Exposure assessment of dietary cadmium: findings from shanghainese over 40 years, China

Cadmium is one of the trace metals widely used in different types of industries. However, cadmium can be a source of soil and water pollution after extraction and discharge. Cadmium contamination is a major environmental health problem, and is potentially harmful to the broader population through the food chain [1]. The half-life of cadmium is 10 to 30 years in humans. Cadmium slowly accumulates in organs, such as the kidneys and bone, and causes chronic damage, such as renal dysfunction and fractures [2]. Long-term exposure to low-dose cadmium has been associated with increased oxidative stress and tubular impairment [3, 4]. Additionally, there is a dose–response relationship between urinary cadmium levels and some chronic diseases, such as diabetes and high blood pressure [5, 6]. In 1993, IARC classified cadmium and its compounds as human carcinogens (primarily lung cancer) [7]. Recent studies have shown that dietary cadmium exposure is associated with the development of postmenopausal breast cancer in women [8].

To prevent health risk from cadmium contamination, a provisional tolerable weekly intake (PTWI) for cadmium of 7 μg/kg body weight was established by the Joint FAO/WHO Expert Committee on Food Additives in 2004 [9]. In 2010, the 73rd JECFA reevaluated cadmium intake levels based on findings from a number of recent epidemiological studies and established a provisional tolerable monthly intake (PTMI) of 25 μg/kg body weight based on the long half-life of cadmium [10].

Cadmium contamination is severe and prevalent in some areas of China. As reported, food is the main source of cadmium exposure for the non-occupationally exposed population (WHO 1992). In recent years, it has been reported that cadmium levels in some food in Shanghai have exceeded the National Maximum Level as follows: 27.2% in aquatic products, 2.7% in animal innards and 8.2% in grains [11]. Therefore, it was necessary to evaluate dietary cadmium exposure and its health risk for future policymaking on cadmium contamination prevention.

This study aimed to assess dietary cadmium exposure and other environmental cadmium exposure compared with the PTWI standard among adult residents (over 40 years of age) in Shanghai, China, and to determine the major contributors of total cadmium exposure. Food, tobacco and water cadmium levels were combined to evaluate the external cadmium exposure, and internal exposure levels were assessed. Therefore, this study may provide scientific evidence for cadmium monitoring, risk management and policymaking.

A cross-sectional survey concerning food consumption was conducted among Shanghai residents. The study was approved by the institutional review board of the School of Public Health, Fudan University. Informed consent from participants was obtained prior to enrollment in the study.

This study provided a detailed assessment of environmental cadmium exposure, especially dietary exposure among middle-aged and senior residents in Shanghai, China. We found that the mean dietary cadmium exposure level of the population was far below the PTDI value, and a very low percentage of the population (6.6%) exceeded the PTDI. Moreover, 93.4% of the residents had no health risks, even if they had consumed all types of food with detectable cadmium levels throughout their lifetime. Men had higher dietary cadmium exposure levels than women, which could be explained by smoking and dietary behaviors, especially consuming food with high cadmium concentrations, such as vegetables and grains. Because the food supply in Shanghai does not meet local demand and much is imported from other provinces or other Asian countries, the food cadmium level did not simply reflect the local pollution, but that of a much wider area. These data could also represent pollution conditions in other non-polluted coastal areas in China other than Shanghai because of similar dietary patterns.

To better understand the food cadmium administration condition in China, the Chinese food cadmium maximum limits (MLs) are listed in Table 6. The Chinese MLs quoted national standard and international MLs were from WHO and European Food Safety Authority (ESFA) [21‐25]. The rice MLs in China were lower than those of the WHO. Rice cadmium levels in our study were much less than the MLs in China (90th percentile?=?0.09 mg/kg). However, it is necessary to reevaluate whether the MLs of rice should be increased to 0.4 mg/kg because rice is the main food of Chinese, and many planting areas exhibit cadmium pollution in China. The meat MLs were higher than those of the WHO. In this study, the 90th percentile of cadmium levels in bovine animals was 0.1 mg/kg, which could be caused by polluted feed and contamination in processing. Higher fish MLs may be caused by water and feed pollution. Table 2 shows that food cadmium levels in this study were lower compared with levels in 2000 in Shanghai, which might indicate a reduction in cadmium contamination.

