- Email: [email protected]
Dietary exposure to inorganic arsenic of the Hong Kong population: Results of the first Hong Kong Total Diet Study
Food and Chemical Toxicology 51 (2013) 379–385
Contents lists available at SciVerse ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
Dietary exposure to inorganic arsenic of the Hong Kong population: Results of the first Hong Kong Total Diet Study Waiky W.K. Wong a, Stephen W.C. Chung b, Benny T.P. Chan b, Y.Y. Ho a, Ying Xiao a,⇑ a
Centre for Food Safety (CFS), Food and Environmental Hygiene Department (FEHD), 43/F, Queensway Government Offices, 66 Queensway, Hong Kong Food Research Laboratory (FRL), Centre for Food Safety (CFS), Food and Environmental Hygiene Department (FEHD), 4/F, Public Health Laboratory Centre, 382 Nam Cheong Street, Shek Kip Mei, Kowloon, Hong Kong
b
a r t i c l e
i n f o
Article history: Received 18 April 2012 Accepted 12 October 2012 Available online 24 October 2012 Keywords: Inorganic arsenic Total diet Dietary exposure
a b s t r a c t Inorganic arsenic, a human carcinogen, can be found in the environment and food. In the first Hong Kong Total Diet Study, the dietary exposure of the Hong Kong people, including various age-gender subgroups, to inorganic arsenic was estimated for assessing the associated health risk. Food samples, which represented the Hong Kong people’s diet, were collected and prepared ‘‘as consumed’’ for analysis. Concentrations of inorganic arsenic, as sum of arsenite (As(III)) and arsenate (As(V)) were determined in 600 composite samples by using inductively coupled plasma mass spectrometry. The dietary exposures were estimated by combining the analytical results with the local food consumption data of the adult population. The mean and 95th percentile of inorganic arsenic exposures of the Hong Kong people were 0.22 and 0.38 lg/kg body weight (bw)/day, respectively. Among the 12 age-gender subgroups, the respective exposures ranged from 0.19 to 0.26 lg/kg bw/day and from 0.33 to 0.46 lg/kg bw/day. The main food category that contributed inorganic arsenic was ‘‘cereals and their products’’ (53.5% of the total exposure), particularly rice. Having considered the carcinogenic risk of inorganic arsenic to humans, it is suggested that efforts should be made to reduce the inorganic arsenic exposure of the Hong Kong population. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Inorganic arsenic is the more toxic form of arsenic, a metalloid, which is found in the environment from both natural occurrence and human activities (JECFA, 2011). Food is recognised as a major source of inorganic arsenic exposure (EFSA, 2009; JECFA, 2011; WHO, 2003). The inorganic arsenic levels found in foods and beverages usually do not exceed 100 lg/kg with mean value of less than 30 lg/kg. However, some food commodities, such as seaweed, rice, some fish and seafood commodities, may contain higher inorganic arsenic levels. The inorganic arsenic as a proportion of the total arsenic also varies among food commodities (Codex, 2011a; JECFA, 2011). Long term exposure to inorganic arsenic may cause bladder, lung and skin cancers, skin lesions, cardiovascular disease, neurotoxicity, and diabetes (EFSA, 2009; IARC, 2009; JECFA, 2011). In 2010, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) determined the inorganic arsenic Benchmark Dose Lower Confidence Limit for a 0.5% increased incidence of lung cancer in human (BMDL0.5) as 3.0 lg/kg body weight (bw)/day (in the region of 2–7 lg/kg bw/day). This BMDL0.5 contains the following uncertainties, namely, the assumptions on total exposure extrapolated ⇑ Corresponding author. Tel.: +852 28675526; fax: +852 28933547. E-mail address: [email protected] (Y. Xiao). 0278-6915/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fct.2012.10.010
from the drinking water; and the extrapolation of the BMDL0.5 to other populations due to the influence of nutritional status such as low protein intake, and other lifestyle factors on the effects observed in the studied population (JECFA, 2011). BMDL0.5 cannot be regarded as a safety reference value. A dietary exposure below the BMDL0.5 does not mean that there is no health risk (Codex, 2011b). Therefore, in order to assess the health risk of dietary exposure the margin of exposure (MOE) approach is used. The MOE for inorganic arsenic is the ratio of BMDL0.5 to the inorganic arsenic dietary exposure estimate; the higher the MOE, the lower the health concern, and vice versa. The MOEs can be used for priority setting for risk management actions, and the level of regulatory or non-regulatory intervention can be determined taking account of the size of the MOE (Codex, 2011b). The Centre for Food Safety (CFS), being the government food safety control authority of Hong Kong, is responsible for assessing the health risks associated with food sold in Hong Kong. The CFS started to conduct its first Hong Kong Total Diet Study (TDS) in 2010 aiming to provide dietary exposure estimates of contaminants and nutrients for the Hong Kong people and various agegender subgroups. TDS approach was adopted because it has been recognised internationally as the most cost effective way to estimate the dietary exposures to chemicals or nutrients (WHO, 2006). It focuses on substances in the whole diet, not individual foods, and aims to assess dietary exposure to substances actually
380
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
ingested by the population, rather than the substance concentration in food (WHO, 2006). This article presents the first Hong Kong TDS results of the dietary exposure of people to inorganic arsenic and assessment of its associated health risk. In this study the actual inorganic arsenic levels in food as consumed were determined, rather than using generalised conversion factors to estimate the inorganic arsenic content from total arsenic measurement, which has been applied in some exposure assessment studies (JECFA, 2011). 2. Material and methods 2.1. Food Consumption data and selection of food samples The food consumption data were taken from the Hong Kong Population-based Food Consumption Survey (FCS) conducted by the CFS in 2005–2007 (FEHD, 2010). Through a quota sampling by gender and age groups, 5008 Hong Kong adults aged 20–84 were invited to complete two non-consecutive 24-h dietary intake questionnaires. The survey results revealed that over 1400 food items were being consumed by the Hong Kong people. When estimating the dietary exposure, a weighting based on age-gender group was applied to adjust for bias arising from the age-gender quotas. The weighting was based on the population distribution by age and gender in the 2006 Population By-census (FEHD, 2011, 2010). One hundred and fifty most commonly consumed food items were selected for the study, based on the food consumption pattern of the Hong Kong people. In order to cover the whole diet of the Hong Kong people, a food mapping process of the 150 TDS food items over the 1400 FCS food items was carried, and it is described under Section 2.4.
2.2. Food sampling and preparation To provide greater flexibility in calculating dietary exposure values for various segments of the population, an individual food approach was used in the study (FAO/WHO, 2009). Three samples of each TDS food item were collected and prepared in four occasions from March 2010 and February 2011. They were homogenised individually and combined into a composite sample. A total of 1800 samples were collected and combined into 600 composite samples. Based on the buying habits of the majority of the Hong Kong people, the samples were collected from a range of retail outlets, such as supermarkets, wet markets, groceries and restaurants, etc., in different areas of Hong Kong. No verification of species/varieties and the country of origin of the samples were made because it was not the focus of the study. The collected food samples were prepared as food normally consumed, i.e. table ready, in a manner consistent with cultural habits, and the preparation methods ranged from simple rinsing to cooking. Distilled water was used for food preparation. No salt and cooking oil were added during food preparation because salt and cooking oil were also considered as food items selected for inorganic arsenic testing, and the consumption amount had been captured separately (FEHD, 2011). This approach is consistent with the TDS conducted by other countries, such as Australia and New Zealand (FSANZ, 2011; NZFSA, 2011). Prepared samples were kept at 18 °C and transferred to a laboratory for analysis (normally less than 3 h). The equipment used for preparing and homogenising the composite samples was thoroughly washed between each preparation (e.g. cleaning with a laboratory-grade detergent, rinsing thoroughly with hot tap water, and rinsing thoroughly with deionised water) to avoid the risk of crosscontamination.
