A longitudinal assessment of aluminum contents in foodstuffs and aluminum intake of residents in Tianjin metropolis

Abstract Aim In this report, we retrieved and analyzed the data of aluminum contents in foodstuffs over a 6‐year span between 2010 and 2015 and assessed the risk of dietary aluminum exposure in residents of Tianjin metropolis. Methods A multistage random clustering method was used to survey Tianjin residents between 2010 and 2015. Samples were mainly purchased from breakfast vendors, farmers’ markets, and supermarkets in Tianjin between 2009 and 2015. A total of 1,814 persons aged at least 2 years from 1,262 households from randomly chosen communities were asked to complete the questionnaire on food consumption. Aluminum contents in the food samples were determined. Results Totally 21.14% of food samples exceeded the recommended aluminum residue limit over the study period. The mean aluminum levels in the food samples over the 6‐year span were 111.9 7 ± 265.26 mg/kg, and the mean P95 was 597.00 mg/kg. Totally 21.14% of the food samples exceeded the recommended aluminum residue limit (100 mg/kg). The lowest mean aluminum levels in food were detected in 2010, and the highest levels were found in 2015. The highest mean aluminum levels were found in jellyfish. The highest total mean aluminum intake in food was 83.61 mg/day in those aged at least 50 years and younger than 66 years. Meanwhile, children aged at least 2 years and less than 8 years had the highest mean weekly aluminum intake (18.19 mg/kg body weight/week); they also had the highest MOS (18.19). Conclusion The findings indicate that despite the implementation since 2014 of the new policy on the use of aluminum food additives in China, residents in Tianjin still face high levels of aluminum exposure in foodstuffs with young children particularly vulnerable. Public awareness of the new policy should be enhanced, and more vigorous supervision of the use of aluminum food additives should be undertaken.


| INTRODUC TI ON
Aluminum is present widely in nature and is in fact the third most abundant element on earth. Toxicities of undue intake by humans of aluminum and aluminum compounds have been well documented (Campbell, Hamai, & Bondy, 2001;Chen et al., 2007).
Aluminum intake via external contacts such as inhalation or dermal uptake contributes little to aluminum toxicity while aluminum, and aluminum compounds, which are used in foodstuffs, constitute a major source of aluminum intake and toxicity. Migration of aluminum from food contact materials to foods also represents a source of aluminum intake (Ma et al., 2016;Stahl et al., 2017aStahl et al., , 2017bStahl et al., , 2017c. Aluminum potassium sulfate and aluminum ammonium sulfate were no longer allowed in steamed bun and steamed cake, and other flour products excluding fried noodles, paste with batter, breading, and frying powder, and the use of aluminum-containing additives in puffed foods were also no longer permitted. Daily intake of aluminum in China has been investigated, and the studies typically focus on distinct geographical areas such as Shanghai (Guo et al., 2015), Zhejiang province (Zhang et al., 2016), the Pearl River Delta of South China , and Hong Kong (Wong, Chung, Kwong, Yin Ho, & Xiao, 2010 (Xu et al., 1993). Tianjin started its program in 2009 for monitoring contaminants in foodstuffs by purchasing and analyzing food samples for aluminum exposure and other toxic contaminants. In this report, we retrieved and analyzed the data of aluminum contents in foodstuffs over a 6-year span between 2010 and 2015 and assessed the risk of dietary aluminum exposure in residents of Tianjin metropolis.

| Sources of food samples
The sampling technique strictly followed the technical requirements for food sampling analysis specified in the "National Food Safety Risk Handbook." Samples analyzed in the current study were mainly purchased from breakfast vendors, farmers' markets, and supermarkets in Tianjin between 2009 and 2015, and included fermented pastry (steamed bun and steamed twisted), deep-fried pastry (deep-fried dough sticks, cakes, or twists), nonfermented pastry (gluten, instant noodles, and noodles), bakery pastry (bread), puffed foods (shrimp chips, French fries, egg crisp cake, and rice cake), jellyfish, algae products, baby foods including infant formula food and infant supplementary food, aquatic products, starches including vermicelli, fine vermicelli, and vermicelli sheets, bean jelly, and steamed cold noodles. specified for other food types. According to international guidelines (GEMS/Food, 1995), the rate of nondetected values was lower than 60% and the nondetected values were assumed to be LOD/2.

| Food consumption survey
A multistage random clustering method was used to survey Tianjin residents between 2010 and 2015. A total of 1,814 persons aged at least 2 years from 1,262 households from randomly chosen communities were asked to complete the questionnaire on food consumption (questionnaire is available upon request).
The study protocol was approved by the ethics committee at the authors' affiliated institutions, and questionnaire respondents provided written consent.

