3,326
Views
252
CrossRef citations to date
0
Altmetric
Original Articles

Dietary exposure estimates of 18 elements from the 1st French Total Diet Study

, , , , &
Pages 624-641 | Received 02 Sep 2004, Accepted 21 Mar 2005, Published online: 20 Feb 2007

Abstract

To estimate the dietary exposure of the main minerals and trace elements from retail food typically consumed by the French population, samples were purchased and then prepared and cooked prior to analysis. A total of 1080 individual food composites samples were collected and analysed for 18 elements (arsenic, lead, cadmium, aluminium, mercury, antimony, chrome, calcium, manganese, magnesium, nickel, copper, zinc, lithium, sodium, molybdenum, cobalt and selenium). Intakes were calculated from different food consumption patterns found in France for average and high consumers among adults and children. Dietary exposures of those consumers estimated from the France 2000 Total Diet Study (FTDS) are reported, and compared with existing nutritional reference values (Lowest threshold Intake, LTI) or toxicological reference values (Provisional Tolerable Weekly Intake, PTWI or Upper Level, UL) of the respective element and from previous French studies. This study confirms for the populations concerned, the low probability of nutritional or health risks due to food consumption.

Introduction

Estimation of food intakes is necessary for risk evaluation, and possibly to determine relationships between adverse effects observed in humans and exposure to particular substances (FAO/WHO Citation1985). These exposure assessments are also useful when making decisions on the regulation of nutritional or chemical products and the safety of food products. The Total Diet Study (TDS) is one method used in many countries to address this issue (WHO, IWTDS, 1999, 2002, 2004). It has the advantage of yielding more refined exposure data, in that foods are analysed “as-consumed” by the eater, which is a different approach from the current French system based on cross-referencing contamination data for raw food materials obtained from state monitoring plans with national individual food consumption data. The Ministry of Agriculture and Fishes and the national institute of agronomical research (INRA) decided in 2000 to implement with the collaboration of the French agency of food safety (AFSSA) the first French total diet study (FTDS) for essential and non-essential trace elements, minerals and mycotoxins. The design of the FTDS based from a representative “individual food market basket”, prepared as consumed has been described in detail elsewhere (Leblanc et al. Citation2005), but basically involves a list of 338 food items corresponding to the analysis of 1080 individual food composites samples. Only results dealing with minerals and trace elements issue from adults and children diet (300 food list item corresponding to 998 individual food composites samples), excluding vegetarian diet (82 individual food composites samples) obtained in the FTDS are presented here.

In order to get representative data concerning the effective amount of elements ingested by French consumers, a sampling plan for foods was applied using a “market basket” approach and foods were prepared on an “as consumed” basis. The 338 foods of the list were obtained from the national individual dietary survey (CREDOC-AFSSA-DGAL Citation2000) after applying the criteria of intake greater than 1 g/day/person for all foods and greater than 0.5 g/day for specific foods less frequently eaten but with a high mineral content, such as offal or nuts. Each of the sampled individual foods in the list was obtained from an individual composite sample of five sub-samples weighted according to the brand market shares and actual distribution channels described by the Secodip household purchasing panel (Combris et al. Citation2000).

Two food groups, three locations and two campaigns of purchasing the foods were defined and programmed in the sampling in order to take into account as far as possible variations due to contamination levels with respect to their production and/or processed methods, eating behaviour and seasonality. The first group includes so-called national foods (N), since these are processed foods sold in every region. These foods were therefore bought in one region only (Paris was chosen). The second group includes so-called regional foods (R), since these are unprocessed and their origin may vary according to regions. Three regions were chosen on the basis of the disparity of the origin of raw foods like seafood or vegetables consumptions modes: in the West Lorient (Rlo), in the South-East Lyon (Rly) and in the Center-North Paris (Rrp). Concerning sampling campaign, the first one covered the spring/summer 2000 and the second the autumn/winter 2001. The “as-consumed” foods were prepared in an experimental kitchen of the Ecole Superieure de Cuisine Française (ESCF) in Paris’ 6th District by a ESCF-qualified cook understanding the rudiments of cooking and the qualitative and quantitative aspects of the purchase and preparation of composed dishes (type and quantities of products, ingredients, etc.). The food were cooked “as-consumed” when necessary, then mixed, ground and blended to obtain a homogenized individual composite sample of the national or regional foods for each season and frozen until analysis. Samples that were bought in different seasons were never mixed prior to analysis. All the sampling brought the number of individual composite samples to 1080 for minerals and trace elements and sent to the French national reference laboratory (NRL) for analysis.

The mean concentrations from the 998 composites samples for the 300 individual foods item of the FTDS's list are used together with data on the consumption at the same aggregate level of these foods to estimate dietary exposure for mean and high percentile (95th and 97.5th) of individual adult and children consumers. According to the Goldberg cut-off limit (Goldberg et al. Citation1991) based on comparison between energy intake and basal metabolic rate, under-reporters were excluded from the analysis.

In general, for an exposure by ingestion of a given nutritional element and/or toxic element, a Lowest Threshold Intake or a minimal nutritional requirement (LTI or BNM, SCF Citation1993, AFSSA-CNERNA-CNRS Citation2001) and/or a Provisional Maximum Tolerable Weekly Intake (PMTDIs) or Provisional Tolerable Weekly Intake (PTWIs) for non-essential trace elements or of Upper Levels (ULs) for minerals and trace elements is defined to assess the safety of foods consumed by the general French population. Lowest Threshold Intake (LTI) is corresponding to the quantity of necessary nutrient to assure maintenance, metabolic and physiological function for an individual in good health. It is defined as being equal to the average requirements – 2 standard deviations. But it must be kept in mind that certain population groups could have greater requirements to this minimal value. It is also important to bear in mind that this paper presents toxicological reference value defined for chronic, or sub-chronic exposure of the general population. The value thus defined corresponds to an estimation of the quantity of the element to which individuals can theoretically be exposed every day of their lives, without this ingestion having adverse impact on their health.