2.3. Chemical analysis of inorganic arsenic Inorganic arsenic refers to the sum of arsenite (As(III)) and arsenate (As(V)). The analytical method used was described by Munoz et al. (1999). In brief, the composite samples (2.5 g) were solubilised in around 3 mL deionised water and 18.4 mL concentrated hydrochloric acid. After reduction with 2 mL hydrogen bromide and 1 mL hydrazine sulphate at 15 mg/mL, arsenite was extracted into 10 mL chloroform twice. The arsenite in the combined chloroform was then back extracted to 10 mL diluted hydrochloric acid. Subsequently, the organic matters remaining in the back-extracted acid solution were removed and destructed by dry-ashing at 425 °C for 12 h. The resulting ash was dissolved in a solution of 0.5 mL deionised water and 5 mL concentrated hydrochloric acid. For quality control, a reference sample was purchased from the Food Analysis Performance Assessment Scheme (FAPAS, York, UK). The analytical quantification of inorganic arsenic was performed, using a hydride generation inductively coupled plasma mass spectrometry (Agilent ICP-MS 7500-Ce). [Note: This procedure was also known to extract a small amount of monomethylarsenic species (FSA, 2009a).] For the comparison of measured and referenced concentrations of inorganic arsenic, each test run included spiked test sam-
ples and a quality control reference material. The limits of detection (LOD) and the limits of quantification (LOQ), respectively, were 3 and 10 lg/kg in food, and 1.5 and 5 lg/kg in drinking water and bottled water. Recovery percentages (n = 218) based on a quality control reference material (FAPAS T0792 with assigned value of 67.2 mg/kg) ranged between 89 and 122. Correction for recovery percentages was not performed. The average spiking recovery percentage was 87% (RSD of ±7%). 2.4. Dietary exposure estimates The result reveals that 49% of the test samples were below the LOD. The mean levels of the four occasions for exposure estimation were calculated under the assumption that the non-detectable results equal to the half of LOD, according to the World Health Organization recommendations (WHO, 1995). The 150 TDS food items were mapped with 1400 food items captured by FCS in order to cover the whole diet of the Hong Kong people. The mean levels of the TDS food items were assigned to the mapped FCS food items with an application of conversion factors taking reference to the differences in water content (FEHD, 2011). To cite an example, cooked white rice in TDS food was mapped to cooked white rice and congee in FCS. As a result, over 99% of the food intake of the Hong Kong people was covered in the dietary exposure estimation after food mapping. Dietary exposure to inorganic arsenic of individual respondent was estimated by combining the food consumption data with inorganic arsenic level found in mapped food items according to the following formula:
Er ¼
n X F r;i C i 1000 bw r i¼1
ð1Þ
where Er is the total daily dietary exposure to inorganic arsenic of the respondent r (lg/kg bw/day), Fr,i is the daily intake of the mapped FCS food item i by the respondent r (g/day), Ci is the inorganic arsenic level of the mapped FCS food item assigned (lg/kg), bwr is the body weight of the respondent r (kg), n is the total number of mapped FCS food items consumed by the respondent r. The mean exposure level among the respondents after weighting by age-gender is used to represent the average dietary exposures of the Hong Kong people. The 95th percentile exposure level is used to represent the dietary exposure of the high consumer of the Hong Kong people. Dietary exposure estimation was performed with the aid of an in-house developed web-based computer system, EASY (Exposure Assessment System), that takes food mapping and weighting of data into consideration.
3. Results and discussion The inorganic arsenic contents of 150 TDS food items are given in Table 1. The highest levels were found in water spinach (mean: 74 lg/kg), salted eggs (mean: 58 lg/kg) and oyster (mean: 58 lg/ kg). The high level of inorganic arsenic in water spinach may be due to the aquatic growing conditions. The presence of inorganic arsenic in salted eggs may be due to the use of plant ash and/or loess (a kind of light-coloured soil) in the production process, and its presence in oyster may be due to the environmental contamination. On the other hand, all samples of ‘‘dairy products’’ and ‘‘fats and oils’’ were not detected with inorganic arsenic, and 95% of non-alcoholic beverage samples including water samples were also not detected with inorganic arsenic. Although it has been reported that water is one of the most significant sources of inorganic arsenic exposure (JECFA, 2011; WHO, 2003), water samples from this study, including drinking water and bottled water collected at various parts of Hong Kong, were not detected with inorganic arsenic. Dietary exposure to inorganic arsenic was estimated by using the inorganic arsenic level found in food items together with the food consumption data. The estimated mean inorganic arsenic dietary exposure of the Hong Kong people was 0.22 lg/kg bw/day and its 95th percentile was 0.38 lg/kg bw/day. For adult males and females (aged 20–84), the mean inorganic arsenic dietary exposure values were 0.23 and 0.21 lg/kg bw/day respectively. Among the 12 age-gender groups, the mean inorganic arsenic dietary exposures of the individual subgroups ranged from 0.19 lg/kg bw/day (female aged 20–29) to 0.26 lg/kg bw/day (male aged 60–69) and the 95th percentile range from 0.33 lg/kg bw/day (female aged 20–29) to 0.46 lg/kg bw/day (male aged 60–69) (Fig. 1). For illustration, using the average body weight of the Hong Kong people (estimated to be 61 kg), a 61-kg person is estimated to be
381
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Table 1 Inorganic arsenic content (lg/kg) in 150 food items found in the First Hong Kong Total Diet Study (1st HKTDS). Food Item
Consumption amount (g/person/day)a
No. composite samplesb
% Composite samples < LOD
Cereals and their products Rice, white Rice, unpolished Corn Noodles, Chinese or Japanese style Pasta, Western style Instant noodles Noodles, rice Bread, plain Bread, raisin ‘‘Pineapple’’ bun Sausage/ham/luncheon meat bun Chinese steamed bread Biscuits Cakes Pastries Pastries, Chinese Oatmeal Breakfast cereals Deep-fried dough
491
76
29
Vegetables and their products Carrot/Radish Potato Potato, fried Broccoli Cabbage, Chinese Cabbage, Chinese flowering Cabbage, European variety Cabbage, Petiole Chinese Celery Chinese kale Chinese spinach Leaf mustard Lettuce, Chinese Lettuce, European Mung bean sprout Spinach Water spinach Watercress Bitter melon Cucumber Hairy gourd Pumpkin Sponge gourd Wax gourd Zucchini Eggplant Sweet pepper Tomato Garlic Onion Spring onion Preserved vegetables Mushroom, dried shiitake Mushrooms Ear fungus
177
Legumes, nuts and seeds and their products Green string beans, with pod Mung bean vermicelli Beancurd Fermented bean products Peanut Peanut butter
18
Fruits Apple Banana Dragon fruit Grapes Kiwi fruit Longan/Lychee Mango Melons
147
140
24
68
Mean (lg/kg)c (ND = LOD/2) [range]
22 43 1.5 1.5 1.5 3 9 5 4 4 5 3 8 8 1.5 17 1.5 6 10
[16–26] [37–46] [ND] [ND] [ND] [ND–4] [6–10] [3–6] [3–5] [3–6] [4–6] [ND–5] [4–11] [4–21] [ND] [8–24] [ND] [3–7] [8–12]
1.5 1.5 2 1.5 2 9 1.5 5 4 3 10 8 3 1.5 4 5 74 19 4 9 1.5 1.5 2 1.5 1.5 3 1.5 1.5 9 2 14 38 45 5 11
[ND] [ND] [ND–5] [ND] [ND–4] [6–15] [ND] [3–9] [ND–7] [ND–6] [9–13] [ND–22] [ND–4] [ND] [ND–5] [ND–7] [35–120] [8–34] [ND–8] [3–18] [ND] [ND] [ND–4] [ND] [ND] [ND–5] [ND] [ND] [8–11] [ND–5] [9–18] [11–48] [36–53] [4–6] [9–14]
1.5 1.5 1.5 9 6 5
[ND] [ND] [ND] [ND–14] [3–11] [ND–13]
6 2 1.5 3 1.5 3 23 6
[ND–10] [ND–5] [ND] [ND–6] [ND] [ND–9] [ND–88] [ND–9]
49
63
78
(continued on next page)
382
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
Table 1 (continued) Food Item
Consumption amount (g/person/day)a