| Exposure assessment
The PTWI of aluminum in foods (2 mg/kg body weight) according to the JECFA (Joint FAO/WHO Expert Committee on Food Additives 2011) was used as a reference (JECFA, 2011). Margin of safety (MOS) for aluminum in foods was the weekly aluminum exposure in foods divided by PTWI, where MOS ≥ 1 indicates unacceptable food safety risk and that risk mitigation measures should be undertaken.
The body weight of infants, children, and adults used their respective standard body weight of 10, 30, and 60 kg. Exposure assessment was made according to "Food Safety Risk Analysis: Chemical Hazard Assessment." (Luo & Wu, 2012) Subjects were categorized into six age groups according to "Chinese Dietary Reference Intakes." (Zhai & Yang, 2006) Group 1 included persons aged equal to or more than 2 years and less than 8 years, group 2 included persons aged equal to or more than 8 years and less than 13 years, group 3 included persons aged equal to or more than 13 years and less than 20 years, group 4 included persons aged equal to or more than 20 years and less than 50 years, group 5 included persons aged equal to or more than 50 years and less than 66 years, and group 6 included persons aged at least 66 years. Infants (birth to 23 months) were not included. A standard body weight of 20, 40, and 50 was used for groups 1, 2, and 3, respectively, and a standard body weight of 60 kg was used for groups 4, 5, and 6.

| Statistical analysis
Categorical data were expressed using number and frequency (%), and numerical data were expressed using mean ± standard deviation, median (quartile), and P95. Student's t test was used for comparison between two groups for normally distributed data. ANOVA was used for comparison among three groups, and the SNK method was used to compare two groups for non-normally distributed data, and the Wilcoxon two-sample test was used to compare two groups, and the Kruskal-Wallis test to compare three groups. The SNK method was used for rank transformation. Statistical analysis software SAS 9.3 was used. The test was two-sided, and p < 0.05 indicated statistically significant difference.

| Aluminum content in main food types in Tianjin
We analyzed totally 1,263 food samples between 2010 and 2015.
The aluminum contents in the food samples are listed in Table 1.
The mean aluminum levels in the food samples over the 6-year span were 111.9 7 ± 265.26 mg/kg, and the mean P95 was 597.00 mg/ kg. Totally 21.14% of the food samples exceeded the recommended aluminum residue limit (100 mg/kg). The lowest mean aluminum levels in food were detected in 2010 (41.20 ± 45.01 mg/kg; range, 12.50-221.00 mg/kg), and the highest levels were found in 2015 (125.31 ± 315.65 mg/kg; range, 0.35-2,810.00 mg/kg). The lowest P95 was seen in 2010 (139.00 mg/kg) while the highest was observed in 2012 (929.50 mg/kg). In 2010, 11.11% of the food samples analyzed exceeded the recommended aluminum residue limit, which was noticeably lower compared to 38.50% in 2012 and 24.11% in 2014.

| Daily food consumption and exposure to aluminum in foods by age
The amount of daily food consumption by residents in Tianjin is shown in  (Table 4 and Table S2).

| Dietary aluminum intake in Tianjin residents
The daily and weekly dietary aluminum intake of Tianjin residents is shown in Table 5. The highest total mean aluminum intake in food was 83.61 mg/day in those aged at least 50 years and younger than 66 years, followed by those aged 66 years or older (74.28 mg/day) and those aged at least 20 years and younger than 50 years (72.04).
Those aged at least 8 years and younger than 13 years had the least mean aluminum intake (49.54 mg/day), followed by those aged at least 2 years and younger than 8 years (51.97 mg/day) and those aged at least 13 years and younger than 20 years (64.41 mg/day). had the highest P95 for mean weekly aluminum intake (57.70 mg/kg body weight/week); they also had the highest MOS (28.85). We found that flour products and vegetables were the two main food types consumed by Tianjin residents. Consistently, flour products and vegetables remained the two major sources of aluminum intake across all age groups and contributed the majority (87.41%) of aluminum exposure in foodstuffs. Our data indicated that Tianjin residents would have a daily aluminum intake anywhere from 49.4 mg/day for those aged at least 8 years and less than 13 years to 83.61 mg/day in those aged at least 50 years and younger than 66 years. This is apparently higher than that (3-10 mg/day) reported in an earlier study in 1993 on Tianjin residents (Xu et al., 1993).  (Jiang, Huang, & Zhang, 2013). Young children are at a critical stage of neurophysiologic development. We found that those aged at least 2 years and less than 8 years had the highest MOS for daily and weekly aluminum intake, suggesting that this particular age group may be vulnerable to excess aluminum intake.

| D ISCUSS I ON
In conclusion, the current study is the first longitudinal assess-

ACK N OWLED G M ENT
None declared.

CO N FLI C T O F I NTE R E S T S TATE M E NT S
The authors declare that they do not have any conflict of interest.

E TH I C A L R E V I E W
This study was approved by the ethics committee of Tianjin Center for Disease Control and Prevention and conducted in accordance with Helsinki's Declaration.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.