This paper reports mean concentration and dietary exposure for 18 elements from the 2000 France Total Diet Study. The elements discussed are: Arsenic, lead, cadmium, aluminium, mercury, antimony, chrome, calcium, manganese, magnesium, nickel, copper, zinc, lithium, sodium, molybdenum, cobalt and selenium.

Materials and methods

During the preparation or transfer of the prepared “as-consumed” food the use of storage containers and cooking utensils made of aluminium, ceramic or enamel, or of any other packaging material, was excluded. Suitable stainless steel cooking equipment was used to prepare the food. In accordance with the good laboratory practices and internal quality procedures of Environmental Inorganic Contaminants and Mineral Unit of the AFSSA-LERQAP, in compliance with the ISO/IEC 17025 norm (ISO/IEC 17025 1999), the equipment used for preparing and grinding the composite samples was thoroughly washed between each preparation to avoid risks of cross-contamination prejudicial to the study, notably for analysis of trace elements. Moreover, in compliance with international WHO/GEMS/Food recommendations, the food was prepared when necessary using drinking water from the regions where the foods (e.g., pasta, coffee, etc.) were bought.

All the sampling brought the number of individual composite samples to 606 for regional foods and 392 for national foods, or a total of 998 composite samples prepared for the study and stored at −20°C in suitable 100 ml containers until their analysis (see for sampling details).

Table I. Sampling of the FTDS for elements.

The elementary analyses (about 18 000 results in all) were carried out by the Environmental Inorganic Contaminants and Mineral Unit of the AFSSA-LERQAP, which is the national reference laboratory. All the 998 individual food composite samples were homogenized and digested (about 0.6 g taken from each sample) in the quartz vessels with suprapure nitric acid (3 ml) using Multiwave closed microwave system (Anton-Paar, Courtaboeuf, France). The total content of all selected essential and non essential trace elements in the foods was determined by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) (VG PlasmaQuad ExCell-Thermo Electron, Coutaboeuf, France), a very powerful technique for quantitative multi-elemental analysis.

All solutions were prepared with analytical reagent grade chemicals and ultra pure water (18 MΩ-cm) generated by purifying distilled water with the Milli-Q™ PLUS system associated to an Elix 5 pre-system (Millipore S.A., St Quentin en Yvelines, France).

A total of 1000 mg l−1 standard stock solutions of magnesium nitrate, calcium nitrate, sodium chloride, iron (III) nitrate, copper (III) nitrate, zinc nitrate, tin (IV) chloride, antimony (III) chloride, ammonium heptamolybdate, a 10 mg l−1 rhodium (III) nitrate and Suprapur HNO3 65% (v/v) were purchased from Merck (Merck-Eurolab, Nogent sur Marne, France). A 10 mg l−1 Multielement Standard 2 Solution (Perkin-Elmer, PE Pure) was used for preparing the working standards of aluminium, arsenic, cadmium, cobalt, chromium, mercury, lithium, manganese, nickel, lead and selenium; and to prepare tuning solution containing indium, lithium, barium and uranium. 1000 mg/L standard stock solutions of scandium, indium and bismuth were purchased from Fisher (Fisher Scientific, Elancourt, France) to prepare the internal standard solution. All standards were used without further purification. Working standards were prepared in 6% (v/v) HNO3. 3 ml HNO3 65% (v/v), used without further purification, was utilized for the digestion procedure. Certified reference materials (CRMs) BCRs 063R (Natural milk powder), 150 and 151 (Skim milk powder), 184 (Bovine muscle), 185R (Bovine liver), 186 (Pig kidney), 191 (brown bread), 278R (Mussel tissue), 279 (sea lettuce), 422 (cod muscle), Simulated diet E and DORM-2 (Dogfish) plus IAEA 140/TM (focus) and 350 (tuna fish flesh) were purchased from Promochem (Molsheim, France). All CRMs were used as provided without further grinding.

Results and discussion

Analytical Quality Assurance (AQA)

All the results were validated with Internal Quality Controls (IQC) and External Quality Controls (EQC), in compliance with the ISO/IEC 17025 norm (ISO/IEC 17025 1999). Once the method was optimized according to certain criteria described before (Noël et al. Citation2003a), all test batches were evaluated using an internal quality approach previously explained (Noël et al. Citation2003b) and validated if they satisfied the defined IQC (Lemahieu N, Guérin T, report for DGAl). For this reason, the results for iron and tin did not satisfy these controls with the analytical methodological compromises chosen and were therefore eliminated of the present paper. As Baxter et al. (Citation1997) suggest, target criteria should be established in advance, although this is not always possible when very complex mixtures of samples are measured for the first time, as in our case.

Appropriate IQC parameters were further considered for single analyte determinations (Thompson and Wood Citation1995) and many of them have been included for evaluating and to develop our IQC for multi-element analyses by ICP-MS on different types of unknown matrix. The criteria retained and used in this large survey are presented in .

Table II. Internal quality control criteria.

The limits of detection (LOD) were used to identify any cross contamination or memory effects from sample preparation. Thus, each test run included one reagent blank control. The LOD of selected essential and non-essential trace elements sought, given that the limit of quantification (LOQ) corresponds to twice the LOD and the percentage of Non-Quantified values (<LOQ) are presented in .

Table III. Limits of quantification for elements.

