidation state is + 3, because Cr compounds with oxidation states lower than + 3 are reducing, and those higher than + 3 are oxidizing. With the oxidation state of +6, chromium forms chromates and dichromates, which posess relatively high toxicity because of their strong oxidizing properties and easy penetration through biological membranes. In acidic solutions, Cr III is soluble, easily forming hexaedral coordinate complexes with suitable available ligands, such as oxalate and sulfate ions. In neutral or basic solutions, however, Cr III undergoes olation, i.e., forms polymerized high-mol wt compounds with hydroxyl ions that precipate. Thus, in biological materials, Cr III is always complexed with organic and inorganic ligands. During the 1970s, it was thought that foods contained a specific biologically active chromium complex called glucose tolerance factor (GTF), which it was claimed was necessary for the biologic activity of Cr (1); however, later studies do not support that hypothesis, and until now, such a compound has not been identified. Therefore, the content of total Cr in foods is presently believed to be responsible for the biological activity of Cr.
CHROMIUM CONTENT OF FOODS A comprehensive evaluation of the Cr content of foods and diets is hampered by the fact that Cr is one of the most difficult elements to determine accurately in biological materials (2). Difficulties encountered during the course of the history of chromium determination in foods are illustrated in data reported on the Cr content of NBS bovine liver (SRM 1577) by various investigators before the certification of this material for Cr content in 1978 (Table 1) (3). As can be seen, only one of the values reported falls within the certified range of 88 + 12 ~g/g established later Biological Trace Element Research
Vol. 32, 1992
C h r o m i u m in F o o d s a n d Diets
11 Table ]
Reported Values for Chromium in NBS Bovine Liver Reference Material (SRM 1577) Before Its Certification for Chromium Content in 1978 (3) Concentration found, ng Cr/g dry wt 3500-1500 1000 87O 490 210 150 130 130 130 120 80.6 50.1 44.9 5
Analytical method" INAA DNAA CL INAA DNAA NAA GFAAS NAA INAA NAA DNAA GFAAS INAA INAA
Reference Nadkami and Morrison, I978 Stella et al., I976 Li and Hercules, 1974 Brill et al., 1973 McClendon, 1978 Donev and Marichkova, I976 Mertz and Roginski, 1969 Tijoe et al., 1977 Plantin, 1979 Behne and Jtirgensen, 1978 Versieck et al., 1979 Christensen et al., 1976 (50) Pierce et al., 1976 Kasperek et al., 1973
"Abbreviations used: INAA = Instrumental Neutron Activation Analysis, DNAA D e s t r u c h v e N e u t r o n A c t i v a t i o n Analysis, N A A - N e u t r o n A c t i v a t i o n Analysis, G F A A S - G r a p h i t e Furnace A t o m i c A b s o r p t i o n S p e c t r o m e t r y , CL = C h e m i l u m i n e s c e n c e . ~'NBS certified value = 88 + 12 ng/g. on. In particular, not only contamination control during sampling, sample homogenization, and analysis, but also m a n y analytical difficulties owing to the low concentration level of Cr in foods as well as other analytical problems have detrimentally affected the accuracy of results reported for the c h r o m i u m content of foods and diets (2). For example, homogenization of fresh meat in a food processor equipped with standard stainless-steel blades almost doubles the chromium content of the meat, whereas a comparable homogenization of dried liver increases the Cr content 10-fold after 3 min of blending (4). Furthermore, a lack of reference materials similar in concentration level and matrix composition to foods and diets has prevented the evaluation of the validity of reported data, and it was not until quite recently that such materials have become available (2). One of the most comprehensive studies on the chromium content of foods was conducted in Finland during the latter part of the 1970s, including approx 450 different food items (5). Unfortunately, however, a subsequent evaluation of the data reported by that study proved it to be inaccurate (6). The concentration levels reported in that study for various foods, however, seem to be of the correct magnitude, and consequently, a rough idea, at least on foods of high Cr content, probably can be obtained on the basis of that study. Accordingly, canned and other processed foods are clearly higher in c h r o m i u m than fresh foods, with the exception of refined sugar, which is Biological TraceElementResearch
VoL32, 1992
12
Kumpulainen
very low in Cr, whereas brown sugar and molasses are high in chromium (5,18,50). Many candies, spices, condiments, tea, coffee, and cocoa, as well as some wines and beers seem to be high in chromium (5). In terms of spices, condiments, tea, and coffee, these results are in agreement with those of Gutherie (7), and for beer with those by Anderson and Bryden (8). Moreover, the results of Offenbacher and Pi-Sunyer (9) showing that Cr is strongly solubilized from stainless-steel cans into acidic foods support the above results. Recent results, obtained as part of the Trace Elements in Food Research Programme of the FAO European Research Network on Trace Elements (10) demonstrate that the Cr content of staple foods is very low. Thus, the Cr content of whole wheat and wheat flour ranges from 5-t0 ~g/g dry wt in nationally representative samples from some Western European countries (11). Similarly, the average Cr content of potatoes and cow's milk is approx 5 ~g/g dry wt (11). The above results on the Cr content of staple foods are lower than those reported earlier by other investigators (4,7,12,21,50).