No. composite samplesb
% Composite samples < LOD
Orange Papaya Peach Pear Persimmon Pineapple Plum Pummelo/Grapefruit Watermelon Meat, poultry and game and their products Beef Mutton Pork Ham Luncheon meat Barbecued pork Roasted pork Pig liver Chicken meat Chicken, soy sauce Roasted duck/goose Meat sausage
113
Eggs and their products Egg, chicken Egg, lime preserved (or century egg) Egg, salted (a kind of traditional Chinese egg products and prepared by immersing with salt)
15
Fish and seafood and their products Fish, Big head Fish, Mandarin fish Fish, Grass carp Fish, Golden thread Fish, Grouper Fish, Horse head Fish, Pomfret Fish, Sole Fish, Tuna Fish, Grey mullet Fish, Salmon Fish, Yellow croaker Fish, Dace, minced Fish ball/fish cake Shrimp/Prawn Crab Oyster Scallop Squid
71
Dairy products Milk, whole Milk, skim Cheese Yoghurt Ice-cream
38
Fats and oils Butter Oil, vegetable
8
Beverages, alcoholic Beer Red wine
33
Beverages, non-alcoholic Tea, Chinese Tea, Milk tea Coffee Malt drink Soybean drink Fruit and vegetable juice Carbonated drink Tea, chrysanthemum Water, bottled, distilled Water, drinking
1625
Mixed dishes
222
48
12
76
20
8
8
40
48
Mean (lg/kg)c (ND = LOD/2) [range] 1.5 5 1.5 1.5 2 2 1.5 2 1.5
[ND] [ND–13] [ND] [ND] [ND–5] [ND–4] [ND] [ND–4] [ND]
12 1.5 1.5 4 6 5 7 3 3 1.5 1.5 4
[6–27] [ND] [ND] [3–5] [3–9] [ND–8] [ND–22] [ND–5] [ND–5] [ND] [ND] [ND–6]
1.5 10 58
[ND] [3–21] [31–93]
5 1.5 1.5 20 19 18 17 3 5 12 8 30 2 8 23 27 58 20 9
[4–6] [ND] [ND] [12–24] [13–24] [14–30] [15–21] [ND–5] [3–9] [9–14] [4–10] [6–55] [ND–4] [6–11] [15–30] [21–35] [49–74] [7–34] [4–19]
1.5 1.5 1.5 1.5 1.5
[ND] [ND] [ND] [ND] [ND]
1.5 1.5
[ND] [ND]
2 5
[ND–5] [ND–7]
1.5 1.5 1.5 1.5 1.5 5 1.5 1.5 0.75 0.75
[ND] [ND] [ND] [ND] [ND] [ND–12] [ND] [ND] [ND] [ND]
54
33
17
100
100
50
95
21
383
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Siu Mai Dumpling, steamed Dumpling, pan-fried Dumpling, including wonton Steamed barbecued pork bun Turnip cake Steamed minced beef ball Glutinous rice dumpling Steamed rice-rolls with filling Steamed rice-rolls, plain Chinese soup Hamburger Snack foods Potato chips
1
4
0
Sugars and confectionery Chocolate Granulated white sugar
5
8
63
Condiments, sauces and herbs Table salt Soya sauce Oyster sauce Tomato paste/ ketchup Cornstarch Total
11
2 976
20
600
8 5 3 3 3 10 8 12 7 10 2 3
[7–8] [3–6] [ND–6] [ND–7] [ND–4] [7–13] [4–19] [9–16] [6–8] [7–13] [ND–5] [ND–5]
8
[6–10]
5 1.5
[ND–8] [ND]
1.5 21 12 7 2
[ND] [3–65] [6–17] [ND–14] [ND–4]
40
49
ND denotes non-detected, i.e. results less than LOD. a Consumption amount in each food group was the weighted average consumption amount of the Hong Kong people extracted from the Hong Kong Population-based Food Consumption Survey (FSC), which were generated by combining the food items in that group after food mapping. Over 99% of food intake is covered. The weight of liquid food was assumed to be 1 g per 1 mL when calculating the consumption amount in each food group. The average body weight of the Hong Kong people is 61 kg. b Each composite sample is mixed with 3 samples purchased from different retail outlets. c As only 49% of results are below limit of detection (LOD), half of LOD is used for all results less than LOD in calculating the mean concentration.
exposed to 13.42 lg/day of inorganic arsenic at the mean exposure level and 23.18 lg/day at the 95th percentile exposure level. All dietary exposure estimates to inorganic arsenic of the Hong Kong people were below the BMDL0.5. The range of the calculated MOEs was from 9 to 32 for the mean exposure and from 5 to 18 for the 95th percentile exposure. Currently there is no international recommendation on the MOE values that are indicated of any health concern. The estimated percentage contribution of different food groups to the mean dietary exposure of the Hong Kong people to inorganic arsenic is given in Fig. 2. It showed that the main dietary source of inorganic arsenic was ‘‘cereals and their products’’ (53.5% of the total exposure), followed by ‘‘beverages, non-alcoholic’’ (13.0%), ‘‘vegetables and their products’’ (10.4%) and ‘‘fish and seafood and their products’’ (7.9%). The findings are similar to those TDS conducted in the UK (FSA, 2009b) and Mainland China (Li et al., 2006), where the main dietary sources of inorganic arsenic are cereals (31–45% of the total exposure), vegetables (17–18%), beverages (18–20%) and fish and seafood (3–4%). Under ‘‘cereals and their products’’, cooked white rice (including congee) (45.2% of the total exposure) is particularly a significant source of inorganic arsenic. The finding is consistent with data reported in other countries where rice is the staple food (JECFA, 2011). Among the samples in this food group, cereals including noodles, bread and oatmeal contained low levels of inorganic arsenic (mean levels ranged from 1.5 to 9 lg/kg). In contrast, cooked white rice and cooked unpolished rice contained relatively high level of inorganic arsenic. The level detected in unpolished rice (mean: 43 lg/kg) was nearly double that of white rice (mean: 22 lg/kg). Codex (2011a) and JECFA (2011) reported similar results. Torres-Escribano et al. (2008) reported that rice contained higher inorganic arsenic level compared with other food products of terrestrial origin, and brown rice had a higher arsenic level than white rice. This study showed that cooked unpolished rice (including congee) only contributed 1.3% of total exposure to inorganic
arsenic. Based on available information it was not possible to conduct a risk and benefit analysis for food safety and nutrition regarding the intake of unpolished rice. Processing and preparation may affect the level of inorganic arsenic in the rice. It has been reported that rinsing or soaking rice and discarding the water before cooking can reduce arsenic levels, especially inorganic forms, provided that uncontaminated water is used for cooking, but cooking rice with low water volume (rice to water ratio 1:1.5–2.5) does not change the arsenic content (Codex, 2011a; JECFA, 2011; Raab et al., 2009; Signes et al., 2008). The rice samples in this study were cooked in accordance with the local practice by rinsing a few times and cooking with low water volume until no water was left. It is believed that certain amount of arsenic in the rice samples had been removed during the washing process. ‘‘Non-alcoholic beverages’’ contributed 13.0% of the total exposure although majority of the samples were not detected with inorganic arsenic. This level of contribution was attributed to assigning ½ LOD value to non-detected samples. Drinking water can be a significant source of arsenic exposure in places where drinking water contains relatively high levels of arsenic (>10 lg/L) (JECFA, 2011; WHO, 2003). However, no water samples from the current study were detected with inorganic arsenic. By assigning ½ LOD value to all non-detected results in exposure estimation, drinking water contributes 6% to the exposure of inorganic arsenic (i.e. 0.014 lg/ kg bw/day). Under this conservative approach, drinking water is not considered as an important source of inorganic arsenic exposure of the Hong Kong people. The third highest contributor was ‘‘vegetables and their products’’ (10.4% of total exposure). Under this food category water spinach contributed 3% to the total dietary exposure due to the high level of inorganic arsenic (mean: 74 lg/kg), and it is considered not an important source of inorganic arsenic exposure of the Hong Kong people. The dietary exposure of the Hong Kong people was also compared with those obtained from other food authorities and the results are given in Table 2. Direct comparison of the data from
384
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