The instrumental drift is verified in each six samples (plus one at the end) by the control of the sensitivity of one calibration standard. Although four internal standards (beryllium, scandium, indium and bismuth) were used to correct matrix effects (non-spectral interferences) during the ICP-MS run, some variations in response were expected and this was assessed by measuring a second aliquot of a mid-range standard each five samples and at the end of each ICP-MS analytical sequence. Each test run included spiked test samples and reference materials for a comparison of measured and certified reference material (CRM) concentrations of the elements of interest. Furthermore, all samples were analysed in duplicate, which were digested and measured in separates batches, to eliminate any batch specific error.

During the study, the analytical laboratory took part of proficiency schemes (CSL-FAPAS), when available, to estimate external quality controls (EQC).

Intake calculations

The mean contents per food group are expressed in mg/kg of fresh matter for elements (). For each of the 300 foods listed, five sub-samples were purchased per national or regional food and subsequently combined into one composite sample corresponding to a total of 998 individual composite samples obtained from 4990 purchased food products as detailed in Leblanc et al. Citation2005. The average levels of concentration used in the calculation of intakes correspond to the weighted average of the individual food composite items for the two seasons.

Table IV. Mean concentrations of minerals, essential trace elements and non-essential trace elements included in the French Total Diet Study.

Due to the considerable place of the Table result, the mean values obtained per individual food item (300) and used at a disaggregated level in our intakes calculus (e.g., apricot, fresh) are not presented here, just the mean aggregated value for the 43 food group (e.g., fruit) are presented in . The mean element levels were used in the intake calculations, since in general at international level, this methodology is recognized as giving an appropriate estimate of the long–term exposure and is considered to be relevant when exposures were compared with TDI or PTWI (FAO/WHO Citation1997). The intakes are expressed in µg or mg per person or per kilogram of body weight per day or per week.

The percentage of non-quantified values for the various element (between 0.2 and 91%, ) were treated according the following statistical approach: for samples in which elements levels were not detected (nd), values were assumed to be half the limit of detection (nd = (1/2)LOD) and for samples in which elements levels were seen to be below the Limit of Quantification, values were assumed to be half the limit of Quantification (<LOQ = (1/2)LOQ).

Intakes estimates shown in and expressed for averages, median and/or high quartiles (95th or 97.5th percentile) were obtained by multiplying individual food consumption data for the French adult and children population by the average concentrations of elements present in the “as-consumed” individual food list (300). The coverage of the total diet, consumed by the populations studied, took into account in intakes estimate of elements is in mean around 81% and the coverage of each food group in the diet is between 10.5 and 100% depending of the food group (see column 2 in ). The total intakes presented for each element in were calculated for each individual, based on actual food consumption and actual declared body weight. Note that the total intake for the high quartiles exposure reported in the Tables does not correspond to the sum of the high quartiles of exposure of each food group taken into account in the diet model, since the high consumers characterized for each food item are not the same in different food items.

Table V. Estimation of daily intakes of minerals and essential trace elements for children from 3 to 14 years (n = 1018) from French Total Diet Study.

Table VI. Estimation of daily intakes of minerals and essential trace elements for adults, 15 years and more (n = 1474) from French Total Diet Study.

Table VII. Estimation of daily intakes of non-essential trace elements for children from 3 to 14 years (n = 1018) from French Total Diet Study.

Table VIII. Estimation of daily intakes of non-essential trace elements for adults over 15 years and more (n = 1474) from French Total Diet Study.

Table IX. Intakes distribution and quantification of the probabilities for general population to be under or over toxicological or nutritional references values established for minerals and essential trace elements.

Table X. Intakes distribution and quantification of the probabilities for general population to be under or over toxicological reference value established for non-essential trace elements.

The average contribution of each product group to the total exposure is expressed here as a percentage. Moreover, when necessary the main vector(s) contributing to this exposure is (are) mentioned in the text, in accordance with the methodological recommendations for fixing food standards in the process of adoption in the Codex Alimentarius (CCFAC Citation2004). The results obtained are expressed as equivalences or as contribution to the nutritional or toxicological reference value.

In this work, the risk was also quantified by estimating the probability for exposure to exceed a fixed safe reference dose like PTWI, PTMI, UL or to be under a fixed nutritional reference dose like LTI, called “d”. This probability was evaluated by the empirical estimator, often called Plug-In.

Denoting by Ki i = 1, … , N, the calculated individual exposure, the estimator is simply

where # (Ki  ≥ d) is the number of exposures that exceed d (Tressou et al. Citation2004). Considering the percentage of the coverage diet took into account for each food group and the statistics treatment made for non-quantified values related to elements, the values obtained with respect to nutritional or toxicological risk quantification should be interpreted with caution.

Minerals and essential trace elements

Calcium. In 998 individual food composite samples, 30 display calcium levels < LOD of 7.65 mg/kg of fresh weight and 13 were between LOD and LOQ. Calcium is found at average levels between 1.0 and 6.7 g/kg in the following food groups, in decreasing order of content: Cheeses, breakfast cereals, nuts and oilseeds, pizzas, quiches and salted cakes, milk, chocolate, substitute meals, ultra-fresh dairy products and salads; the majority of other food groups contain less than 0.8 g/kg ().

and show that the estimated average daily intake of the French population is 721 mg for adults aged 15 years or more and 729 mg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 300 mg (equivalent LTI of 75%) and 1312 mg (equivalent UL of 66%). For children, the exposures are 274 mg (equivalent LTI of 90–140%) and 1324 mg (equivalent UL of 66%) (see ). Compared to existing French data, the results are lower on average and for the median by about 100 mg (CREDOC-AFSSA-DGAL) or higher by about 100 mg (Noël et al. 2003, 642 mg). The milk/dairy products group is the vector contributing most (49–54%) to the exposure of the populations. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 8.6% of adults aged 15 years or more and 4.3% of children aged 3–14. The proportion exceeding the upper level is estimated to be 0.01% for both groups.