CHROMIUM CONTENT OF DIETS The lack of knowledge, proper instrumentation, and suitable reference materials resulted in poor estimates of the Cr intake by various populations during the early years of dietary Cr research. In general, Cr intake estimates seem to be overly high as shown by Table 2, which lists results published before 1978 when the National Bureau of Standards certified their Bovine Liver reference material for Cr content. This was the first material with a certified Cr content at the same concentration level as diets and most foods, and has been of great importance in helping scientists in this field to refine their analytical and contamination control methods. Most estimates for dietary Cr intake listed in Table 2 range from 100-200 ~g/d. The judgment that most of these values are probably too high is supported by the results listed in Table I showing that most values reported during that time were overly high for Cr content in biological materials. The dietary Cr intake estimates reported before 1977 (Table 2) can be compared with those published after that time as shown in Table 3. Most of the average intake estimates representing various populations living in 14 different countries fall between 30--60 ~g/d (Table 3), whereas the earlier estimates (Table 2) are approximately three times higher, obviously reflecting extraneous contaminations during sampling, sample homogenization, and analysis. Furthermore, the sample handling and analytical quality control measures taken in many of the more recent studies are very strict, indeed (4,6,25,26,29,33,34). Approximately half of the mean intake estimates fall at or below the lower limit of the present estimated safe and adequate daily dietary intake (ESADDI) as established Biological Trace Element Research
Vol. 32, 1992
Chromium in Foods and Diets
.~
13
~
~, ~
.
,_4
~
~
~
~,
.~
o
o
9
~
9
~ ~ -
-~
t~
;> o
c~ r~
b
r-
Biological Trace Element Research
Vol. 32, 1992
14
Kumpulainen
~r3 v
o
v
~
b., p,.
,a,...i
0
m
>
ee3
"a 9 cl 0
o
> o
9
r.-
o
==
=m=~
~
=-~~'~
m
L)
5
Biological Trace Element Research
Vot 32, 1992
Chromium in Foods and Diets
15
Table 4 Estimated Safe and Adequate Daily Dietary Intake (ESADDI) of Chromium as Proposed by the Food and Nutrition Board of the US National Academy of Sciences in 1989 (35) Category Infants Infants Children and Children and Children and Children and Adults
adolescents adolescents adolescents adolescents
Age, yr 0-0.5 0.5-1 1-3 4-6 7-10 11 +
ESADDI, I~g Cr/d 1040 20-60 20-80 30-120 50-200 50-200 50-200
by the Food and Nutrition Board of the US National Academy of Sciences (Table 4) (35). The reasons for the observed differences in Cr intakes of various populations are not known, however, it seems that certain populations living in developing countries, such as the Sudan, Brazil, and Iran, do have higher dietary Cr intakes than those in developed countries, such as Sweden, Finland, England, or the US.
