Dietary intake of nickel and zinc by young children – Results from food duplicate portion measurements in comparison to data calculated from dietary records and available data on

ARTICLE IN PRESS

0946-672X/$ - s

doi:10.1016/j.jte

�CorrespondE-mail addr

Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194

www.elsevier.de/jtemb

NUTRITION

Dietary intake of nickel and zinc by young children – Results from food

duplicate portion measurements in comparison to data calculated from

dietary records and available data on levels in food groups

Jurgen Wittsiepea,�, Kerstin Schnella, Annett Hilbigb, Petra Schreya,Mathilde Kerstingb, Michael Wilhelma

aRuhr-Universitat Bochum, Abteilung fur Hygiene, Sozial- und Umweltmedizin, Universitatsstraße 150, D-44801 Bochum, GermanybForschungsinstitut fur Kinderernahrung, Heinstuck 11, D-44225 Dortmund, Germany

Received 25 September 2008; accepted 25 March 2009

Abstract

The daily dietary intake of nickel (Ni) and zinc (Zn) by 42 young children, 21 boys and 21 girls, from 4 to 7 years ofage, living in urban and rural areas of Germany and having different food consumption behaviour, was determined bythe duplicate method with a 7-day sampling period. Dietary records were also kept by the children’s parents for the7-day sampling period. Individual reported food items were identified, assigned to food groups and, together withknown Ni and Zn concentrations of foodstuffs, daily intake rates were calculated. The same method was used forcalculations of the energy, fat, protein and carbohydrate intake rates.

The levels in the food duplicates, determined by atomic absorption spectrometry, were in the range of 69–2000 mgNi/kgdry weight (geometric mean (GM): 348) and 7.1–43mg Zn/kgdry weight (GM: 17.5). Daily intake rates based on the294 individual food duplicate analyses were 12–560 mgNi/d (GM: 92.3) and 1.5–11mgZn/d (GM: 4.63). The resultsfrom the dietary records were 35–1050 mg Ni/d (GM: 123) and 1.7–15mg Zn/d (GM: 5.35).

The results of the daily intake rates from both methods showed a correlation with regard to Zn (r ¼ 0.56), but nocorrelation was found between either the Ni intake rates determined with both methods or between the Ni intake ratesmeasured by the duplicate method and calculated intake rates from the dietary records of energy, fat, protein,carbohydrates or drinking water. In the case of nickel, the discrepancies between the methods lead one to suppose thatthe main factors influencing Ni intake by food are not directly caused by easily assessable food ingredients themselves.It is possible that other factors, such as contaminated drinking water or the transition of Ni from kettles or otherhousehold utensils made from stainless steel into the food, may be more relevant. In addition there are some foodstuffswith great variations in concentrations, often influenced by the growing area and environmental factors. Further, somefood groups naturally high in Nickel like nuts, cocoa or teas might not have been kept sufficient within the records. Insummary, the dietary record method gave sufficient results for Zn, but is insufficient for Ni.

Based on the food duplicate analysis, children living in urban areas with consumption of food products from afamily-owned garden or the surrounding area and/or products from domestic animals of the surrounding area hadabout one-third higher Ni levels in their food than children either living in an urban area or children consumingproducts exclusively from the supermarket. Only slight differences were found with regard to Zn.

Compared to the recommendations of the German Society of Nutrition (DGE) (25–30 mgNi/d and 5.0mgZn/d), theparticipants of the study had a clearly increased Ni and, in view of the geometric mean value, a nearly adequate Zn

ee front matter r 2009 Elsevier GmbH. All rights reserved.

mb.2009.03.007

ing author.

ess: [email protected] (J. Wittsiepe).

www.elsevier.de/jtemb

dx.doi.org/10.1016/j.jtemb.2009.03.007

mailto:[email protected]

ARTICLE IN PRESSJ. Wittsiepe et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194184

intake. Health risks are especially given with regard to the influence of nickel intake by food on dermatitis for nickel-sensitive individuals.r 2009 Elsevier GmbH. All rights reserved.

Keywords: Nickel; Zinc; Dietary intake; Children; Germany

Introduction

Nickel is a metal with no clearly identified biologicalfunction in humans until now. It is thought to interactas a cofactor of metalloenzymes. Human exposure tonickel and the subsequent health effects have beensummarized in several reviews [1,2]. Nickel and nickelcompounds are well-recognized carcinogens. Nickel isalso a common sensitizing agent with high prevalence ofallergic contact dermatitis. For example, by using patchtesting with nickel sulphate in 6-year-old children livingin industrial and rural areas of West Germany, we founda prevalence of allergic sensitization ranging between5.0% and 30.7% [3,4]. There is increasing evidence thatthe oral intake of nickel can induce systemic contactdermatitis in nickel-sensitive individuals [4–8]. A recentmeta-analysis indicates that oral nickel exposure viafood and drinking water may lead to systemic contactdermatitis in a dose-dependent manner [9]. Nutritionthat is low in nickel can be used to treat the disease [5].The main exposure path of non-occupationally exposedindividuals is by the consumption of vegetables [2].However, there is lack of data on the dietary intakeof nickel. This is especially true for children. TheGerman Society of Nutrition (Deutsche Gesellschaft furErnahrung, DGE), the Austrian Society of Nutrition(Osterreichische Gesellschaft fur Ernahrung, OGE), theSwiss Society of Nutrition Research (SchweizerischeGesellschaft fur Ernahrungsforschung, SGE) and theSwiss Organization of Nutrition (Schweizerische Ver-einigung fur Ernahrung, SVE) recommend an estimatedappropriate intake of 25–30 mg/d [10] and the US Foodand Nutrition Board, FNB and Institute of Medicine(IOM) [11] published a tolerable upper intake levelof 200 (1–3 y) or 300 mg/d (4–8 y). In light of the highprevalence of nickel-sensitized children [3] and thesignificance of the oral nickel exposure on systemiccontact dermatitis, we re-examined samples and ques-tionnaires from our former duplicate study in whichsamples were collected in 1998 [12–14]. In addition tothe determination of Ni by the duplicate method, wealso calculated the dietary intake from the dietaryrecords of that investigation. In contrast to selenium [15]or zinc [16], it is not known if the assessment of dietaryrecords is suitable to estimate the nickel intake.