Chromium

In 998 individual food composite samples, 192 display chromium levels < LOD of 15 µg/kg of fresh weight and 188 were between LOD and LOQ. Chromium is found in the majority of foods at average levels between 0.10 and 0.34 mg/kg in the following food groups, in decreasing order of content: Chocolate, cereals and cereal products, cheeses, offal and delicatessen, sugars and confectionery, salads and condiments and sauces; the majority of other groups contain less than 90 µg/kg (see ).

and show that the estimated average daily intake of the French population is 77 µg for adults aged 15 years or more and 68 µg for children aged 3–14. The daily 97.5th percentile exposure for adults is 126 µg (equivalent UL of 13%) and 124 µg for children (equivalent UL of 12%) (). Compared to existing French data, the results are on average of the same order of magnitude (Biego et al. Citation1998) or lower by a factor of about 2 (Noël et al. Citation2003a). The food groups including bread, rusk, non-alcoholic beverages and prepared dishes are the vectors contributing most (7–17%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the upper level is estimated to be 0% of adults aged 15 years or more and of children aged 3–14.

Cobalt

In 998 individual food composite samples, 568 display cobalt levels < LOD of 1 µg/kg of fresh weight and 57 were between LOD and LOQ. Cobalt is found in the majority of foods at average levels between 25 and 50 µg/kg in the following food groups, in decreasing order of content: Chocolate, shellfish, nuts and oilseed, pasta and salads; the majority of other food groups contain less than 20 µg/kg ().

and show that the estimated average daily intake of the French population is 7.5 µg for adults aged 15 years or more and 7.3 µg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 3.5 µg and 14 µg. For children, these exposures are 3.0 µg and of 15 µg (). To be able to compare the exposure to cobalt with reference nutritional or toxicological values, it is necessary to know the proportion of foods containing cobalt in the form of vitamin B12, information that we do not have. Compared to existing French data, the results are on average lower by a factor of 4 (Biego et al. Citation1998) or higher by a factor of 2 (Noël et al. Citation2003b). The food groups including cereals and cereal products, fruits, sugars and confectionery, salads and prepared dishes are the vectors contributing most (7–12%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure.

Copper

In 998 individual food composite samples, 22 display an average copper content < LOD of 19 µg/kg of fresh weight and 28 were between LOD and LOQ. The main food groups containing copper are shellfish, nuts and oilseed and offals at respective levels of 7.0, 7.15 and 13.5 mg/kg; other food groups contain less than 3 mg/kg ().

and show that the estimated average daily intake of the French population is 0.98 mg for adults aged 15 years or more and 0.81 mg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 0.5 mg (equivalent LTI of 83%) and 1.7 mg (equivalent UL of 17%). For children, these exposures are 0.4 mg (equivalent LTI of 130–200%) and 1.6 mg (equivalent UL of 16%) (). Compared to existing French data, the results are on average equivalent (Noël et al. Citation2003b, 0.93 mg) or lower by a factor of 1.5–2 (Biego et al. Citation1998, AFSSA Citation2001). The food groups including bread, rusks, biscuits, vegetables, starchy vegetables and fruits are the vectors contributing most (5–16%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 7.3% of adults aged 15 years or more and 0.01% of children aged 3–14. The proportion exceeding the upper level is estimated to be 0% for both groups.

Lithium

In 998 individual food composite samples, 345 display average lithium content < LOD of 3 µg/kg of fresh weight and 115 were between LOD and LOQ. The main food groups containing lithium are shellfish (123 µg/kg) and drinking water (100 µg/l); other food groups contain less than 40 µg/kg ().

and show that the estimated average daily intake of the French population is 28.5 µg for adults aged 15 years or more and 14.5 µg for children aged 3–14. The daily 97.5th percentile exposure is 144 µg for adults and 38 µg for children (). Compared to existing French data, the results are on average higher by a factor of 3 (Noël et al. Citation2003b). The food groups including drinking waters and soups are the vectors contributing most (respectively 25–41% and 14–15%) to the exposure of the populations; other vectors contribute less than 10% of the total food exposure.

Magnesium

In 998 individual food composite samples, two display magnesium levels < LOD of 0.25 mg/kg of fresh weight and none between LOD and LOQ. The main food groups containing magnesium are shellfish, cereals, and nuts and oilseed groups at respective average levels of 510, 896 and 1150 mg/kg; other food groups contain less than 500 mg/kg ().

and show that the estimated average daily intake of the French population is 224 mg for adults aged 15 years or more and 197 mg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures of adults are 130 mg (equivalent LTI of 87%) and 359 mg (equivalent UL of 64%). For children, these exposures are 105 mg (equivalent LTI of 95–310%) and 334 mg (equivalent UL of 60%) (). Compared to existing French data, the results are on average of the same order of magnitude (CREDOC-AFSSA-DGAL, Noël et al. Citation2003a). The food groups including bread, rusks, milk, vegetables, meat, fruits and coffee are the vectors contributing most (6–13%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 7.2% of adults aged 15 years or more and 1.3% of children aged 3–14. The proportion exceeding the upper level is estimated to be 0% for both groups.