CHROMIUM CONTENT OF HUMAN MILK AND INTAKE OF INFANTS Table 5 lists the values reported for the Cr content of mature human milk showing that a difference of over two orders of magnitude exists between the highest and lowest values reported. A similar decline with time as that reported for Cr concentration values in human milk has occurred for other body fluids, too (48). However, the three lowest values reported, one from Finland (37) and two others from the US (36,,t7), several years later are surprisingly similar. This is in line with the fact that they all meet very high analytical quality and contamination control standards. Also, the above studies (36,37,47) showed that the milk Cr concentration did not decline with advancing stages of lactation (Table 6). Furthermore, the average total dietary Cr intake of exclusively breast-fed Finnish infants was determined to be only 0.27 _+ 11 IJ,g/d (Table 7), whereas the total daily Cr output via milk in US women was still lower, only 0.15 IJ,g/d (Table 7). The above values can be compared with the ESADDI values for infants 0-6 mos of age, from 10-40 IJ,g/d (Table 4) (35). Thus, the average dietary Cr intake of exclusively breast-fed Finnish or American infants seems to be approximately two orders of magnitude lower than that proposed by the present ESADDI value for such infants (36,37,,t7). However, the Finnish infants studied were healthy and grew according to Finnish norms (37). Thus, although the bioavailability of Cr from human milk may be better than that from cow's milk or other infant foods, it is very likely that the present ESADD1 values for Cr intake of infants are Biological Trace Element Research
Vol. 32, 1992
16
Kumpulainen
Table 5 Reported Chromium Concentrations in Mature Human Milk Mean ng Cr/mL
Reference Casey et al., 1985 (47) Casey and Hambidge, 1984 (36) Kumpulainen and Vuori, 1980 (37) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Iyengar, 1985 (38) Gunshin et al., 1984 (39) Casey, 1976 (40) Hambidge, 1971 (41) Cigna et al., 1977 (42) Varo et al., 1980 (43) Grimanis et al., 1979 (44) Medvedeva, 1966 (45) Carter et al., 1968 (46)
Range ng Cr/mL
0.25 0.30 0.39 1.3 1.50 1.87 3.65 5.1 7 10
0.22-0.34 0.06-1.56 0.19-0.69 (Guatemala) (Zaire) (Sweden) (Philippines) (Nigeria) 1-21
12
6-19 3-37 5-40 18-40
20 27 29 60
43-80
Table 6 Concentration of Chromium from Healthy American and Finnish Mothers with Advancing Stages of Lactation Stage of lactation, d or mo post partum 0-14 d 15-28 d 1-3 mo 4--6 mo 7+ mo
ng Cr/mL American mothers, Finnish mothers, mean" mean `',/' 0.25' 0.25' 0.28'~
0.26 J 0.46'~
"Overall mean + SD. 0.30 + 0.17 t'From Kumpulainen and Vuori, 1980 (see ref. 37) 'From Casey et al., 1985 (see ref. 47) 'tFrom Casey and Hambidge, 1984 (see ref. 36)
0.43 0.39 0.34 0.39 _+ 0.15
approximately two orders of m a g n i t u d e too high. Obviously outdated, erroneously high values reported for h u m a n milk Cr content were employed as the basis for decision making in establishing the ESADDI values and, consequently, o u g h t to be r e m o v e d for the time being. The view that the dietary Cr requirement of infants is very likely < 0.5 ~g/d is supported by similar levels as those for h u m a n milk Cr content, recently found in the liver and spleen tissues of infants w h o accidentally died (49). Furthermore, normal whole blood, serum, and urine Cr concentration levels (48) are similar to or lower than that f o u n d in h u m a n milk. Thus, the present author's view that the dietary Cr r e q u i r e m e n t of infants from 0-6 mo of age is < 0.5 p~g/d is well-justified. Biological Trace Element Research
VOl.32, 1992
C h r o m i u m in F o o d s a n d Diets
17
Table 7 Average Daily Chromium Intake of Healthy, Exclusively Breast-Fed Finnish and American Infants Infant age, d or mo
Finnish infants, mean p,g/d'''b
American infants, mean ~g/d ''~"
0.27 0.26
0.15
8--12 d 1-2 mo
"Mean -+ SD. 0.27 + 0.11 ~'From Kumpulainen and Vuori, 1980 (see ref. 37) 'From Casey et al., 1985 (see ref. 47)
0.15
H o w e v e r , revised ESADDI values s h o u l d n o t be established until e n o u g h reliable data o n the bioavailability of Cr f r o m h u m a n milk are available.