provide better estimates of long-term dietary exposure and food consumption, particularly for those rarely consumed items such as seaweeds and the Chinese herbs/medicines. Seaweeds such as hijiki are reported to contain high inorganic arsenic content (i.e. seaweed other than hijiki: normally below 2 000 lg/kg, hijiki: 30 000–130 000 lg/kg) (JECFA, 2011). The TDS food list did not include any seaweed items because of the low consumption amount (0.3 g/person/day), and it may lead to underestimation of the dietary exposure to inorganic arsenic from seaweed. In addition, occurrence of various substances in food may vary. The application of a single set of food mapping for all substances tested may cause uncertainty in the estimates of dietary exposure to individual substances. In summary, based on the data obtained in this study the mean dietary exposure to inorganic arsenic of the Hong Kong people is
different sources has to be done with caution because of the following factors, namely the difference in study time periods, research methodology, consumption data collection methods, chemical analysis methods and treatment of non-detected results. In addition, most studied analysed arsenic as total arsenic, and sometimes exposure to inorganic arsenic was estimated using the levels derived from total arsenic by using generalised conversion factors. Therefore, such estimates may be subjected to bias. In spite of these factors, it can be seen that the dietary exposure estimated in this study compares favourably with the exposure estimates reported by other food authorities. This study encountered several limitations. The exposure estimation in the study was based on the food consumption derived from two non-consecutive 24-h dietary records of the Hong Kong people. More comprehensive data on multiple-day intakes may
Dietary Exposure (µg/kg bw/day)
0.5 Mean
95th percentile
0.4 0.3 0.2 0.1
0–
–8
s (2 ale fem
All
Age-gender Groups
84)
4)
ale
(20 les
4F All
ma
–8 70
–8
4M
em
ale
ale –6
60
70
9F
9M
em
ale
ale –6 60
50
–5
9F
9M
em
ale
ale –5 50
9F –4 40
40
–4
9M
em
ale
ale em
ale
–3
9F
9M 30
–3 30
9F –2 20
20
–2
9M
em
ale
ale
0.0
Fig. 1. Mean and 95th percentile of inorganic arsenic dietary exposures among individual age-gender groups.
Meat, poultry and game and their products 3.2% Others* 3.3% Fruits 3.3% Mixed dishes 5.4% Fish and seafood and their products 7.9%
Vegetables and their products 10.4%
Cereals and their products 53.5%
Beverages, non-alcoholic 13.0% * Others include Beverages, alcoholic; Condiments, sauces and herbs; Dairy products; Eggs and their products; Fats and oils; Legumes, nuts and seeds and their products; Snack foods; Sugars and confectionery Fig. 2. Percentage contribution to the mean dietary exposure to inorganic arsenic of the population by food groups.
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Table 2 A comparison of dietary exposures to inorganic arsenic. Places
Dietary exposure of adult (lg/kg bw/day) Mean
95th or 97.5th percentile
UKa FSA (2009b) Franceb JECFA (2011) USAa JECFA (2011) Hong Konga,c New Zealandb MAF (2011) Canadab JECFA (2011) Europeb JECFA (2011) Japana,b JECFA (2011) Chinaa JECFA (2011)
0.03–0.09 0.1 0.08–0.20 0.22 0.24–0.29 0.29 0.21–0.61 0.36–0.46 0.24–0.76
0.07–0.17 (97.5th percentile) 0.27 (95th percentile) 0.16–0.34 (95th percentile) 0.38 (95th percentile)
0.36–0.99 (95th percentile) 0.83–1.29 (95th percentile)
a
Exposure data were estimated based on the detection of inorganic arsenic. Exposure data were estimated based on the detection of total arsenic and the use of conversion factors. c Data are extracted from the current study. b
0.22 lg/kg bw/day and the 95th percentile is 0.38 lg/kg bw/day. All dietary exposure estimates including those of different agegender subgroups are below the BMDL0.5. The range of the calculated MOE of inorganic arsenic is from 9 to 32 for the mean exposure and from 5 to 18 for the 95th percentile exposure. Rice, being the staple food of Hong Kong people, is estimated as a major contributor of the dietary exposure to inorganic arsenic. Having considered the carcinogenic risk of inorganic arsenic to humans, it is suggested that efforts should be made to reduce the exposure to inorganic arsenic for the population in Hong Kong. Conflict of Interest The authors declare that there are no conflicts of interest. References Codex, 2011a. Discussion Paper on Arsenic in Rice (CX/CF 11/5/10) for the Fifth Session of the Codex Committee on Contaminants in Foods, the Hague, the Netherlands, 21–25 March 2011. Available at: ftp://ftp.fao.org/codex/cccf5/ cf05_10e.pdf. Codex, 2011b. Discussion Paper on Guidance for Risk Management Options on How to Deal with the Results from New Risk Assessment Methodologies (CX/CF 11/5/ 11) for the Fifth Session of the Codex Committee on Contaminants in Foods, the Hague, the Netherlands, 21–25 March 2011. Available at: ftp://ftp.fao.org/ codex/cccf5/cf05_11e.pdf. EFSA (European Food Safety Authority), 2009. Scientific Opinion on Arsenic in Food; EFSA Panel on Contaminants in the Food Chain. The EFSA Journal 7 (10), 1351, EFSA, Parma, Italy.
385
FAO/WHO, 2009. Principles and Methods for the Risk Assessment of Chemicals in Food. Environmental Health Criteria 240. Available at: http://www.who.int/ foodsafety/chem/principles/en/index1.html. FEHD, 2010. Food and Environmental Hygiene Department. Hong Kong PopulationBased Food Consumption Survey 2005–2007 Final Report. FEHD, Hong Kong. FEHD, 2011. The First Hong Kong Total Diet Study: Methodology. FEHD, Hong Kong. Available at:. FSA, 2009a. Food Standard Agency of UK. Survey of total and inorganic arsenic in rice drinks. Food Survey Information Sheet No. 02/09. FSA. Available at: . FSA, 2009b. Measurement of the Concentrations of Metals and other Elements from the 2006 UK Total Diet Study. Food Survey Information Sheet No. 01/09. FSA. Available at: . FSANZ, 2011. Food Standards Australia New Zealand. Appendices of the 23rd Australian Total Diet Study. FSANZ. Available at: . IARC (International Agency of Research on Cancer), 2009. Special Report: Policy – A review of human carcinogens – Part C: metals, arsenic, dusts, and fibres. WHO IARC Monograph Working Group. Lancet Oncol. 10, 453–454. JECFA, 2011. Evaluation of Certain Contaminants in Food: Seventy-second Report of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report Series 959. Available at: . Li, X., Gao, J., Wang, Y., Chen, J., 2006. 2000 Chinese Total Dietary Study – the dietary arsenic intakes. J. Hyg. Res. 35 (1), 63–66 (in Chinese). MAF, 2011. Ministry of Agriculture and Forestry. 2009 New Zealand Total Diet Study. Agricultural Compound Residues, Selected Contaminant and Nutrient Elements. MAF, New Zealand. Available at: . Munoz, O., Vélez, D., Montoro, R., 1999. Optimization of the solubilization, extraction and determination of inorganic arsenic [As(III) + As(V)] in seafood products by acid digestion, solvent extraction and hydride generation atomic absorption spectrometry. Analyst 124, 601–607. NZFSA, 2011. New Zealand Food Safety Authority. 2009 New Zealand Total Diet Study. NZFSA. Available at: . Raab, A., Baskaran, C., Feldmann, J., Meharg, A., 2009. Cooking rice in a high water to rice ratio reduces inorganic arsenic content. J. Environ. Monit. 11, 41–44. Signes, A., Mitra, K., Bruló, F., Carbonell-Barrachina, A.A., 2008. Effect of two different rice dehusking procedures on total arsenic concentration in rice. Eur. Food Res. Technol. 226, 561–567. Torres-Escribano, S., Leal, M., Vélez, D., Montoro, R., 2008. Total and inorganic arsenic concentrations in rice sold in Spain, effect of cooking, and risk assessments. Environ. Sci. Technol. 42, 3867–3872. WHO, 1995. GEMS/Food-EURO Second Workshop on Reliable Evaluation of Lowlevel Contamination of Food – Report of a Workshop in the Frame of GEMS/ Food-EURO. WHO. Available from URL: . WHO, 2003. Arsenic in drinking water: Background document for development of WHO Guidelines for Drinking-water Quality. WHO. Available at: . WHO, 2006. Total Diet Studies: A Recipe for Safer Food. GEMS/Food, Food Safety Department, WHO.