Manganese

In 998 individual food composite samples, 23 display manganese levels < LOD of 8 µg/kg of fresh weight and 13 were between LOD and LOQ. The main food groups containing manganese are breakfast cereals, nuts and oilseed, and other cereals at respective average levels of 10.8, 10.9 and 19.6 mg/kg; other food groups contain less than 5 mg/kg ().

and show that the estimated average daily intake of the French population is 2.3 mg for adults aged 15 years or more and 1.9 mg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 1.0 mg (equivalent LTI of 133%) and 4.8 mg (equivalent UL of 48%). For children, these exposures are 0.6 mg (equivalent LTI of 80%) and 4.2 mg (equivalent UL of 42%) (). Compared to existing French data, the results are on average of the same order of magnitude (Biego et al. Citation1998; Noël et al. Citation2003b). The food groups including bread, rusk and fruits are the vectors contributing most (respectively 20–30% and 15–17%) to the exposure of the populations; other vectors contribute less than 8% of the total food exposure. The proportion of individuals whose theoretical intake is less than the MNI is estimated to be 0.4% of adults aged 15 years or more and 0.8% of children aged 3–14. The proportion exceeding the upper level is estimated to be 0% in adults and 0.01% in children.

Molybdenum

In 998 individual food composite samples, 164 display molybdenum levels < LOD of 3 µg/kg of fresh weight and 87 were between LOD and LOQ. The main food groups containing molybdenum are nuts and oilseed, offals, pulses and other cereals at respective average levels of 1.3, 0.8 and 0.7 mg/kg; other food groups contain less than 0.5 mg/kg ().

and show that the estimated average daily intake of the French population is 138 µg for adults aged 15 years or more and 114 µg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 65 µg (equivalent LTI of 370%) and 270 µg (equivalent UL of 45%). For children, these exposures are 48 µg (equivalent LTI of 274%) and 216 µg (equivalent UL of 54–108%) (). Compared to existing French data, the results are on average lower by a factor of 2 (Biego et al. Citation1998) or of the same order of magnitude (Noël et al. Citation2003b). The food groups including bread, rusk, starchy vegetables, vegetables and pulses are the vectors contributing most (respectively 13–22%, 16–18%, 9–13% and 7%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 0% of adults aged 15 years or more and children aged 3–14. The proportion exceeding the upper level is estimated to be 0% of adults and 0.5% of children.

Nickel

In 998 individual food composite samples, 424 display nickel levels < LOD of 32 µg/kg of fresh weight and 261 were between LOD and LOQ. The main food groups containing nickel are nuts and oilseed, chocolate and breakfast cereals at respective average levels of 1.15, 0.63 and 0.55 mg/kg; other food groups contain less than 0.5 mg/kg ().

and show that the estimated average daily intake of the French population is 94 µg for adults aged 15 years or more and 92 µg for children aged 3–14. The daily 97.5th percentile exposure for adults is 166 µg (equivalent UL of 28%) and for children 174 µg (equivalent UL of 29%) (). Compared to existing French data, the results are on average lower by a factor of 2.5 (Biego et al. Citation1998) or of the same order of magnitude (Noël et al. Citation2003b). The food groups including vegetables, drinking water, breakfast cereals, biscuits, bread, rusk, milk, cheeses, non-alcoholic beverages and prepared dishes are the vectors contributing most (respectively 5% and 10%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the upper level is estimated to be 0% of adults aged 15 years or more and 0% of children aged 3–14.

Selenium

In 998 individual food composite samples, 671 display selenium levels < LOD of 22 µg/kg of fresh weight and 141 were between LOD and LOQ. The main food groups containing selenium are poultry and game, fish, delicatessen and meat at average levels between 95 and 170 µg/kg; other food groups contain less than 50 µg/kg ().

and show that the estimated average daily intake of the French population is 42 µg for adults aged 15 years or more and 31 µg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 22 µg (equivalent LTI of 110%) and 70 µg (equivalent UL of 23%). For children, these exposures are 15 µg (equivalent LTI of 150–375%) and 55 µg (equivalent UL of 22–61%) (). Compared to existing French data, the results are on average lower by a factor of 1.5 (Noël et al. Citation2003b) or of the same order of magnitude (AFSSA Citation2001). The food groups including vegetables, drinking water, breakfast cereals, biscuits, bread, rusk, milk, cheeses, non-alcoholic beverages and prepared dishes are the vectors contributing most (respectively 5 and 10%) to the exposure of the population; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 1.5% of adults aged 15 years or more and 0.2% of children aged 3–14. The proportion exceeding the upper level is estimated to be 0% for both groups.

Sodium

In 998 individual food composite samples, 40 display sodium levels < LOD of 2.92 mg/kg of fresh weight and 13 were between LOD and LOQ. The main food groups containing sodium are cereal products (1.7–6.8 g/kg), meat products (1.2–6.9 g/kg), fishing products (2.1–5.4 g/kg), cheeses (5.6 g/kg) and composed products such as pizzas, salt cakes, quiches, sandwiches, prepared dishes, soups and salads (1.5–6.0 g/kg) (see ).

and show that the estimated average daily intake of the French population is 2.3 g for adults aged 15 years or more and 1.8 g for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 1.1 g (equivalent LTI of 193%) and 4.3 g (equivalent UL of 90–122%). For children, these exposures are 0.8 g (equivalent LTI of 140%) and 3.8 g (equivalent UL of 81–108%) (). The results compared to existing French data are on average and at the 95th percentile lower by about 20–25% (CREDOC-AFSSA-DGAL; Noël et al. Citation2003b). The food groups including bread, rusk, and prepared dishes are the vectors contributing most (respectively 20–40% and 10%) to the exposure of the population; other vectors contribute less than 6% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 0.07% of adults aged 15 years or more and 0.3% of children aged 3–14. The proportion of individuals exceeding the upper level is estimated to be between 8% and 13% of adults aged 15 years or more and between 3.5% and 5% of children aged 3–14. This frequency is higher if food table salt is included.