For comparison, we also studied the dietary intake ofzinc. Zinc is an essential trace element in humannutrition because it is a component of multiple enzymes

involved in the maintenance of the structural integrityof proteins and in the regulation of gene expression [17].Food of animal origin is a major path of intake into thehuman body. The supply of zinc to young childrenhas been shown to be suboptimal in the last decades[16,18,19]. The recommended intake rates as proposedby the DGE, OGE, SGE and SVE [10] (1–4 y: 3.0mg/d;4–7 y: 5.0mg/d; 7–10 y: 7.0mg/d) or by the Foodand Nutrition Board [11] (1–3 y: 3.0mg/d (tolerableupper intake level (UL): 7.0mg/d); 4–8 y: 5.0mg/d(UL: 12.0mg/d)) are often not achieved.

In general, several suitable and successful tools existfor the estimation of dietary intakes of substances.These include the duplicate method [20,21] or evaluationof dietary records [22]. Each of these methods have theirown advantages and disadvantages in expense, relia-bility and practicality. The results of the intake of metalsor metalloids measured by both methods have beencompared [15,16], but depend on the study designand the element. The dietary intake may also beinfluenced by regional differences in the elementcontents of various foods caused by anthropogenic orgeogenic sources. In the case of nickel and zinc, we havea further comparison of two elements where, on the onehand, the main food related route of human exposure isby vegetables and, on the other hand, by food of animalorigin.

In the present study, we have therefore used both, theduplicate method and the evaluation of dietary recordsto determine the intake of nickel and zinc in groupsof young children (age 4–7 years) with different foodconsumption behaviour and different places of residencewithin Germany.

Subjects and methods

Study design and subjects

The general study design of the duplicate studyperformed in 1998 and the results with regard to dioxinexposure [12] or intake of arsenic, cadmium, lead andmercury [13] or gold and platinum [14] have beendescribed in detail previously.

In brief: Three study groups with 14 children each,7 male and 7 female, at the age of 48–63 months, wereconducted in two areas of North-Rhine Westphalia

ARTICLE IN PRESS

Fig. 1. Study design.

J. Wittsiepe et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194 185

(Germany). Groups I and III consisted of children livingeither in an urban area (Ruhr District) or in a rural area(Steinfurt District) with a major portion of their foodconsumption coming from the families’ own vegetablegardens or the surrounding area and/or products fromtheir own domestic animals. Group II was made up ofchildren living in an urban area consuming foodexclusively from the supermarket (Fig. 1). The children’sweight and height were measured by the interviewer atthe beginning of the sampling period and were withinthe normal range for their ages (3rd–97th percentile).

During the sampling period, food duplicates of thetotal food intake, including drinks and snacks, werecollected on 7 consecutive days following the WHOguidelines [23]. The samples were homogenized, lyophi-lized and frozen at �20 1C until analysis. During thistime, the parents also kept a record of the type andamount of the children’s total food intake in standar-dized schedules using household measures (cup, plate,spoon, etc.), units (slices, peaces, etc.) and portion sizes(small, medium, large). All children consumed a mixeddiet and none of the children were on a special dietor were vegetarian. All further chemical analyses of thefood duplicates or data analyses of the dietary recordswere performed for each of the 294 individual samplingdays.

Analytical procedures for the determination of nickel

and zinc in the food duplicate samples

Approximately 500mg of the lyophilisate wereweighed in 90mL PTFE-tubes. After the addition of2.0mL of HNO3 (65%, puriss. p.a., Fluka, Buchs,Switzerland) and 1.0mL of H2O2 (30%, Baker, Deven-ter, Netherlands), the samples were digested with a MLS1200 mega microwave digestion system (MLS GmbH,Leutkirch, Germany) using the following temperatureprogram: 200W (2min), 0W (1min), 350W (2min),0W (1min), 450W (2min), 0W (1min), and 550W(2min). The digested sample was filled to 10.0mL withbi-distilled water in a polypropylene tube. This solutionwas used for the determination of Ni or Zn, respectively.

Nickel concentrations were measured by graphitefurnace atomic absorption spectrometry (AAS) using aPerkin-Elmer model SIMAA 6000 spectrometer

equipped with an autosampler AS-72 using standardconditions. The determination of Zn was carried out byflame AAS using a Analytik Jena AAS 6 vario spectro-meter, as well under standard conditions. For internalquality control, simulated diets, type C, D and E(National Food Administration, Uppsala, Sweden)were analyzed together with each digestion run andgood agreements were found between the certified andmeasured values (Nickel (mg/kg): type C: measured:N ¼ 19, 0.08670.016, certified: 0.10870.022; type D:measured N ¼ 11, 0.11770.018, certified 0.10070.029,type E: measured N ¼ 10, 0.11970.02, certified 0.12470.025; Zinc (mg/kg): type C: measured: N ¼ 12,37.671.9, certified 40.673.1; type D: measured N ¼ 7,33.474.2, certified 32.273.2; type E: measured N ¼ 3,36.170.5, certified 39.573.1).

Calculation of dietary intake and statistics

The daily intake per child in relation to the bodyweight was calculated on the basis of the measuredconcentrations of Ni and Zn in the food duplicates.A log-transformation was applied to all intake data toachieve an approximately normal distribution. Groups(regions/food consumption behaviour and sex) werecompared by a t test on log-transformed data. Associa-tions between the data were analyzed by Pearson’scorrelation. All statistical calculations were made usingthe Statistica data analysis software system, version 8.0(StatSoft, Inc., 2008, Tulsa, OK, USA).