Contents lists available at SciVerse ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
Dietary exposure to inorganic arsenic of the Hong Kong population: Results of the first Hong Kong Total Diet Study Waiky W.K. Wong a, Stephen W.C. Chung b, Benny T.P. Chan b, Y.Y. Ho a, Ying Xiao a,⇑ a
Centre for Food Safety (CFS), Food and Environmental Hygiene Department (FEHD), 43/F, Queensway Government Offices, 66 Queensway, Hong Kong Food Research Laboratory (FRL), Centre for Food Safety (CFS), Food and Environmental Hygiene Department (FEHD), 4/F, Public Health Laboratory Centre, 382 Nam Cheong Street, Shek Kip Mei, Kowloon, Hong Kong
b
a r t i c l e
i n f o
Article history: Received 18 April 2012 Accepted 12 October 2012 Available online 24 October 2012 Keywords: Inorganic arsenic Total diet Dietary exposure
a b s t r a c t Inorganic arsenic, a human carcinogen, can be found in the environment and food. In the first Hong Kong Total Diet Study, the dietary exposure of the Hong Kong people, including various age-gender subgroups, to inorganic arsenic was estimated for assessing the associated health risk. Food samples, which represented the Hong Kong people’s diet, were collected and prepared ‘‘as consumed’’ for analysis. Concentrations of inorganic arsenic, as sum of arsenite (As(III)) and arsenate (As(V)) were determined in 600 composite samples by using inductively coupled plasma mass spectrometry. The dietary exposures were estimated by combining the analytical results with the local food consumption data of the adult population. The mean and 95th percentile of inorganic arsenic exposures of the Hong Kong people were 0.22 and 0.38 lg/kg body weight (bw)/day, respectively. Among the 12 age-gender subgroups, the respective exposures ranged from 0.19 to 0.26 lg/kg bw/day and from 0.33 to 0.46 lg/kg bw/day. The main food category that contributed inorganic arsenic was ‘‘cereals and their products’’ (53.5% of the total exposure), particularly rice. Having considered the carcinogenic risk of inorganic arsenic to humans, it is suggested that efforts should be made to reduce the inorganic arsenic exposure of the Hong Kong population. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Inorganic arsenic is the more toxic form of arsenic, a metalloid, which is found in the environment from both natural occurrence and human activities (JECFA, 2011). Food is recognised as a major source of inorganic arsenic exposure (EFSA, 2009; JECFA, 2011; WHO, 2003). The inorganic arsenic levels found in foods and beverages usually do not exceed 100 lg/kg with mean value of less than 30 lg/kg. However, some food commodities, such as seaweed, rice, some fish and seafood commodities, may contain higher inorganic arsenic levels. The inorganic arsenic as a proportion of the total arsenic also varies among food commodities (Codex, 2011a; JECFA, 2011). Long term exposure to inorganic arsenic may cause bladder, lung and skin cancers, skin lesions, cardiovascular disease, neurotoxicity, and diabetes (EFSA, 2009; IARC, 2009; JECFA, 2011). In 2010, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) determined the inorganic arsenic Benchmark Dose Lower Confidence Limit for a 0.5% increased incidence of lung cancer in human (BMDL0.5) as 3.0 lg/kg body weight (bw)/day (in the region of 2–7 lg/kg bw/day). This BMDL0.5 contains the following uncertainties, namely, the assumptions on total exposure extrapolated ⇑ Corresponding author. Tel.: +852 28675526; fax: +852 28933547. E-mail address: [email protected] (Y. Xiao). 0278-6915/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fct.2012.10.010
from the drinking water; and the extrapolation of the BMDL0.5 to other populations due to the influence of nutritional status such as low protein intake, and other lifestyle factors on the effects observed in the studied population (JECFA, 2011). BMDL0.5 cannot be regarded as a safety reference value. A dietary exposure below the BMDL0.5 does not mean that there is no health risk (Codex, 2011b). Therefore, in order to assess the health risk of dietary exposure the margin of exposure (MOE) approach is used. The MOE for inorganic arsenic is the ratio of BMDL0.5 to the inorganic arsenic dietary exposure estimate; the higher the MOE, the lower the health concern, and vice versa. The MOEs can be used for priority setting for risk management actions, and the level of regulatory or non-regulatory intervention can be determined taking account of the size of the MOE (Codex, 2011b). The Centre for Food Safety (CFS), being the government food safety control authority of Hong Kong, is responsible for assessing the health risks associated with food sold in Hong Kong. The CFS started to conduct its first Hong Kong Total Diet Study (TDS) in 2010 aiming to provide dietary exposure estimates of contaminants and nutrients for the Hong Kong people and various agegender subgroups. TDS approach was adopted because it has been recognised internationally as the most cost effective way to estimate the dietary exposures to chemicals or nutrients (WHO, 2006). It focuses on substances in the whole diet, not individual foods, and aims to assess dietary exposure to substances actually
380
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
ingested by the population, rather than the substance concentration in food (WHO, 2006). This article presents the first Hong Kong TDS results of the dietary exposure of people to inorganic arsenic and assessment of its associated health risk. In this study the actual inorganic arsenic levels in food as consumed were determined, rather than using generalised conversion factors to estimate the inorganic arsenic content from total arsenic measurement, which has been applied in some exposure assessment studies (JECFA, 2011). 2. Material and methods 2.1. Food Consumption data and selection of food samples The food consumption data were taken from the Hong Kong Population-based Food Consumption Survey (FCS) conducted by the CFS in 2005–2007 (FEHD, 2010). Through a quota sampling by gender and age groups, 5008 Hong Kong adults aged 20–84 were invited to complete two non-consecutive 24-h dietary intake questionnaires. The survey results revealed that over 1400 food items were being consumed by the Hong Kong people. When estimating the dietary exposure, a weighting based on age-gender group was applied to adjust for bias arising from the age-gender quotas. The weighting was based on the population distribution by age and gender in the 2006 Population By-census (FEHD, 2011, 2010). One hundred and fifty most commonly consumed food items were selected for the study, based on the food consumption pattern of the Hong Kong people. In order to cover the whole diet of the Hong Kong people, a food mapping process of the 150 TDS food items over the 1400 FCS food items was carried, and it is described under Section 2.4.