Zinc

In 998 individual food composite samples, 11 display zinc content < LOD of 81 µg/kg of fresh weight and 45 were between LOD and LOQ. The main food groups containing zinc are shellfish (66 mg/kg), meat (37 mg/kg), cheese and offal (31 mg/kg), other cereals and nuts and oilseed (19–26 mg/kg); other food groups contain less than 17 mg/kg ().

and show that the estimated average daily intake of the French population is 8.7 mg for adults aged 15 years or more and 7.6 mg for children aged 3–14. The 2.5th and 97.5th percentile daily exposures for adults are 4.6 mg (equivalent LTI of 92–115%) and 15 mg (equivalent UL of 25–100%). For children, these exposures are 3.6 mg (equivalent LTI of 100–225%) and 14 mg (equivalent UL of 47–140%) (). Compared to existing French data, the results are on average lower by a factor of 1.5 (Biego et al. Citation1998) or of the same order of magnitude (Noël et al. Citation2003b). The food groups including meat, milk, cheeses, bread and rusk are the vectors contributing most (respectively 22–23%, 8–17% and 6–11%) to the exposure of the population; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake is less than the lowest threshold intake is estimated to be 2.6% of adults aged 15 years or more and 0.9% of children aged 3–14. The proportion of individuals exceeding the upper level is estimated to be between 0 and 2.0% of adults and between 0 and 2.1% of children.

Non-essential trace elements

Aluminium

In 998 individual food composite samples, 151 display aluminium levels < LOD of 109 µg/kg of fresh weight and 216 were between LOD and LOQ. Aluminium is found at a level exceeding 3 mg/kg only in the following food groups: bread, rusk, breakfast cereals, biscuits, vegetables, nuts and oilseed, ice-cream, chocolate, salads and shellfish ().

and show that the estimated average daily intake of the French population is 1.6 mg for adults aged 15 years or more and 1.3 mg for children aged 3–14. The 97.5th percentile exposure for adults is 0.4 mg/kg b.w./week (equivalent PTWI of 6% – JECFA-SCF) and for children 0.7 mg/kg b.w./week (equivalent PTWI of 10% – JECFA-SCF) (). Compared to existing French data, the results are lower by a factor of 1.3 (Noël et al. Citation2003b), 1.75 (InVS-AFSSA-AFSSAPS, Citation2003) or 2.5 (Biego et al. Citation1998). The food groups including vegetables, bread and rusk are the vectors contributing the most (15–20%) to the exposure of the population; other vectors contribute less than 10% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the tolerable weekly dose of 7 mg/kg b.w./week is estimated to be 0% of adults aged 15 years or more and 0% of children aged 3–14.

Antimony

In 998 individual food composite samples, 881 display antimony levels < LOD of 0.6 µg/kg of fresh weight and 26 were between LOD and LOQ. Antimony is found at an average level between 1.7 and 2.4 µg/kg in the food groups including meats, delicatessen, shellfish, vegetables and starchy vegetables, fruits and nuts and oilseed; the majority of other groups contain less than 0.6 µg/kg ().

and show that the estimated average daily intake of the French population is 1.0 µg for adults aged 15 years or more and 0.8 µg for children aged 3–14. The daily 97.5th percentile exposure is 2 µg for adults and 1.6 µg for children (). The results compared to existing French data are lower by a factor of 3 (Noël et al. Citation2003b). The food groups including meats, vegetables and starchy vegetables, fruits and drinking water are the vectors contributing most (10–15%) to the exposure of the populations; other vectors contribute less than 10% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the TDI established for antimony is estimated to be 0% of adults aged 15 years or more and 0% of children aged 3–14.

Arsenic

In 998 individual food composite samples, 557 display arsenic levels < LOD of 5 µg/kg of fresh weight and 95 were between LOD and LOQ. Arsenic is found at an average level of 2 mg/kg in the fish and shellfish food group; the majority of other groups contain less than 50 µg/kg. The scientific literature reports that 0.4–5.3% of arsenic present in products from the sea are in the form of inorganic arsenic (Schoof et al. Citation1999) ().

and show that the estimated average daily intake of the French population is 62 µg for adults aged 15 years or more and 43 µg for children aged 3–14. The 97.5th percentile exposure is 31 µg/kg b.w./week for adults and 42 µg/kg b.w./week for children. To permit a comparison of the exposure versus PTWI based on toxicity of inorganic arsenic, it is estimated that around 10% of the arsenic found in foods are inorganic forms, given an equivalent PTWI of inorganic arsenic of 3.1% for adults and 4.2% for children (). Compared to existing French data, the results are lower by a factor of 1.8 to 2.4 ((Noël et al. Citation2003b; Leblanc et al. Citation2000; SCOOP 3.2.11 Citation2004). The food group including fish, shellfish and fruits are the vectors contributing most (respectively 49–50%, 8–13% and 15–17%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the PTWI established for inorganic arsenic is estimated to be 0% of adults aged 15 years or more and 0% of children aged 3–14.

Cadmium

In 998 individual food composite samples, 737 display cadmium levels < LOD of 0.7 µg/kg of fresh weight and 80 were between LOD and LOQ. Cadmium is found at an average level between 0.05 and 0.1 mg/kg in offals and shellfish; the majority of other food groups contain less than 0.02 mg/kg ().

and show that the estimated average daily intake of the French population is 2.7 µg for adults aged 15 years or more and 2.0 µg for children aged 3–14. The 97.5th percentile exposure for adults is 0.7 µg/kg b.w./week (equivalent PTWI of 10%) and for children 1.2 µg/kg b.w./week (equivalent PTWI of 17%) children (). Compared to existing French data, the results are on average comparable (Noël et al. Citation2003b) or lower by a factor 4 to 10 (SCOOP 3.2.11 Citation2004; Leblanc et al. Citation2000; Decloitre Citation1998; Biego et al. Citation1998). The food group including vegetables and starchy vegetables are the vectors contributing most (respectively 21–23% and 21–27%) to the exposure of the populations; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the PTWI established for cadmium is estimated to be 0% of adults aged 15 years or more and 0% of children aged 3–14.