Data analysis of dietary records

The data reported in the dietary records kept by theparents were processed for further analysis. First, theamounts were converted into mass units using standar-dized conversion factors [24]. Dishes, recipes andcommercial food, including baby food, were brokendown into their individual ingredients. The individuallyreported food items were identified and assigned to sub-groups. The daily intake of zinc was calculated basedon this data, together with the known typical zincconcentrations of foodstuffs, dishes, recipes and productsstored in the LEBTAB database [25] at the ResearchInstitute of Child Nutrition (FKE), Dortmund. The same


method was used for the calculation of energy, fat,protein and carbohydrate intake rates. Appropriate datafor nickel concentrations in different foodstuffs were notavailable at this time. Hence, the estimation of the nickelintake was performed using a simplified approach.Foodstuffs with relevant shares on total intake andwell-known nickel concentrations [26] were summari-zed in 16 food groups (nickel concentrations (mg/kg)in parenthesis): bread (94), fish (34), meat and meatproducts (143), vegetables excluding asparagus (144),cereals, rice and oats (367), pulse (2189), cocoa contain-ing products (1582), cake (no reliable data available andtherefore excluded), milk and milk products (55), pasta(190), fruits (108), seed and nuts (2018), black tea (6500),asparagus (110) and other foodstuffs (20). The dailyfood intakes of each food group as evaluated fromthe dietary records were multiplied with these nickelconcentrations and added to give the estimated totaldaily nickel intake. The calculations were either donewith SAS statistics software, version 8.2 or MicrosoftExcel, version 2003.

Results

Descriptive statistical data according to study groupsand gender on the total daily food intake based on foodduplicate measurements and values calculated fromdietary records are given in Table 1.

The measured food intake was significantly (po0.05)higher for Group I than for Group II. This was not the

Table 1. Daily food uptake based on food duplicate measurement

N MIN 5 10 50 90 95 MAX A

Total foodd (g/d)

Group I 98 684 797 905 1348 1806 2021 2386 1

Group II 98 615 689 829 1161 1733 1798 2139 1

Group III 98 644 765 858 1304 1750 1889 2224 1

Boys 147 615 858 938 1422 1812 2035 2386 1

Girls 147 644 766 805 1158 1582 1712 2021 1

All Groups 294 615 775 869 1278 1748 1865 2386 1

Total foodc (g/d)

Group I 98 721 895 950 1508 2039 2325 2755 1

Group II 98 720 925 1038 1374 1813 2101 2130 1

Group III 98 694 964 1056 1473 1998 2109 2557 1

Boys 147 694 944 1056 1554 2101 2226 2755 1

Girls 147 721 895 1000 1339 1800 1894 2325 1

All Groups 294 694 923 1038 1451 1981 2114 2755 1

Remarks: N ¼ number of samples; MIN ¼ minimum; 5, 10, 50, 90, 95 ¼ p

deviation of AM; GM ¼ geometric mean; CI GM ¼ 95% confidence intervaaFor log-transformed values in t-test; I/II comparison between Groups I

between Groups II and III; b/g comparison between boys and girls; p-value

case for the calculated food intake. Higher (po0.001)food intakes were found for boys in comparison togirls (geometric means (GMs): 1348/1159 g/d (duplicatemethod); 1522/1347 g/d (dietary records)) using bothmethods. Good correlation was found for all partici-pants (r ¼ 0.7645) and all subgroups (r: 0.6971–0.8166)between both determination methods (see Table 6and Fig. 2). The mean daily food intake rates obtainedfrom the dietary records were higher than thoseobtained from the duplicate measurements. In addition,general data on energy (3.17–12.26MJ/d; GM: 6.33MJ/d), protein (10.0–92.9 g/d; GM: 43.1 g/d), fat (13.1–203.2 g/d; GM: 62.7) and carbohydrate (66.0–354 g/d;GM: 175 g/d) intake based on the dietary records werecalculated (ranges and GM for all daily dietary records,N ¼ 294).

In the same way, Table 2 shows descriptive statisticaldata of the Ni and Zn concentrations in the foodduplicates. Significantly higher nickel concentrations(po0.001) were found in Group I (urban area with foodfrom region; GM: 426.8 mg/kgdry weight) in comparison toGroups II or III, whereas the mean concentrations inGroups II and III were nearly identical (geometricmean: 313.9 or 314.2 mg/kgdry weight). The zinc concen-trations in the food duplicates were in the range of7.1–43mg/kgdry weight and slightly higher values wereonly found in Group I in comparison to Group III(GMs: 18.50/16.29mg/kgdry weight). No differences in Nior Zn concentrations were found between boys andgirls.

The same observations – significantly higher Ni intakefor Group I – were made for the calculated daily intakes

s (d) and values calculated from dietary records (c).

M SD GM CI GM p-valuesa

I/II I/III II/III b/g

355 360 1307 1237–1381

233 335 1188 1124–1256

299 332 1256 1192–1325

396 360 1348 1289–1409

195 298 1159 1112–1207

296 345 1250 1211–1289 0.0167 0.3064 0.1509 0.0000

517 410 1462 1383–1545

409 319 1373 1312–1437

505 359 1462 1392–1535

574 399 1522 1458–1589

380 302 1347 1299–1396

477 366 1432 1391–1474 0.0868 0.9934 0.0642 0.0000

ercentile; MAX ¼ maximum; AM ¼ arithmetic mean; SD ¼ standard

l of GM.

and II; I/III comparison between Groups I and III; II/III comparison

s o0.05 are underlined.

ARTICLE IN PRESS

Fig. 2. Correlation between daily intake of total food (left) and zinc (right) as measured by the duplicate method (d) and as

calculated from dietary records and available data on levels in food groups (c).

Table 2. Nickel and zinc concentrations in the food duplicate samples.

N MIN 5 10 50 90 95 MAX AM SD GM CI GM p-valuesa


Ni (mg/kgdry weight)

Group I 98 130 190 200 430 790 830 1800 480.1 250.2 426.8 386.9–470.8

Group II 98 69 120 160 310 610 900 1600 367.9 243.4 313.9 280.9–350.7

Group III 98 85 140 150 310 630 790 2000 364.7 245.2 314.2 282.8–349.2

Boys 147 85 140 180 320 710 840 2000 394.5 266.3 337.3 309.0–368.2

Girls 147 69 140 180 360 740 820 1800 414.0 235.8 358.7 328.3–392.1

All Groups 294 69 140 180 340 720 830 2000 404.3 251.3 347.8 326.9–370.1 0.0001 0.0000 0.9885 0.3293

Zn (mg/kgdry weight)