REFERENCES 1. E. W. Toepfer, W. Mertz, E. E. Roginski, and M. M. Polansky, 1. Agric. Food Chem. 21, 69 (1973). 2. ]. Kumpulainen, Quantitative Trace Analysis of Biological Materials, H. A. McKenzie and L. E. Smythe, eds., Elsevier Science Publishers, Amsterdam, 1988, p. 451. 3. J. Kumpulainen, Composition and Physiological Properties of Human Milk, J. Schaub, ed., Elsevier Science Publishers, Amsterdam, 1985, p. 105. 4. J. Kumpulainen, E. Vuori, S. M/ikinen, and R. Kara, Br. J. Nutr. 44, 257 (1980). 5. P. Koivistoinen, Acta Agr. Scand. Suppl. 22 (1980). 6. J. Kumpulainen, M. Mutanen, M. Paakki, and J. Lehto, V~r Ffida suppl. 1/87, 39, 75 (1987). 7. B. A. Guthrie, N. Z. Med. J. 82, 418 (1975). 8. R. A. Anderson and N. Bryden, ]. A,~ric. Food Chem. 31, 308 (1983). 9. R. Offenbacher and X. Pi-Sunyer, ]. Agric. Food Chem. 31, 89 (1983). 10. Anon. Report of the 1989 Consultation of the FAO European Cooperative Research Network on Trace Eleme, ts, Lausanne, September 5-9, 1989. FAO, Rome (1989). 11. J. Kumpulainen, unpublished results (1991). 12. E. E. Ca W and O. E. Olsen, J. Assoc. Off. Anal. Chem. 58, 433 (1975). 13. H. A. Schroeder, H. Balassa, and I. H. Tipton, J. Chronic. Dis. 15, 941 (1962). 14. M. L. Deutsch, D. Duffy, H. D. Pillsbury, and H. W. Lay, J. Assoc. Offic. Anal. Chem. 46, 759-762 (1963). 15. Y. Murakami, T. Suzuki, T. Yamagata, and N. Yamagata, ]. Ra&at. Res. 6, 105 (1965). 16. R. A. Levine, D. H. Streeten, and R. J. Doisy, Metabolism 17, 114 (1968). 17. I. H. Tipton, P. L. Steward, and ]. Dickson, Health Phys. 16, 455-462 (1969). 18. W. R. Wolf, W. Mertz, and R. Masironi, J. Agric. Food Chem. 22, 1037 (1974). 19. S. D. Soman, V. K. Panday, K. T. Joseph, and S. J. Raut, Health Phys. 17, 36 (1969). 20. E. W. Murphy, L. Page, and B. K. Watt, J. Amer. Diet. Assoc. 58, 115 (1971). 21. H. A. Schroeder, Am. J. Clin. Nutr. 24, 562 (1971). 22. ]. C. Meranger and D. C. Smith, Ca,. ]. Publ. Health 63, 53 (1972). Biological Trace Element Research
Vol. 32, 1992
18
t(umpulainen
23. B. Gutherie, Proc. Univ. Otago Med. Sch. 51, 47 (1973). 24. G. F. Clemente, L. C. Rossi, and G. P. Santaroni, J. Radioanal. Chem. 37, 549 (1977). 25. R. Schelenz, J. Radioanal. Chem. 37, 539 (1977). 26. J. T. Kumpulainen, W. R. Wolf, C. Veillon, and W. Mertz, J. Agric. Food Chem. 27, 490 (1979). 27. R. S. Gibson and C. A. Scythes, J. Biol. Trace Element. Res. 6, 105 (1984). 28. W. Bunker, M. S. Lawson, H. T. Delves, and B. E. Clayton, Am. J. Clin. Nutr. 39, 797 (1984). 