The local total daily intake (TDI?=?12.73 μg/day) of cadmium was lower than the levels in other rice-dependent Asian countries. The cadmium TDI of Bangladeshis was estimated as 34.55 μg/day, which was 57.8% of the PTDI [26]. Average daily cadmium exposure was 14 μg in the South Korean population, representing approximately 27% of the TDI [27]. Dietary cadmium intake of the general Japanese population was approximately 25 to 30 μg/day [28]. Other units of exposure, such as kg per bodyweight, are used as the unit of cadmium exposure in Western countries. Therefore, for better comparability, these data were converted by assuming an adult average bodyweight of 60 kg. The TDI was 22.2 μg/day in Germany [29], 8.4 μg/day in Catalonia (Spain) [30], 8.6 μg/day in Sweden [31], 8.4 μg/day in Belgium [32] and 20.9 μg/day in France [33]. In these countries, the highest contribution to total cadmium intake corresponded to tubers, such as potatoes, cereals, and pulses, and aquatic products [29‐35]. Although rice, which is the main food in Shanghai, was prone to cadmium accumulation, the data did not show a significant difference between Shanghai and Western countries in cadmium accumulation in other foods.

Compared with other cities in China, the local total daily intake (TDI?=?12.73 μg/day) of cadmium in Shanghai was lower than that in the Guangdong province (43.22 μg/day), where the soil cadmium background value was higher than other places [36] but slightly higher than Wuhan (10.8 μg/day), another non-polluted city in the middle of China [37]. The average TDI was 15.25 μg/day for adults in northern China and 27.6 μg/day in three southern areas (Shanghai, Jiangxi Province and Fujian Province) according to Chinese national dietary research in 2000 [17]. One reason for the decrease in food cadmium exposure might be because intake of certain foods, especially rice, has been decreasing in the past decades due to changing dietary patterns in China [38]. Our study found that vegetables, rice and aquatic products contributed most to dietary cadmium exposure, which was consistent with findings from Chinese national dietary research in 2000 [18]. The relative dietary contributions differed in other regions, indicating dietary pattern differences. Sensitivity analysis showed that cadmium in vegetables contributed most to dietary cadmium exposure, followed by rice and aquatic products. Other Asian studies have found that cadmium in rice and vegetables contributed most to dietary exposure [26, 27]. Therefore, controlling the cadmium concentration in these particular food items could most likely reduce dietary cadmium exposure in Shanghai.

Of the total population older than 15 years, 35.8% were smokers according to the National Survey in 2002 [39]. The percentage of men and women smokers were 66.0% and 3.1%, respectively, which was confirmed by our study (66.2% for men and 3.3% for women). Tobacco cadmium through inhalation is another important source of non-occupational cadmium exposure, of which the concentration could be approximately 0.5 to 5 mg/kg [40]. According to this study, exposure through tobacco cadmium inhalation appeared to be the second greatest source of non-occupational cadmium exposure in the population. Our study demonstrated that tobacco cadmium inhalation accounted for 25% of the food cadmium intake and correlated with BCd. Smoking inhabitants were 14.0% more likely to exceed the PTDI compared with non-smokers. The study also confirmed that tobacco cadmium exposure (13.8 μg/day) was approximately equivalent to the total dietary cadmium exposure (14.0 μg/day). The same conclusion was found by Van Assche who constructed a model indicating that approximately 50% of the total cadmium intake was derived from cigarettes in smokers, whereas ingestion accounted for 95% of total cadmium intake in non-smokers [41]. Madeddu et al. concluded that smoking was the most important risk factor for blood cadmium concentration in the selected population [42]. National Health and Nutrition Examination Surveys (NHANES) from 1988 to 2008 indicated that declining smoking rates and changes in exposure to tobacco smoke might have played an important role in the decline of urine cadmium concentrations [43]. Additionally, a significant and higher difference in blood cadmium concentrations was observed in smokers compared with non-smokers. The study indicated that one effective way to decrease the risk of non-occupational cadmium exposure might be to reduce the smoking rate.