Group I 98 7.1 10 12 19 26 30 35 19.31 5.590 18.50 17.42–19.65

Group II 98 7.7 9.7 12 18 26 28 43 18.53 5.828 17.65 16.56–18.81

Group III 98 7.4 9.1 10 16 25 26 40 17.17 5.649 16.29 15.25–17.40

Boys 147 7.1 9.6 11 18 26 27 43 18.48 5.879 17.56 16.64–18.52

Girls 147 7.6 10 12 17 26 28 35 18.19 5.612 17.36 16.50–18.26

All Groups 294 7.1 9.7 12 18 26 28 43 18.34 5.739 17.46 16.83–18.11 0.2877 0.0053 0.0848 0.7609

Remarks: N ¼ number of samples; MIN ¼ minimum; 5, 10, 50, 90, 95 ¼ percentile; MAX ¼ maximum; AM ¼ arithmetic mean; SD ¼ standard

deviation of AM; GM ¼ geometric mean; CI GM ¼ 95% confidence interval of GM.aFor log-transformed values in t-test; I/II comparison between Groups I and II; I/III comparison between Groups I and III; II/III comparison

between Groups II and III; b/g comparison between boys and girls; p-values o0.05 are underlined.


(Table 3) and body weight-related daily intake rates(Table 4), each based on food duplicate analysis. The Niintake for all children participating in the study wasin the range of 12–560 mg/d (GM: 91.27 mg/d) or0.63–31 mg/(kgbw d) (GM: 4.712 mg/(kgbw d)), respec-tively. With regard to zinc, we found slight differencesin daily intake rates between Group I and Groups II orIII and between boys and girls (Table 3) and slightdifferences in body weight-related daily intake rates

between Groups I and II, Groups II and III and betweenboys and girls (Table 4). The Zn intake for allparticipants was in the range of 1.5–11mg/d (GM:4.634mg/d) or 85–670 mg/(kgbw d) with a geometricmean of 236.2 mg/(kgbw d).

Results from the calculated daily Ni and Zn intakebased on the dietary records and available data on thelevels in food groups are given in Table 5. In contrast tothe corresponding data based on the duplicate samples,

ARTICLE IN PRESS

Table 3. Daily nickel and zinc uptake based on food duplicate analysis.



Ni (mg/d)

Group I 98 36 51 64 110 240 310 560 134.6 79.7 117.3 105.7–130.0

Group II 98 12 25 41 79 180 250 480 93.76 65.61 78.67 69.91–88.53

Group III 98 26 34 46 80 160 190 460 98.33 62.87 85.15 76.71–94.51

Boys 147 24 41 47 86 190 250 480 109.5 73.03 93.22 85.25–101.9

Girls 147 12 36 47 86 190 220 560 108.2 71.02 91.32 82.97–100.5

All Groups 294 12 39 47 86 190 240 560 108.9 71.91 92.27 86.44–98.48 0.0000 0.0000 0.3205 0.7565

Zn (mg/d)

Group I 98 2.4 2.9 3.3 5.1 8.1 9.0 10 5.372 1.815 5.081 4.749–5.436

Group II 98 1.6 2.2 2.7 4.5 7.6 8.2 11 4.758 1.795 4.433 4.104–4.789

Group III 98 1.5 2.1 2.7 4.5 6.8 7.9 9.4 4.723 1.694 4.417 4.094–4.765

Boys 147 1.6 2.4 3.1 4.9 7.9 9.0 11 5.219 1.937 4.856 4.556–5.177

Girls 147 1.5 2.6 2.8 4.5 6.8 7.9 8.8 4.684 1.587 4.421 4.179–4.678

All Groups 294 1.5 2.5 2.8 4.6 7.7 8.5 11 4.951 1.788 4.634 4.440–4.836 0.0090 0.0068 0.9458 0.0302


deviation of AM; GM ¼ geometric mean; CI GM ¼ 95% confidence interval of GM.aFor log-transformed values in t-test; I/II comparison between Groups I and II; I/III comparison between group I and III; II/III comparison


Table 4. Daily nickel and zinc uptake in relation to the body weight based on food duplicate analysis.



Ni (mg/(kgbw d))

Group I 98 1.7 2.8 3.1 5.7 12 16 31 6.838 4.389 5.900 5.311–6.553

Group II 98 0.63 1.5 2.3 4.3 9.6 13 31 5.112 3.832 4.268 3.791–4.805

Group III 98 1.2 1.6 2.1 4.0 7.9 9.7 26 4.821 3.254 4.155 3.738–4.618

Boys 147 1.2 2.1 2.4 4.4 9.6 13 31 5.601 3.963 4.774 4.375–5.21

Girls 147 0.63 1.7 2.4 4.7 9.7 12 31 5.580 3.933 4.651 4.216–5.13

All Groups 294 0.63 1.9 2.4 4.5 9.6 12 31 5.590 3.942 4.712 4.414–5.03 0.0001 0.0000 0.7368 0.6938

Zn (mg/(kgbw d))

Group I 98 120 150 170 250 400 430 510 269.4 89.15 255.4 239.1–272.8

Group II 98 85 120 130 250 390 410 670 256.4 94.58 239.6 222.1–258.5

Group III 98 87 110 120 230 320 370 450 229.1 77.48 215.4 200.1–231.8

Boys 147 85 120 170 250 390 410 670 264.1 91.46 248.5 234.4–263.5

Girls 147 87 120 130 230 360 400 510 239.2 84.27 224.5 211.6–238.2

All Groups 294 85 120 140 240 380 410 670 251.6 88.67 236.2 226.5–246.3 0.2084 0.0007 0.0464 0.0166




J. Wittsiepe et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194188

the Ni and Zn intake rates based on the dietary recordscannot reflect expected regional differences by differentcontaminated foodstuffs, because all data were calcu-lated with the same concentration factors. The differ-ences observed can therefore only be caused by differentfood composition. Differences in nickel intake were notobserved between any of the three subgroups. Only the

calculated values for girls were higher than those ofboys. In the case of zinc, we found differences betweenGroups I/II and I/III. Fig. 3 shows histograms of dailynickel and zinc intake measured by the duplicate methodin comparison to recommended intakes and tolerableupper intake levels by the DGE [10] and the FNB &IOM [11].

ARTICLE IN PRESS

Table 5. Daily nickel and zinc uptake calculated from dietary records and available data on levels in food groups.