2.2. Food sampling and preparation To provide greater flexibility in calculating dietary exposure values for various segments of the population, an individual food approach was used in the study (FAO/WHO, 2009). Three samples of each TDS food item were collected and prepared in four occasions from March 2010 and February 2011. They were homogenised individually and combined into a composite sample. A total of 1800 samples were collected and combined into 600 composite samples. Based on the buying habits of the majority of the Hong Kong people, the samples were collected from a range of retail outlets, such as supermarkets, wet markets, groceries and restaurants, etc., in different areas of Hong Kong. No verification of species/varieties and the country of origin of the samples were made because it was not the focus of the study. The collected food samples were prepared as food normally consumed, i.e. table ready, in a manner consistent with cultural habits, and the preparation methods ranged from simple rinsing to cooking. Distilled water was used for food preparation. No salt and cooking oil were added during food preparation because salt and cooking oil were also considered as food items selected for inorganic arsenic testing, and the consumption amount had been captured separately (FEHD, 2011). This approach is consistent with the TDS conducted by other countries, such as Australia and New Zealand (FSANZ, 2011; NZFSA, 2011). Prepared samples were kept at 18 °C and transferred to a laboratory for analysis (normally less than 3 h). The equipment used for preparing and homogenising the composite samples was thoroughly washed between each preparation (e.g. cleaning with a laboratory-grade detergent, rinsing thoroughly with hot tap water, and rinsing thoroughly with deionised water) to avoid the risk of crosscontamination.
2.3. Chemical analysis of inorganic arsenic Inorganic arsenic refers to the sum of arsenite (As(III)) and arsenate (As(V)). The analytical method used was described by Munoz et al. (1999). In brief, the composite samples (2.5 g) were solubilised in around 3 mL deionised water and 18.4 mL concentrated hydrochloric acid. After reduction with 2 mL hydrogen bromide and 1 mL hydrazine sulphate at 15 mg/mL, arsenite was extracted into 10 mL chloroform twice. The arsenite in the combined chloroform was then back extracted to 10 mL diluted hydrochloric acid. Subsequently, the organic matters remaining in the back-extracted acid solution were removed and destructed by dry-ashing at 425 °C for 12 h. The resulting ash was dissolved in a solution of 0.5 mL deionised water and 5 mL concentrated hydrochloric acid. For quality control, a reference sample was purchased from the Food Analysis Performance Assessment Scheme (FAPAS, York, UK). The analytical quantification of inorganic arsenic was performed, using a hydride generation inductively coupled plasma mass spectrometry (Agilent ICP-MS 7500-Ce). [Note: This procedure was also known to extract a small amount of monomethylarsenic species (FSA, 2009a).] For the comparison of measured and referenced concentrations of inorganic arsenic, each test run included spiked test sam-
ples and a quality control reference material. The limits of detection (LOD) and the limits of quantification (LOQ), respectively, were 3 and 10 lg/kg in food, and 1.5 and 5 lg/kg in drinking water and bottled water. Recovery percentages (n = 218) based on a quality control reference material (FAPAS T0792 with assigned value of 67.2 mg/kg) ranged between 89 and 122. Correction for recovery percentages was not performed. The average spiking recovery percentage was 87% (RSD of ±7%). 2.4. Dietary exposure estimates The result reveals that 49% of the test samples were below the LOD. The mean levels of the four occasions for exposure estimation were calculated under the assumption that the non-detectable results equal to the half of LOD, according to the World Health Organization recommendations (WHO, 1995). The 150 TDS food items were mapped with 1400 food items captured by FCS in order to cover the whole diet of the Hong Kong people. The mean levels of the TDS food items were assigned to the mapped FCS food items with an application of conversion factors taking reference to the differences in water content (FEHD, 2011). To cite an example, cooked white rice in TDS food was mapped to cooked white rice and congee in FCS. As a result, over 99% of the food intake of the Hong Kong people was covered in the dietary exposure estimation after food mapping. Dietary exposure to inorganic arsenic of individual respondent was estimated by combining the food consumption data with inorganic arsenic level found in mapped food items according to the following formula:
Er ¼
n X F r;i C i 1000 bw r i¼1
ð1Þ
where Er is the total daily dietary exposure to inorganic arsenic of the respondent r (lg/kg bw/day), Fr,i is the daily intake of the mapped FCS food item i by the respondent r (g/day), Ci is the inorganic arsenic level of the mapped FCS food item assigned (lg/kg), bwr is the body weight of the respondent r (kg), n is the total number of mapped FCS food items consumed by the respondent r. The mean exposure level among the respondents after weighting by age-gender is used to represent the average dietary exposures of the Hong Kong people. The 95th percentile exposure level is used to represent the dietary exposure of the high consumer of the Hong Kong people. Dietary exposure estimation was performed with the aid of an in-house developed web-based computer system, EASY (Exposure Assessment System), that takes food mapping and weighting of data into consideration.