Lead

In 998 individual food composite samples, 267 display lead levels < LOD of 5 µg/kg of fresh weight and 320 were between LOD and LOQ. Lead is found at an average level between 0.05 and 0.1 mg/kg in offal and shellfish; the majority of other food groups contain less than 0.04 mg/kg ().

and show that the estimated average daily intake of the French population is 18 µg for adults aged 15 years or more and 13 µg for children aged 3–14. The 97.5th percentile exposure for adults is 3.6 µg/kg b.w./week (equivalent PTWI of 14%) and for children 6.4 µg/kg b.w./week (equivalent PTWI of 26%) (). Compared to existing French data, the results are lower by a factor of 2 to 4 (Noël et al. Citation2003b; SCOOP 3.2.11 Citation2004; Leblanc et al. Citation2000; Decloitre Citation1998). The following food groups are the vectors contributing most (5–11%) to the exposure of the populations: bread, rusk, soups, vegetables, fruits, drinking water, non-alcoholic beverages, alcoholic beverages and sugars and confectionery; other vectors contribute less than 5% of the total food exposure. The proportion of individuals whose theoretical intake exceeds the PTWI established for lead is estimated to be 0% of adults aged 15 years or more and 0% for children aged 3–14.

Mercury

In 998 individual food composite samples, 674 display mercury levels < LOD of 6 µg/kg of fresh weight and 129 were between LOD and LOQ. Mercury is found at an average level of 62 µg/kg in the fish food group and 42 µg/kg in chocolate group; other food groups contain less than 17 µg/kg ().

and show that the estimated average daily intake of the French population is 9.7 µg for adults aged 15 years or more and 7.9 µg for children aged 3–14. The 97.5th percentile exposure for adults is 1.8 µg/kg b.w./week and for children is 4.1 µg/kg b.w./week for mercury an. To permit a comparison of the exposure versus PTWI based on toxicity of methylmercury, it is estimated that around 100% of the mercury found in fish and shellfish are methylmercury forms, given for adults and children a respective exposure estimated of 0.7 µg/kg b.w./week and 1.0 µg/kg b.w./week (equivalent PTWI of 44% and 63% for people consuming fishing products) (). Compared to existing French data, the results are on average comparable (Noël et al. Citation2003b; SCOOP 3.2.11 Citation2004) or lightly lower by a factor of 1.5 (Decloitre Citation1998). The proportion of individuals whose theoretical intake exceeds the PTWI established for methylmercury is estimated to be 0% of adults aged 15 years or more and 1% of children aged 3–14.

Conclusion

The implementation of the TDS, which is one of the methods used in many countries to evaluate food safety risks, offers for risk managers involved in food safety several advantages. First it has the advantage of providing more realistic exposure data since foods are analysed “as-consumed” by the eater. Secondly, it provides a good tool for identifying the population or age groups most exposed (children, old people, etc.), identifying the main food or/food groups contributing to the exposure when food sampling is based on an individual approach and monitoring food and identifying useful trends for guiding and orienting food safety programmes. Finally, it facilitates data which is usable for international comparisons, since the study applies a methodological approach similar to that already used in other countries (USA, Australia, New Zealand, China, Canada, England, Ireland, Netherlands, Spain, Czech Republic).

This study reveals that, for the French population, the observed levels of the mineral contaminants studied in representative diets are overall satisfactory with respect to current regulations. Concerning minerals, trace elements and non-essential trace elements, the study shows that for the populations studied, the probability of exposure to nutritional and/or health risks from such elements are overall small.

Acknowledgements

The authors offer special thanks to the Ministry of Agriculture for financing this study and also to H. Ferry-Wilczek, Deputy Director of the DGAL at the Ministry of Agriculture, for his support during the initiation of this project. His interest in the evaluation of food risks enabled the necessary resources and energy to be mobilized to allow this study to progress effectively.