Ni (mg/d)

Group I 98 45.8 44.3 68.4 121 279 117 626 147.9 101.3 125.1 112.0–139.8

Group II 98 45.9 48.2 65.6 104 299 107 1050 152.0 161.6 118.0 104.5–133.3

Group III 98 35.0 49.1 66.1 123 242 99.4 538 142.4 76.74 126.6 115.0–139.4

Boys 147 49.3 50.8 70.1 112 196 112 538 126.9 67.78 114.8 107.1–123.0

Girls 147 35.0 45.0 63.9 116 320 99.8 1050 168.0 150.5 132.2 119.1–146.8

All Groups 294 35.0 46.0 66.5 115 257 108 1050 147.4 118.3 123.2 115.7–131.2 0.4840 0.8708 0.3707 0.0268

Zn (mg/d)

Group I 98 3.0 3.5 3.8 5.6 9.3 10 15 6.128 2.184 5.796 5.426–6.192

Group II 98 1.9 2.8 3.3 5.3 7.0 9.0 12 5.326 1.676 5.071 4.755–5.407

Group III 98 1.7 2.6 2.9 5.5 8.8 10 12 5.617 2.165 5.209 4.807–5.645

Boys 147 1.7 3.2 3.6 5.6 9.0 9.9 12 5.901 2.099 5.535 5.214–5.877

Girls 147 2.0 2.9 3.2 5.3 7.6 9.1 15 5.48 1.969 5.170 4.890–5.466

All Groups 294 1.7 2.9 3.3 5.5 8.3 9.8 15 5.691 2.042 5.350 5.136–5.573 0.0044 0.0429 0.6046 0.1002




Fig. 3. Daily intake of nickel (mg/d) (left) and zinc (mg/d) (right) as measured by the duplicate method in comparison to the

recommended uptake rates by the DGE [10] and the FNB & IOM [11] and tolerable upper intake levels by the FNB [11] for children.


Moderate correlations were found between Zn intakesmeasured by the duplicate method versus calculationsfrom the dietary records for all participants (r ¼ 0.5585,see Table 6 and Fig. 2) and between the measurementson total food and Zn intake for both methods (see Fig.3). Poor correlations were found between most othertotal food, Zn, or Ni intake rates shown in Table 6.However, unexpectedly, no correlations were found forall groups between the two Ni intake rates, or between

the Ni intake measured by the duplicate method and anyof the calculated intake rates of total food, energy,protein, fat, carbohydrates or drinking water.

Discussion

To evaluate the necessary supply and the risk due tothe dietary intake of Ni and Zn, the data were compared

ARTICLE IN PRESS

Table 6. Pearson’s correlation coefficients (r) for subgroups and all participants between the daily intake of total food, Ni and Zn

as measured by the duplicate method (d) and the same parameters and energy, protein, fat, carbohydrates and drinking water as

calculated from dietary records and available data on levels in food groups (c).

Group N Total foodc Nid Nic Znd Znc

Total Foodd Group I 98 0.7347 0.1324 0.2763 0.2826 0.4100

Group II 98 0.8166 0.2507 0.0425 0.5904 0.5192

Group III 98 0.7513 0.1243 0.1450 0.5358 0.3995

Boys 147 0.6971 0.2888 0.2440 0.4460 0.3111

Girls 147 0.7763 0.1250 0.1515 0.4558 0.5373

All Groups 294 0.7645 0.1916 0.1280 0.4711 0.4463

Total Foodc Group I 98 0.0736 0.3458 0.1907 0.4689

Group II 98 0.1486 �0.0031 0.4636 0.5695

Group III 98 �0.0138 0.1252 0.4298 0.5230

Boys 147 0.2393 0.1988 0.3107 0.4237

Girls 147 �0.0212 0.2375 0.3348 0.5792

All Groups 294 0.0877 0.1313 0.3500 0.5181

Nid Group I 98 0.2043 0.3948 0.3243

Group II 98 0.0260 0.1878 0.0440

Group III 98 �0.0305 0.1988 0.0405

Boys 147 0.1425 0.4363 0.4317

Girls 147 �0.0641 0.1982 �0.0292

All Groups 294 0.0690 0.3006 0.1905

Nic Group I 98 0.1569 0.4946

Group II 98 0.2513 0.2117

Group III 98 0.3055 0.3499

Boys 147 0.3527 0.3804

Girls 147 0.1608 0.3137

All Groups 294 0.2220 0.3065

Znd Group I 98 0.4442

Group II 98 0.6349

Group III 98 0.6039

Boys 147 0.5456

Girls 147 0.5595

All Groups 294 0.5585

Energyc All Groups 294 0.5287 0.0748 0.2273 0.3753 0.6119

Fatc All Groups 294 0.3257 0.0603 0.1154 0.3212 0.5488

Proteinc All Groups 294 0.4123 0.0549 0.2834 0.4222 0.7865

Carbohydratesc All Groups 294 0.5016 0.0788 0.2108 0.2308 0.3058

Drinking Waterc All Groups 294 0.5154 0.0710 �0.1474 0.1837 0.0929

p-valueso0.05 are underlined.

J. Wittsiepe et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194190

with the recommendations of the DGE [10] and theFNB & IOM [11] with regard to children (Fig. 3).For Ni, 96.2% of the daily intakes determined by theduplicate method exceeded the recommended intake of25–30mg/d, and the tolerable upper intake level of 300mg/dwas exceeded by at least 8 samples. Based on the findingsfrom the duplicate study, Ni exposure of young childrenvia food can generally be classified as high.

The 36% higher Ni concentrations in the foodduplicates from Group I, in comparison to Groups IIor III, indicate that there might be an environmentallycaused additional intake through home-grown vegetablesor domestic animals from the urban area. Within the

residential area of Group I, a hot spot with elevatednickel ambient air levels from the steel producingindustry is known. Taking into account the nickeluptake by plants, such as that known from mossmonitoring measurements [27,28], the enrichment inthe food chain caused by elevated nickel ambient airlevels might be the reason for these observations.

Recent research [3] on children living at steel millhot spot areas in the Ruhr District revealed a highprevalence of allergic sensitizations, especially againstnickel. Within that study, the data were associated withthe current internal nickel exposure (urine levels), andnickel in ambient air was positively associated with the

ARTICLE IN PRESS

Table 7. Comparison of Ni and Zn uptake by children with literature data.