3. Results and discussion The inorganic arsenic contents of 150 TDS food items are given in Table 1. The highest levels were found in water spinach (mean: 74 lg/kg), salted eggs (mean: 58 lg/kg) and oyster (mean: 58 lg/ kg). The high level of inorganic arsenic in water spinach may be due to the aquatic growing conditions. The presence of inorganic arsenic in salted eggs may be due to the use of plant ash and/or loess (a kind of light-coloured soil) in the production process, and its presence in oyster may be due to the environmental contamination. On the other hand, all samples of ‘‘dairy products’’ and ‘‘fats and oils’’ were not detected with inorganic arsenic, and 95% of non-alcoholic beverage samples including water samples were also not detected with inorganic arsenic. Although it has been reported that water is one of the most significant sources of inorganic arsenic exposure (JECFA, 2011; WHO, 2003), water samples from this study, including drinking water and bottled water collected at various parts of Hong Kong, were not detected with inorganic arsenic. Dietary exposure to inorganic arsenic was estimated by using the inorganic arsenic level found in food items together with the food consumption data. The estimated mean inorganic arsenic dietary exposure of the Hong Kong people was 0.22 lg/kg bw/day and its 95th percentile was 0.38 lg/kg bw/day. For adult males and females (aged 20–84), the mean inorganic arsenic dietary exposure values were 0.23 and 0.21 lg/kg bw/day respectively. Among the 12 age-gender groups, the mean inorganic arsenic dietary exposures of the individual subgroups ranged from 0.19 lg/kg bw/day (female aged 20–29) to 0.26 lg/kg bw/day (male aged 60–69) and the 95th percentile range from 0.33 lg/kg bw/day (female aged 20–29) to 0.46 lg/kg bw/day (male aged 60–69) (Fig. 1). For illustration, using the average body weight of the Hong Kong people (estimated to be 61 kg), a 61-kg person is estimated to be
381
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Table 1 Inorganic arsenic content (lg/kg) in 150 food items found in the First Hong Kong Total Diet Study (1st HKTDS). Food Item
Consumption amount (g/person/day)a
No. composite samplesb
% Composite samples < LOD
Cereals and their products Rice, white Rice, unpolished Corn Noodles, Chinese or Japanese style Pasta, Western style Instant noodles Noodles, rice Bread, plain Bread, raisin ‘‘Pineapple’’ bun Sausage/ham/luncheon meat bun Chinese steamed bread Biscuits Cakes Pastries Pastries, Chinese Oatmeal Breakfast cereals Deep-fried dough
491
76
29
Vegetables and their products Carrot/Radish Potato Potato, fried Broccoli Cabbage, Chinese Cabbage, Chinese flowering Cabbage, European variety Cabbage, Petiole Chinese Celery Chinese kale Chinese spinach Leaf mustard Lettuce, Chinese Lettuce, European Mung bean sprout Spinach Water spinach Watercress Bitter melon Cucumber Hairy gourd Pumpkin Sponge gourd Wax gourd Zucchini Eggplant Sweet pepper Tomato Garlic Onion Spring onion Preserved vegetables Mushroom, dried shiitake Mushrooms Ear fungus
177
Legumes, nuts and seeds and their products Green string beans, with pod Mung bean vermicelli Beancurd Fermented bean products Peanut Peanut butter
18
Fruits Apple Banana Dragon fruit Grapes Kiwi fruit Longan/Lychee Mango Melons
147
140
24
68
Mean (lg/kg)c (ND = LOD/2) [range]
22 43 1.5 1.5 1.5 3 9 5 4 4 5 3 8 8 1.5 17 1.5 6 10
[16–26] [37–46] [ND] [ND] [ND] [ND–4] [6–10] [3–6] [3–5] [3–6] [4–6] [ND–5] [4–11] [4–21] [ND] [8–24] [ND] [3–7] [8–12]
1.5 1.5 2 1.5 2 9 1.5 5 4 3 10 8 3 1.5 4 5 74 19 4 9 1.5 1.5 2 1.5 1.5 3 1.5 1.5 9 2 14 38 45 5 11
[ND] [ND] [ND–5] [ND] [ND–4] [6–15] [ND] [3–9] [ND–7] [ND–6] [9–13] [ND–22] [ND–4] [ND] [ND–5] [ND–7] [35–120] [8–34] [ND–8] [3–18] [ND] [ND] [ND–4] [ND] [ND] [ND–5] [ND] [ND] [8–11] [ND–5] [9–18] [11–48] [36–53] [4–6] [9–14]
1.5 1.5 1.5 9 6 5
[ND] [ND] [ND] [ND–14] [3–11] [ND–13]
6 2 1.5 3 1.5 3 23 6
[ND–10] [ND–5] [ND] [ND–6] [ND] [ND–9] [ND–88] [ND–9]
49
63
78
(continued on next page)
382
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
Table 1 (continued) Food Item
Consumption amount (g/person/day)a
No. composite samplesb
% Composite samples < LOD
Orange Papaya Peach Pear Persimmon Pineapple Plum Pummelo/Grapefruit Watermelon Meat, poultry and game and their products Beef Mutton Pork Ham Luncheon meat Barbecued pork Roasted pork Pig liver Chicken meat Chicken, soy sauce Roasted duck/goose Meat sausage
113
Eggs and their products Egg, chicken Egg, lime preserved (or century egg) Egg, salted (a kind of traditional Chinese egg products and prepared by immersing with salt)
15
Fish and seafood and their products Fish, Big head Fish, Mandarin fish Fish, Grass carp Fish, Golden thread Fish, Grouper Fish, Horse head Fish, Pomfret Fish, Sole Fish, Tuna Fish, Grey mullet Fish, Salmon Fish, Yellow croaker Fish, Dace, minced Fish ball/fish cake Shrimp/Prawn Crab Oyster Scallop Squid
71
Dairy products Milk, whole Milk, skim Cheese Yoghurt Ice-cream
38
Fats and oils Butter Oil, vegetable
8
Beverages, alcoholic Beer Red wine
33
Beverages, non-alcoholic Tea, Chinese Tea, Milk tea Coffee Malt drink Soybean drink Fruit and vegetable juice Carbonated drink Tea, chrysanthemum Water, bottled, distilled Water, drinking
1625
Mixed dishes
222
48
12
76
20
8
8
40
48
Mean (lg/kg)c (ND = LOD/2) [range] 1.5 5 1.5 1.5 2 2 1.5 2 1.5
[ND] [ND–13] [ND] [ND] [ND–5] [ND–4] [ND] [ND–4] [ND]
12 1.5 1.5 4 6 5 7 3 3 1.5 1.5 4
[6–27] [ND] [ND] [3–5] [3–9] [ND–8] [ND–22] [ND–5] [ND–5] [ND] [ND] [ND–6]
1.5 10 58
[ND] [3–21] [31–93]
5 1.5 1.5 20 19 18 17 3 5 12 8 30 2 8 23 27 58 20 9
[4–6] [ND] [ND] [12–24] [13–24] [14–30] [15–21] [ND–5] [3–9] [9–14] [4–10] [6–55] [ND–4] [6–11] [15–30] [21–35] [49–74] [7–34] [4–19]
1.5 1.5 1.5 1.5 1.5
[ND] [ND] [ND] [ND] [ND]
1.5 1.5
[ND] [ND]
2 5
[ND–5] [ND–7]
1.5 1.5 1.5 1.5 1.5 5 1.5 1.5 0.75 0.75
[ND] [ND] [ND] [ND] [ND] [ND–12] [ND] [ND] [ND] [ND]
54
33
17
100
100
50
95
21
383
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Siu Mai Dumpling, steamed Dumpling, pan-fried Dumpling, including wonton Steamed barbecued pork bun Turnip cake Steamed minced beef ball Glutinous rice dumpling Steamed rice-rolls with filling Steamed rice-rolls, plain Chinese soup Hamburger Snack foods Potato chips
1
4
0
Sugars and confectionery Chocolate Granulated white sugar
5
8
63
Condiments, sauces and herbs Table salt Soya sauce Oyster sauce Tomato paste/ ketchup Cornstarch Total
11
2 976
20
600
8 5 3 3 3 10 8 12 7 10 2 3
[7–8] [3–6] [ND–6] [ND–7] [ND–4] [7–13] [4–19] [9–16] [6–8] [7–13] [ND–5] [ND–5]
8
[6–10]
5 1.5
[ND–8] [ND]
1.5 21 12 7 2
[ND] [3–65] [6–17] [ND–14] [ND–4]
40
49
ND denotes non-detected, i.e. results less than LOD. a Consumption amount in each food group was the weighted average consumption amount of the Hong Kong people extracted from the Hong Kong Population-based Food Consumption Survey (FSC), which were generated by combining the food items in that group after food mapping. Over 99% of food intake is covered. The weight of liquid food was assumed to be 1 g per 1 mL when calculating the consumption amount in each food group. The average body weight of the Hong Kong people is 61 kg. b Each composite sample is mixed with 3 samples purchased from different retail outlets. c As only 49% of results are below limit of detection (LOD), half of LOD is used for all results less than LOD in calculating the mean concentration.