29. R. A. Anderson and A. S. Kozlowsky, Am. J. Clin. Nutr. 41, 1177 (1985). 30. N. K. Aras, Middle East Technical University, Ankara, Turkey, unpublished results (1987). 31. S. Wyttenbach, S. Bajo, and L. Tobler, Trace Element--Analytical Chemistry in Medicine and Biology, vol. 4, P. Bratter and P. Schramel, eds., Walter de Gruyter & Co., Berlin, 1987, p. 169. 32. Anon. Food Monitoring in Denmark--Nutrients and Contaminants 1983-1987, The National Food Agency of Denmark, ed., 1990, p. 176. 33. W. Becker and J. Kumpulainen, Br. ]. Nutr. 66 (in press) (1991). 34. R. M. Parr, M. Abdulla, N. K. Aras, A. R. Byrne, C. Camara-Rica, S. Finnie, A. G. Gharib, G. Ingrao, G. V. Iyengar, F. A. Khangi, S. S. Krishnan, J. Kumpulainen, S. Liu, R. Schelenz, S. Srianujata, J. T. Tanner and W. Wolf. Proceedings of the TEMA-7 symposium in Dubrovnik, May 20-25, 1990. B. Mom~ilovi4, ed. IMI, Zagreb, 1991, p. 13-3. 35. Anon., Food and Nutrition Board. Recommended Dietary Allowances, 10th ed., National Academy of Sciences, Washington D.C. (1989). 36. C. E. Casey and K. M. Hambidge, Br. J. Nutr. 52, 73 (1984). 37. J. Kumpulainen and E. Vuori, Am. J. Clin. Nutr. 33, 2299 (1980). 38. G. V. Iyengar, Composition and Physiological Properties of Human Milk, J. Schaub, ed., Elsevier Science Publishers, Amsterdam, 1985, p. 17. 39. H. Gunshin, M. Yoshikowa, T. Doudou, and N. Kato, Agric. Biol. Chem. 49, 21 (1985). 40. C. Casey, Proc. Univ. Otago Med. School 54, 7 (1976). 41. K. M. Hambidge, W. Mertz, and W. E. Cornatzer, eds., Newer Trace Elements in Nutrition, Marcel Dekker, New York, 1971, p. 169. 42. C. R. Cigna, G. F. Clemente, G. Mastimi, F. Petri, and G. P. Santaroni, Proc. 3rd Int. Conf. Nucl. Meth. in Environ. and Energy Res., Univ. Missouri, Columbia, Oct. 10-13, 1977, p. 619. 43. P. Varo, M. Nuurtamo, E. Saari, and P. Koivistoinen, Acta Agric. Scand. Suppl. 22, 115 (1980). 44. A. P. Grimanis, M. Vassilaki-Grimani, D. Alexiou, C. Papadatos, Proc. Nucl. Activ. Tech. Life Sci. 1978, IAEA, Viemta, 1979, p. 241. 45. V. I. Medvedeva, Dobl. Acad. BSSR, 10, 98 (1966). (Ref. Chem. Abstr. 64, 20316). 46. J. P. Carter, A. Kattab, L. Abid, K. Hadi, and J. T. Davis, Am. ]. Clin. Nutr. 21, 195 (1968). 47. C. E. Casey, M. Hambidge, and M. Neville, Am J. Clin. Nutr. 41, 1193 (1985). 48. J. Versieck, CRC Crit. Rev. Clin. Lab. Sci. 22, 97 (1985). 49. E. Vuori and J. Kumpulainen, Trace-Elem. Med. 4, 88 (1987). 50. J. J. Christensen, P. A. Hearty, and R. M. Izatt, ]. Agric. Food Chem. 24, 811 (1976).
Biological TraceElementResearch
Vol.32, 1992