Country Year Study type Age (years) P50 AM Reference

Nickel mg/dGermany 1998 Duplicate study 4–6 86 108.9 This study

Dietary records 115 147

Canada 1986–88 Basket study 1–4 190 [32]

5–11 271

France 2000 Total diet/basket study 3–14 87 [33]

Sweden ? Duplicate study 11–14 410 [34]

United Kingdom 1984 Duplicate study 1.75–2.2 88 [35]

USA 1984 Total diet study 2 90.4 [36]

USA 1991–97 Recall 4–8 99.29 [37]

Zinc mg/d

Germany 1998 Duplicate study 4–6 4.6 4.95 This study

Dietry records 5.49 5.69

Germany 1998 Consumer sample 4–6 ~ 6.0 [38]

# 6.4

Germany 1995 Dietary records 1.5–5.3 5.7 6.38 [19]

Germany 1988/89 Duplicate study 5–9 5.3 [16]

Dietary records 5.6

Belgium ? Duplicate study 2–3 7.5 [39]

France 2000 Total diet/basket study 3–14 7.3 [33]

India 1986–1994 Duplicate study 6–10 6.7 (geometric mean) [40]

Japan 2000 Duplicate study 3–6 4.93 [41]

Netherlands 1987/88 Total diet/basket study 4–7 ~ 6.7 [42]

# 7.1

Sweden ? Duplicate study 11–14 8.18 [34]

Sweden 1998 Duplicate records 4.770.8 7.1 [43]

United Kingdom 1992 Duplicate study 0.3–1.2 2.8 (geometric mean) [44]

United Kingdom 1993 Dietary records 4–6 4.62 [45]

USA 1988–94 Recall (NHANES III) 4–6 7.6 7.7 [46]

USA 1999–2000 NHANES o6 7.3 8.1 [47]


frequency of allergic symptoms. The investigations andthe knowledge about the influence of nickel exposurevia food on nickel allergies leads one to suppose thatregional contaminated food may be partly responsiblefor these findings.

The assessment of the Zn levels in the food duplicatesand of the resulting daily intake rates lead to a morepositive balance: the median and geometric mean of theintake of about 4.6mg/d are near to the recommendedintakes by the FNB & IOM or the DGE of 5.0mg/d,respectively. The recommended values are reached by44.6% or 25.2% of all participants and no value exceedsthe tolerable upper intake level of 12.0mg/d, as given bythe FNB & IOM. The differences between the threesubgroups are only moderate, so that no meaningfulgroup-specific exposure is observable. Zinc is ubiquitousin the environment. Within the food chain, high zincconcentrations are normally found in food of animalorigin, especially internal organs like liver or kidney.This is in good agreement with the correlations wefound between the Zn and protein intakes.

As previously mentioned, the duplicate and dietaryrecord methods have their own general method specificcharacteristics. In addition, as known from other studies[15,16], we found lower mean daily intake rates with theduplicate method (108.9 mg/dNi and 4.95mg/dZn) incomparison to the dietary record method (147.4mg/d Niand 5.69mg/d Zn), even if both datasets representexactly the same study days in the present study. Somespecific problems have to be discussed, especially withregard to nickel intake: Based on the study design, it canbe expected that a regional different nickel intake causedby the environmental contamination of home-grownvegetables or domestic animals might have taken place.Detailed data on regional specific food contaminationwere not available and could therefore not be includedin the calculations. Further, some plants and thefoodstuffs produced from them, such as nuts, cocoaand chocolate or some green, black, fruit or other herbteas, typically have very high nickel concentrations.Intakes of these foodstuffs are often not preciselydeterminable and the concentration can vary depending


on the growing areas. As a third point of discussion, theintake of nickel by drinking water cannot be exactlydetermined. As known from our own measurements,drinking water stagnating in pipes overnight oftenshows high nickel concentrations up to some hundredmg/L, especially caused by nickel emission from thefittings. The concentrations decline again to normallevels when the water is run for a while. Further nickelcan be emitted by kettles [29,30] and other householdappliances made from stainless steel [31]. All theseadditional sources of possible nickel intake can varyextremely and cannot be exactly quantified with regardto the evaluation of dietary records. The listing showsthat the exact determination of nickel intake rates withthe dietary record method is difficult.

The duplicate method records all these differentexposure routes. Only during the process of homogeniza-tion did our food duplicates have contact with a blendermade of steel, but blank sample experiments showed nocontamination. When comparing both study types, thevalues measured by the duplicate method seem to be morereliable. Since we found no correlations with the amountof food consumption, with any parameters of the foodcomposition, or with the values calculated with the dietaryrecord method, it must be assumed that the major partof nickel intake into food does not take place by thefoodstuffs themselves, but probably by drinking watercontamination during cooking, or rather that it isinfluenced only by some highly contaminated ingredients(e.g., cocoa, nuts, tea, etc.).

Table 7 gives a comparison of our findings withresults from recent studies on dietary intake of Ni andZn by young children. Only a few studies measured theintake of nickel. Except for a Swedish duplicate studyreporting about four-fold higher levels for 11–14-year-old children, and a Canadian basket study with twice ashigh intake rates, all other reported intake rates were inthe same range as the measurements presented in thepresent study. The intake of Zn has been determined onmore occasions and, astonishingly, the different meanintake rates found worldwide are close together in arange of about 3–8 mg/d.

Conclusion

The study can be summarized as follows:

�
The estimated appropriate intake level for Ni(25–30 mg/d) published by the DGE was clearlyexceeded by most participants. Even the tolerableupper intake level of 300 mg/d given by the FNB &IOM was exceeded in some sampling days. However,it must be mentioned that intakes lower than 30 mg/dwould be difficult unless nutritious foods like nuts,pulses, and vegetables were eliminated from the diet.
�
The mean daily intake of Zn is just under therecommended value of DGE for young children, butcan be assessed as sufficient. � Exposure of children living in the urban area of the
Ruhr District with food consumption includinghome-grown vegetables and domestic animals havea one-third higher intake of nickel (probably causedby emissions from the steel producing industries)in comparison to those living in a rural area or thosewith consumption of foodstuffs exclusively from thesupermarket. Nickel exposure by food naturally highin Nickel might be responsible for the exacerbation ofallergic symptoms found for children living in thisregion.
� Zinc is a ubiquitous element and only minor
differences between the three study groups werefound for its intake.
� The use of the dietary record method for the
determination of Zn intake gave sufficiently reliabledata in comparison to those from the duplicatemethod.
� In the case of nickel, the results from the dietary
record study did not agree with those from theduplicate study. For estimation of Ni intake usingdietary records, the variability of the nickel contentsin food groups of different origins, the survey of evensmall amounts of highly contaminated food groups,and/or the necessary database on contents in food-stuffs is not sufficient at present day. Further, otherrelevant and varying factors, such as the intake bycontaminated drinking water, and the contaminationby cooking water or during the cooking process, canin principle not be considered by the method.

Acknowledgements

Our thanks are expressed to the children involved inthe study, their parents and to Hans-Jurgen Kozlowski,Sylvia Kmiecik, Thomas Koster, Eva Schmidt, KarstenHolinski and Jorg Restemeyer, who helped to collect,prepare and analyze the samples. The zinc measure-ments were performed at the Department of AnalyticalChemistry of the Ruhr-University Bochum, and theauthors thank Karin Bartholomaus from the sameinstitution for technical assistance. We thank the Ger-man Federal Environmental Agency (Umweltbunde-samt) for financial support (FKZ-No. 29765124).

References

[1] Barceloux DG. Nickel. Clin Toxicol 1999;37(2):239–58.

[2] Cempel M, Nikel G. Nickel: a review of its sources and

environmental toxicology. Pol J Environ Stud 2006;15(3):

375–82.

ARTICLE IN PRESSJ. Wittsiepe et al. / Journal of Trace Elements in Medicine and Biology 23 (2009) 183–194 193

[3] Wilhelm M, Eberwein G, Holzer J, Gladtke D, Angerer J,

Marczynski B, et al. Influence of industrial sources

on children’s health – hot spot studies in North Rhine

Westphalia, Germany. Int J Hyg Environ Health 2007;

210(5):591–9.

[4] Nielsen GD, Soderberg U, Jorgensen PJ, Templeton DM,

Rasmussen SN, Andersen KE, et al. Absorption and

retention of nickel from drinking water in relation to food

intake and nickel sensitivity. Toxicol Appl Pharmacol

1999;154(1):67–75.

[5] D’Ambrosio FP, Bagnato GF, Guarneri B, Musarra A,

Di Lorenzo G, Dugo G, et al. The role of nickel in foods

exacerbating nickel contact dermatitis. Allergy 1998;

53(46):143–5.

[6] Christensen JM, Kristiansen J, Nielsen NH, Menne T,

Byrialsen K. Nickel concentrations in serum and urine of

patients with nickel eczema. Toxicol Lett 1999;108(2–3):

185–9.

[7] Sharma AD. Relationship between nickel allergy and

diet. Indian J Dermatol Venereol Leprol 2007;73(5):

307–12.

[8] Jensen CS, Menne T, Lisby S, Kristiansen J, Veien NK.

Experimental systemic contact dermatitis from nickel: a

dose–response study. Contact Dermatitis 2003;49(3):

124–32.

[9] Jensen CS, Menne T, Johansen JD. Systemic contact

dermatitis after oral exposure to nickel: a review with a

modified meta-analysis. Contact Dermatitis 2006;54(2):

79–86.

[10] DGE. Referenzwerte fur die Nahrstoffzufuhr. In:

Deutsche Gesellschaft fur Ernahrung, OsterreichischeGesellschaft fur Ernahrung, Schweizerische Gesellschaft

fur Ernahrungsforschung, Schweizerische Vereinigung furErnahrung, eds. Konzeption und Entwicklung: Arbeits-

gruppe ‘‘Referenzwerte fur die Nahrstoffzufuhr’’. 1 ed.

Frankfurt am Main: Umschau/Braus 2000.

[11] Trumbo P, Yates AA, Schlicker S, Poos M. Dietary

reference intakes: vitamin A, vitamin K, arsenic, boron,

chromium, copper, iodine, iron, manganese, molybde-

num, nickel, silicon, vanadium, and zinc. J Am Diet Assoc

2001;101(3):294–301.

[12] Wittsiepe J, Schrey P, Wilhelm M. Dietary intake of

PCDD/F by small children with different food consump-

tion measured by the duplicate method. Chemosphere

2001;43(4–7):881–7.

[13] Wilhelm M, Wittsiepe J, Schrey P, Hilbig A, Kersting M.

Consumption of homegrown products does not increase

dietary intake of arsenic, cadmium, lead, and mercury by

young children living in an industrialized area of

Germany. Sci Total Environ 2005;343(1–3):61–70.

[14] Wittsiepe J, Schrey P, Wilhelm M, Begerow J, Dunemann

L. Dietary intake of platinum and gold by children from

Germany using duplicate portion sampling. J Trace Elem

Med Biol 2003;17(2):117–22.

[15] Duffield AJ, Thomson CD. A comparison of methods of

assessment of dietary selenium intakes in Otago, New

Zealand. Br J Nutr 1999;82(2):131–8.

[16] Laryea MD, Schnittert B, Kersting M, Wilhelm M,

Lombeck I. Macronutrient, copper, and zinc intakes of

young German children as determined by duplicate food

samples and diet records. Ann Nutr Metab 1995;39(5):

271–8.

[17] Maret W, Sandstead HH. Zinc requirements and the risks

and benefits of zinc supplementation. J Trace Elem Med

Biol 2006;20(1):3–18.

[18] Michaelsen KF, Samuelson G, Graham TW, Lonnerdal

B. Zinc intake, zinc status and growth in a longitudinal

study of healthy Danish infants. Acta Paediatr 1994;

83(11):1115–21.

[19] Schrey P, Wittsiepe J, Budde U, Heinzow B, Idel H,

Wilhelm M. Dietary intake of lead, cadmium, copper

and zinc by children from the German North Sea

island Amrum. Int J Hyg Environ Health 2000;203(1):

1–9.

[20] Thomas KW, Sheldon LS, Pellizzari ED, Handy RW,

Roberds JM, Berry MR. Testing duplicate diet sample

collection methods for measuring personal dietary ex-

posures to chemical contaminants. J Expo Anal Environ

Epidemiol 1997;7(1):17–36.