References

References

  • AFSSA 2002 Rapport GT sel, Evaluation et recommandations Août 2002, www/afssa.fr
  • AFSSA-CNERNA-CNRS 2001 Apports nutritionnels conseillés pour la population française 3ème édition, Coord. Martin A. Eds. tech & doc
  • Baxter MJ Crews HM Robb P Strutt PR 1997 Quality control in the multi-element analysis of foods using ICP-MS. Plasma Source Mass Spectrometry In: Holland G, Tanner SD, editors Developments and applications Cambridge The Royal Society of Chemistry pp 95–108
  • Biego , GH , Joyeux , M , Hartemann , P and Debry , G . 1998 . Daily intake of essential minerals and metallic micropollutants from foods in France . Science of the Total Environment , 217 : 227 – 232 .
  • Codex Committee on Foods Additives and Contaminants (CCFAC) 2004 Proposed draft CCFAC policy for exposure assessment of contaminants and toxins in foods or food groups for adoption step 8 to the CAC through CCGP (36éme session, CDR3, annex II)
  • Combris P Delobel D Boizot C 2000 La consommation alimentaire en 1997: Distribution des quantités consommées à domicile Rapport du laboratoire de recherches sur la consommation, INRA
  • Conseil supérieur d’hygiène publique de France (CSHPH) 1996 Martin, A., Les limites de sécurité dans les consommations alimentaires des vitamines et minéraux Tech et doc, Lavoisier Paris pp 35–36
  • CRÉDOC-AFSSA-DGAL 2000 Enquête INCA individuelle et nationale sur les consommations alimentaires Coord Volatier JL. Eds Tec & Doc Paris
  • Decloitre , F . 1998 . La part des différents aliments dans l’exposition au plomb, au cadmium et au mercure en France . Cahiers de Nutrition et de Diététique , 33 : 3
  • FAO/WHO 1985 Guidelines for the study of dietary intakes of chemical contaminants Geneva WHO Offset publication no 87
  • FAO/WHO 1997 Food consumption and exposure assessment of chemicals Report of a FAO-WHO Consultation, Geneva. 10–14 February 1997
  • Goldberg , GR , Black , AE , Jebb , SA , Cole , TJ , Murgatroyd , PR , Coward , WA and Prentice , AM . 1991 . Critical evaluation of energy intake data using fundamental principles of energy physiology: 1 . Derivation of cut-off limits to identify under-recording. European Journal of Clinical Nutrition , 45 ( 12 ) : 569 – 581 .
  • InVS-AFSSA-AFSSAPS 2003 Evaluation des risques sanitaires lies à l’exposition de la population française à l’aluminium Novembre 2003
  • JECFA 2003 Safety evaluation of certain food additives and contaminants Prepared by the 61st meeting of the joint FAO/WHO Expert Committee on Food Additives Rome WHO June
  • International Organization for Standardization/International Electrotechnical Commission 17025 1999 General Requirements for the competence of testing and calibration laboratories (1st edition, 1999), www.iso.org
  • JECFA 1982 Safety evaluation of certain food additives and contaminants Prepared by the 36th meeting of the joint FAO/WHO Expert Committee on Food Additives Technical reports series no 683 WHO
  • JECFA 1989 Safety evaluation of certain food additives and contaminants Prepared by the 33rd meeting of the joint FAO/WHO Expert Committee on Food Additives Technical reports series No 776
  • JECFA 1998 Safety evaluation of certain food additives and contaminants Prepared by the 49th meeting of the joint FAO/WHO Expert Committee on Food Additives WHO food additives series 40 – (WHO Geneva) pp 359–369
  • JECFA 1999 Safety evaluation of certain food additives and contaminants Prepared by the 53rd meeting of the joint FAO/WHO Expert Committee on Food Additives WHO food additives series 44 – (WHO Geneva) pp 393–483
  • Leblanc J-Ch Tard Alexandra Volatier J-L Verger Ph 2005 Dietary exposure estimates of main food mycotoxins from the 1st French Total Diet Study Food Additives and Contaminants (in press)
  • Leblanc , J-Ch , Malmauret , T , Guérin , F , Bordet , B and Boursier , Ph . 2000 . Verger, Estimation of the dietary intakes of pesticides residues, lead, cadmium, arsenic and radionucleides in France . Food Additives and Contaminants , 17 : 925 – 932 .
  • Noël , L , Guérin , T , Huet , H , Kolf-Clauw , M and Frémy , JM . 2003a . Optimized simultaneous determination of several elements in human intestinal caco-2 TC7 cells by ICP-MS after closed vessel microwave digestion . Journal of AOAC International , 86 : 1225 – 1231 .
  • Noël , L , Leblanc , JC and Guérin , T . 2003b . Determination of several elements in duplicate meals from catering establishment using closed vessel microwave digestion with inductively coupled plasma mass spectrometry detection: Estimation of daily dietary intake . Food Additives and Contaminants , 20 ( 1 ) : 44 – 56 .
  • Risk evaluation of essential trace elements, essential versus toxic levels of intake, report of a Nordic project group 1995 Alexander et al. Nordic Council of Ministers, Copenhagen
  • SCF 1993 Reports of the scientific committee for food. 31st series. Food science and techniques, nutrient and energy intakes for the European community Commission of the European communities, 1993
  • SCF 2000a Opinion of the scientific committee on food on the tolerable upper intake level of molybdenum European Commission
  • SCF 2000b Opinion of the scientific committee on food on the tolerable upper intake level of manganese European Commission
  • SCF 2000c Opinion of the scientific committee on food on the tolerable upper intake level of selenium European Commission
  • SCF 2001 Opinion of the scientific committee on food on the tolerable upper intake level of magnesium European Commission
  • SCF 2003a Opinion of the scientific committee on food on the tolerable upper intake level of calcium European Commission
  • SCF 2003b Opinion of the scientific committee on food on the tolerable upper intake level of zinc European Commission
  • SCF 2003c Opinion of the scientific committee on food on the tolerable upper intake level of trivalent chromium European Commission
  • Schoof , RA , Yost , LJ , Eickhoff , J , Crecelius , EA , Cragin , DW , Meacher , DM and Menzel , DB . 1999 . A market basket survey of inorganic arsenic food . Food and Chemical Toxicology , 37 : 839 – 846 .
  • SCOOP reports on Tasks 3.2.11 2004 Assessment of dietary intake of arsenic, cadmium, lead and mercury by the population of EU members’ states May, 2004
  • Thompson , M and Wood , R . 1995 . Harmonized guidelines for internal quality control in analytical chemistry laboratories . Pure and Applied Chemistry , 67 ( 4 ) : 649 – 666 .
  • Tressou , J , Bertail , P , Crépet , A , Feinberg , M and Leblanc , J-Ch . 2004 . Probabilistic exposure assessment to food chemicals based on extreme value theory: Application to heavy metals from fish and sea products . Food Chemical and Toxicology , 42 : 1349 – 1358 .
  • WHO 2003 Revision of recommendations for drinking water quality
  • WHO 2004 IWTDS, International Total Diet workshop, 1st IWTDS, Kansas City (1999) 2nd IWTDS, Brisbane (2002) and 3rd IWTDS Paris (2004)

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.