External exposure represented a cadmium exposure from environmental sources, whereas the total internal cadmium was estimated in body organs and the bloodstream. There is no unified diagnostic cadmium criterion for the general population in China. The diagnostic criterion of occupational cadmium poisoning (GBZ17-2002) is 5 μg/g creatinine (5 μg/L) for urinary cadmium, whereas the discriminant standard for health hazard caused by environmental cadmium pollution (BGT17221-1998) is 15 μg/g creatinine, which is the recommended upper limit for an environmental cadmium diagnosis in China [44]. Moreover, American Conference of Industrial Hygienists established a similar standard that the biological exposure index for cadmium is 5 μg/L in blood and 5 μg/g creatinine in urine for workers [45]. A health-based limit of 10 μg/L cadmium in blood and 10 μg/g creatinine in urine was established by the WHO [46]. Local participants did not appear to be adversely effected by recent environmental exposure according to urinary cadmium levels, which reflected the body burden over long-term exposure. Body cadmium levels (i.e., UCd) increased with age because of the accumulation of cadmium. A positive correlation between BCd and environmental cadmium exposure indicated recent exposure status.

Exposure assessment, which is an important step in risk assessment, describes the contaminant level in humans instead of extrapolating from animal models. One of the strengths of our study is that we conducted both point and probabilistic estimations to assess cadmium intake including from water and tobacco. Many other references assessed single food items or used point estimations only. For example, the national cadmium exposure survey 2000 in China selected 12 provinces to assess total dietary cadmium exposure, but performed only point estimates without cadmium detection in urine and blood. The advantage of using probabilistic estimation is that all of the available information about variability and inherent uncertainty are included in the assessment, whereas for point estimation most of the information about the variability and uncertainty is discarded, and only one point value is selected [47]. Distribution functions for exposure or risk estimates display a range of exposure or risk and the probability associated with each value. Second, the point estimation may calculate only the conservative or high-end estimated value, which may be far above realistic estimates when determining an upper bound of risk. However, the probabilistic estimation provides a reasonable value that is calculated by the distribution of exposure [48]. Third, sensitivity analysis can be used in the probabilistic estimation to assess the contribution of each risk factor. More data and accurate assumptions are required to build a probabilistic estimation model, which is a shortcoming. We applied Monte Carlo simulation for probabilistic assessment to analyze the cadmium dietary exposure distribution and compared the contributions of different food types with exposure. This model is a widely used computational method with risk parameters, and was first described in 1949 by Metropolis and Ulam [49]. It can be concluded that with appropriate data and simulation conducted with a sufficiently large number of “iterations”, the results will simulate actual conditions [9]. Furthermore, BCd and UCd were collected in the study to directly evaluate cadmium accumulation in the body and illustrate potential health outcomes.

This study had some limitations that should be considered in further investigations. Considering the precision and validity of the FFQ, we selected adult participants (age?>?40 years) but excluded children who were also susceptible to health hazards, such as early kidney toxicity and neurotoxic effects, from cadmium exposure [50]. Major food categories were considered in the food safety audit, but specific food items, which are rarely consumed, were not excluded. Therefore, the results might be slightly lower than the actual level for the PTDI. Cadmium exposure during food processing, which might have led to an overestimated exposure level, was not considered. Biomarkers that indicated early renal dysfunction were not measured. Other trace metals were not detected in this study and might have similar mechanisms of action as cadmium. The Monte Carlo simulation, which depends on the accuracy of parameters, may be inaccurate at the extreme upper and lower ends of the distribution [9]. A larger sample size might have allowed the use of a more suitable probabilistic distribution model.

This study assessed total cadmium exposure including food, water and tobacco in a non-occupational population. The dietary structure of the Chinese is complex, and chemical toxicant concentrations vary by food type. Therefore, the dietary survey conducted in this study helped identify items accurately and calculated the total amount of cadmium contaminant exposure. The risk level of inhabitants was determined by point and probabilistic estimations. Overall, the study has provided scientific evidence for future studies, as well as for contaminant monitoring, risk management and policymaking.

Our study determined that there was no increased health risk among adult residents in Shanghai, China, considering recent total cadmium exposure. Because studies have shown that primary foodstuffs, such as rice in Asian countries and tubers in Western countries, may be the main source of dietary cadmium exposure, controlling cadmium levels in these foods may be a reasonable way to control cadmium intake for non-occupational exposure. A good correlation between internal and external cadmium exposure demonstrated that combining food survey and food contamination detection could provide a reliable estimation for environmental contamination exposure. Moreover, the data in this study demonstrated that tobacco cadmium exposure was as important as dietary cadmium exposure in smokers. Therefore, increased concern regarding tobacco processing and smoking cessation could also reduce cadmium exposure. Further studies could be designed to include other susceptible populations, such as children and pregnant women, considering the additive effect of other contaminants.