[21] Kroes R, Muller D, Lambe J, Lowik MRH, van Klaveren

J, Kleiner J, et al. Assessment of intake from the diet.

Food Chem Toxicol 2002;40(2–3):327–85.

[22] Rockett HRH, Berkey CS, Colditz GA. Evaluation of

dietary assessment instruments in adolescents. Curr Opin

Clin Nutr Metabol Care 2003;6(5):557–62.

[23] WHO. Guidelines for the study of dietary intakes of

chemical contaminants. Geneva: World Health Organiza-

tion; 1985.

[24] Zacharias R, Durr H. Lebensmittelverarbeitung im

Haushalt. 5th ed. Stuttgart: Verlag Eugen Ulmer; 1992.

[25] Kroke A, Manz F, Kersting M, Remer T, Sichert-Hellert

W, Alexy U, et al. The DONALD Study. History, current

status and future perspectives. Eur J Nutr 2004;43(1):45–54.

[26] Souci SW, Fachmann W, Kraut H. Food composition

and nutrition tables. 6th ed. Stuttgart: Medpharm

Scientific Publishers; 2000 [see also: /http://www.sfk-

online.netS].

[27] Herpin U, Siewers U, Markert B, Rosolen V, Breulmann

G, Bernoux M. Second German heavy-metal survey by

means of mosses, and comparison of the first and second

approach in Germany and other European countries.

Environ Sci Pollut Res Int 2004;11(1):57–66.

[28] Harmens H, Norris DA, Koerber GR, Buse A, Steinnes

E, Ruhling A. Temporal trends in the concentration of

arsenic, chromium, copper, iron, nickel, vanadium and

zinc in mosses across Europe. Atmos Environ 2007;

41(31):6673–87.

[29] Helmers E. Migration of nickel from water cookers into

drinking water. Deut Lebensm Rundsch 1998;94(2):50–2.

[30] Berg T, Petersen A, Pedersen FA, Petersen J, Madsen C.

The release of nickel and other trace elements from

electric kettles and coffee machines. Food Addit Contam

2000;17(3):189–96.

[31] Kuligowski J, Halperin KM. Stainless steel cookware as a

significant source of nickel, chromium, and iron. Arch

Environ Contam Toxicol 1992;23(2):211–5.

[32] Dabeka RW, McKenzie AD. Survey of lead, cadmium,

fluoride, nickel, and cobalt in food composites and

estimation of dietary intakes of these elements by

Canadians in 1986–1988. J AOAC Int 1995;78(4):897–909.

http://www.sfk-online.net

http://www.sfk-online.net


[33] Leblanc JC, Guerin T, Noel L, Calamassi-Tran G,

Volatier JL, Verger P. Dietary exposure estimates of 18

elements from the 1st French total diet study. Food Addit

Contam 2005;22(7):624–41.

[34] Abdulla M, Behbehani A, Dashti H. Dietary-intake and

bioavailability of trace-elements. Biol Trace Elem Res

1989;21:173–8.

[35] Smart GA, Sherlock JC, Norman JA. Dietary intakes of

lead and other metals: a study of young children from an

urban population in the UK. Food Addit Contam

1988;5(1):85–93.

[36] Pennington JA, Jones JW. Molybdenum, nickel, cobalt,

vanadium, and strontium in total diets. J Am Diet Assoc

1987;87(12):1644–50.

[37] Food and Nutrition Board (FNB), Institute of Medicine

(IOM). Dietary reference intakes for vitamin A, vitamin

K, arsenic, boron, chromium, copper, iodine, iron,

manganese, molybdenum, nickel, silicon, vanadium, and

zinc. Washington, DC: National Academies Press; 2001.

[38] DGE. Ernahrungsbericht 2004. Bonn: Deutsche Ge-

sellschaft fur Ernahrung e. V.; 2004.

[39] Bosscher D, Van Cauwenbergh R, Robberecht H, Van

Caillie-Bertrand M, Deelstra H. Daily dietary iron, zinc

and copper intake of infants in Belgium. Eur Food Res

Technol 2002;215:275–8.

[40] Raghunath R, Tripathi RM, Kumar AV, Sathe AP,

Khandekar RN, Nambi KS. Assessment of Pb, Cd, Cu,

and Zn exposures of 6- to 10-year-old children in

Mumbai. Environ Res 1999;80(3):215–21.

[41] Aung NN, Yoshinaga J, Takahashi JI. Dietary intake

of toxic and essential trace elements by the children

and parents living in Tokyo Metropolitan Area, Japan.

Food Addit Contam 2006;23(9):883–94.

[42] van Dokkum W, Brussard JH. Estimating dietary

exposure to contaminants. In: Kardinaal AFM, LowikMRH, van der Heij DG, editors. Dietary exposure to

contaminants and additives: risk assessment in Europe.

Noordwijkerhout, Netherlands: TNO Topics in Nutrition

and Food Research; 1995 p. 31–4.

[43] Sepp H, Lennernas M, Pettersson R, Abrahamsson L.

Children’s nutrient intake at preschool and at home. Acta

Paediatr 2001;90(5):483–91.

[44] Richmond J, Strehlow CD, Chalkley SR. Dietary-intake

of Al, Ca, Cu, Fe, Pb, and Zn in infants. Br J Biomed Sci

1993;50(3):178–86.

[45] Watt RG, Dykes J, Sheiham A. Socio-economic

determinants of selected dietary indicators in British

pre-school children. Public Health Nutr 2001;4(6):

1229–33.

[46] Briefel RR, Bialostosky K, Kennedy-Stephenson J,

McDowell MA, Ervin RB, Wright JD. Zinc intake of

the US population: findings from the third National

Health and Nutrition Examination Survey, 1988–1994.

J Nutr 2000;130(5S Suppl.):1367S–73S.

[47] Ervin RB, Wang CY, Wright JD, Kennedy-Stephenson J.

Dietary intake of selected minerals for the United

States population: 1999–2000. Adv Data 2004;27(341):

1–5.

Documents

Dietary intake of nickel and zinc by young children – Results from food duplicate portion measurements in comparison to data calculated from dietary records and available data on