exposed to 13.42 lg/day of inorganic arsenic at the mean exposure level and 23.18 lg/day at the 95th percentile exposure level. All dietary exposure estimates to inorganic arsenic of the Hong Kong people were below the BMDL0.5. The range of the calculated MOEs was from 9 to 32 for the mean exposure and from 5 to 18 for the 95th percentile exposure. Currently there is no international recommendation on the MOE values that are indicated of any health concern. The estimated percentage contribution of different food groups to the mean dietary exposure of the Hong Kong people to inorganic arsenic is given in Fig. 2. It showed that the main dietary source of inorganic arsenic was ‘‘cereals and their products’’ (53.5% of the total exposure), followed by ‘‘beverages, non-alcoholic’’ (13.0%), ‘‘vegetables and their products’’ (10.4%) and ‘‘fish and seafood and their products’’ (7.9%). The findings are similar to those TDS conducted in the UK (FSA, 2009b) and Mainland China (Li et al., 2006), where the main dietary sources of inorganic arsenic are cereals (31–45% of the total exposure), vegetables (17–18%), beverages (18–20%) and fish and seafood (3–4%). Under ‘‘cereals and their products’’, cooked white rice (including congee) (45.2% of the total exposure) is particularly a significant source of inorganic arsenic. The finding is consistent with data reported in other countries where rice is the staple food (JECFA, 2011). Among the samples in this food group, cereals including noodles, bread and oatmeal contained low levels of inorganic arsenic (mean levels ranged from 1.5 to 9 lg/kg). In contrast, cooked white rice and cooked unpolished rice contained relatively high level of inorganic arsenic. The level detected in unpolished rice (mean: 43 lg/kg) was nearly double that of white rice (mean: 22 lg/kg). Codex (2011a) and JECFA (2011) reported similar results. Torres-Escribano et al. (2008) reported that rice contained higher inorganic arsenic level compared with other food products of terrestrial origin, and brown rice had a higher arsenic level than white rice. This study showed that cooked unpolished rice (including congee) only contributed 1.3% of total exposure to inorganic
arsenic. Based on available information it was not possible to conduct a risk and benefit analysis for food safety and nutrition regarding the intake of unpolished rice. Processing and preparation may affect the level of inorganic arsenic in the rice. It has been reported that rinsing or soaking rice and discarding the water before cooking can reduce arsenic levels, especially inorganic forms, provided that uncontaminated water is used for cooking, but cooking rice with low water volume (rice to water ratio 1:1.5–2.5) does not change the arsenic content (Codex, 2011a; JECFA, 2011; Raab et al., 2009; Signes et al., 2008). The rice samples in this study were cooked in accordance with the local practice by rinsing a few times and cooking with low water volume until no water was left. It is believed that certain amount of arsenic in the rice samples had been removed during the washing process. ‘‘Non-alcoholic beverages’’ contributed 13.0% of the total exposure although majority of the samples were not detected with inorganic arsenic. This level of contribution was attributed to assigning ½ LOD value to non-detected samples. Drinking water can be a significant source of arsenic exposure in places where drinking water contains relatively high levels of arsenic (>10 lg/L) (JECFA, 2011; WHO, 2003). However, no water samples from the current study were detected with inorganic arsenic. By assigning ½ LOD value to all non-detected results in exposure estimation, drinking water contributes 6% to the exposure of inorganic arsenic (i.e. 0.014 lg/ kg bw/day). Under this conservative approach, drinking water is not considered as an important source of inorganic arsenic exposure of the Hong Kong people. The third highest contributor was ‘‘vegetables and their products’’ (10.4% of total exposure). Under this food category water spinach contributed 3% to the total dietary exposure due to the high level of inorganic arsenic (mean: 74 lg/kg), and it is considered not an important source of inorganic arsenic exposure of the Hong Kong people. The dietary exposure of the Hong Kong people was also compared with those obtained from other food authorities and the results are given in Table 2. Direct comparison of the data from
384
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385
provide better estimates of long-term dietary exposure and food consumption, particularly for those rarely consumed items such as seaweeds and the Chinese herbs/medicines. Seaweeds such as hijiki are reported to contain high inorganic arsenic content (i.e. seaweed other than hijiki: normally below 2 000 lg/kg, hijiki: 30 000–130 000 lg/kg) (JECFA, 2011). The TDS food list did not include any seaweed items because of the low consumption amount (0.3 g/person/day), and it may lead to underestimation of the dietary exposure to inorganic arsenic from seaweed. In addition, occurrence of various substances in food may vary. The application of a single set of food mapping for all substances tested may cause uncertainty in the estimates of dietary exposure to individual substances. In summary, based on the data obtained in this study the mean dietary exposure to inorganic arsenic of the Hong Kong people is
different sources has to be done with caution because of the following factors, namely the difference in study time periods, research methodology, consumption data collection methods, chemical analysis methods and treatment of non-detected results. In addition, most studied analysed arsenic as total arsenic, and sometimes exposure to inorganic arsenic was estimated using the levels derived from total arsenic by using generalised conversion factors. Therefore, such estimates may be subjected to bias. In spite of these factors, it can be seen that the dietary exposure estimated in this study compares favourably with the exposure estimates reported by other food authorities. This study encountered several limitations. The exposure estimation in the study was based on the food consumption derived from two non-consecutive 24-h dietary records of the Hong Kong people. More comprehensive data on multiple-day intakes may
Dietary Exposure (µg/kg bw/day)
0.5 Mean
95th percentile
0.4 0.3 0.2 0.1
0–
–8
s (2 ale fem
All
Age-gender Groups
84)
4)
ale
(20 les
4F All
ma
–8 70
–8
4M
em
ale
ale –6
60
70
9F
9M
em
ale
ale –6 60
50
–5
9F
9M
em
ale
ale –5 50
9F –4 40
40
–4
9M
em
ale
ale em
ale
–3
9F
9M 30
–3 30
9F –2 20
20
–2
9M
em
ale
ale
0.0
Fig. 1. Mean and 95th percentile of inorganic arsenic dietary exposures among individual age-gender groups.
Meat, poultry and game and their products 3.2% Others* 3.3% Fruits 3.3% Mixed dishes 5.4% Fish and seafood and their products 7.9%
Vegetables and their products 10.4%
Cereals and their products 53.5%
Beverages, non-alcoholic 13.0% * Others include Beverages, alcoholic; Condiments, sauces and herbs; Dairy products; Eggs and their products; Fats and oils; Legumes, nuts and seeds and their products; Snack foods; Sugars and confectionery Fig. 2. Percentage contribution to the mean dietary exposure to inorganic arsenic of the population by food groups.
W.W.K. Wong et al. / Food and Chemical Toxicology 51 (2013) 379–385 Table 2 A comparison of dietary exposures to inorganic arsenic. Places
Dietary exposure of adult (lg/kg bw/day) Mean
95th or 97.5th percentile
UKa FSA (2009b) Franceb JECFA (2011) USAa JECFA (2011) Hong Konga,c New Zealandb MAF (2011) Canadab JECFA (2011) Europeb JECFA (2011) Japana,b JECFA (2011) Chinaa JECFA (2011)
0.03–0.09 0.1 0.08–0.20 0.22 0.24–0.29 0.29 0.21–0.61 0.36–0.46 0.24–0.76
0.07–0.17 (97.5th percentile) 0.27 (95th percentile) 0.16–0.34 (95th percentile) 0.38 (95th percentile)
0.36–0.99 (95th percentile) 0.83–1.29 (95th percentile)
a
Exposure data were estimated based on the detection of inorganic arsenic. Exposure data were estimated based on the detection of total arsenic and the use of conversion factors. c Data are extracted from the current study. b
0.22 lg/kg bw/day and the 95th percentile is 0.38 lg/kg bw/day. All dietary exposure estimates including those of different agegender subgroups are below the BMDL0.5. The range of the calculated MOE of inorganic arsenic is from 9 to 32 for the mean exposure and from 5 to 18 for the 95th percentile exposure. Rice, being the staple food of Hong Kong people, is estimated as a major contributor of the dietary exposure to inorganic arsenic. Having considered the carcinogenic risk of inorganic arsenic to humans, it is suggested that efforts should be made to reduce the exposure to inorganic arsenic for the population in Hong Kong. Conflict of Interest The authors declare that there are no conflicts of interest. References Codex, 2011a. Discussion Paper on Arsenic in Rice (CX/CF 11/5/10) for the Fifth Session of the Codex Committee on Contaminants in Foods, the Hague, the Netherlands, 21–25 March 2011. Available at: ftp://ftp.fao.org/codex/cccf5/ cf05_10e.pdf. Codex, 2011b. Discussion Paper on Guidance for Risk Management Options on How to Deal with the Results from New Risk Assessment Methodologies (CX/CF 11/5/ 11) for the Fifth Session of the Codex Committee on Contaminants in Foods, the Hague, the Netherlands, 21–25 March 2011. Available at: ftp://ftp.fao.org/ codex/cccf5/cf05_11e.pdf. EFSA (European Food Safety Authority), 2009. Scientific Opinion on Arsenic in Food; EFSA Panel on Contaminants in the Food Chain. The EFSA Journal 7 (10), 1351, EFSA, Parma, Italy.
385
FAO/WHO, 2009. Principles and Methods for the Risk Assessment of Chemicals in Food. Environmental Health Criteria 240. Available at: http://www.who.int/ foodsafety/chem/principles/en/index1.html. FEHD, 2010. Food and Environmental Hygiene Department. Hong Kong PopulationBased Food Consumption Survey 2005–2007 Final Report. FEHD, Hong Kong. FEHD, 2011. The First Hong Kong Total Diet Study: Methodology. FEHD, Hong Kong. Available at: