Concentration of cadmium and lead in vegetables and fruits

Chemical contamination of foods pose a significant risk to consumers. A source of this risk is due to the consumption of products contaminated with heavy metals such as cadmium (Cd) and lead (Pb). The aim of the study was to research the levels of Cd and Pb contamination of selected species of vegetables and fruits in the form of fresh, frozen, dried and processed products. The goal was to verify which of these food groups was more contaminated with heavy metals. The study covered 370 samples of fruits and vegetables including apples, pears, grapes, raspberries, strawberries, cranberries, as well as beetroots, celeries, carrots and tomatoes. The content of Cd and Pb was determined by atomic absorption spectrometry. Quantitative results were analyzed using statistical models: analysis of variance, outlier analysis, post-hoc multiple comparison Tukey test. The tests showed that the levels of Cd and Pb concentration in samples of fresh, processed, frozen and dried fruits and vegetables varied substantially. The highest concentrations were recorded in dried products. Several fruit and vegetable samples exceeded the maximum permissible concentrations of Cd and Pb. The contamination of these products could be a significant source of consumer exposure to heavy metals when these products are a part of the diet.

Food quality and safety are perhaps the most important public health issues. Food available on the markets should be free of all chemical contaminants which pose a risk to consumer health, and its safety is not only the responsibility of food producers, but also state governments and agencies that systematically monitor and control food quality 1 . In Poland, this function is performed by the State Sanitary Inspection, which supervises food quality 2 .
A significant risk to the health of potential consumers is food contaminated with heavy metals, such as cadmium (Cd) and lead (Pb), exceeding the maximum permissible limits for food products.  3,4 . Heavy metals can be the cause of many chronic diseases whose symptoms are different depending on the level of toxicity of an element, as well as the duration and level of exposure 5,6 . Kidneys and liver are the main organs especially sensitive to Cd toxicity 7 . In the human body, Cd most often causes damage to both of these organs, as well as the testicles, lungs and bones. In addition, it causes a carcinogenic effect, initiating cancers of the prostate, kidneys, pancreas and testicles 5,8 . This element negatively affects the function of the skeletal system by disturbing the metabolism of calcium, magnesium, zinc, copper and iron ions 5,6 . In turn, Pb is a neurotoxic element. In the general public, but specially in children, elevated levels of Pb in the blood may cause changes in the brain, manifested by: lowering of the IQ level, a problem with proper perception and concentration and a hyperactivity 8,9 . Chronic exposure to Pb can be associated with an increased risk of developing neurodegenerative diseases 10 . Moreover, it has been demonstrated that Pb could have a role in the pathogenesis of deep vein thrombosis of lower limbs 11 .
In the case of vegetables and fruits, the source of their contamination with heavy metals may be the environmental conditions in which the cultivation was carried out 12 . The food product contamination with heavy metals may have also resulted from the migration of these elements from the packaging material. Contamination may have occurred during the technological processes that prepare the products for consumption, for example as the result of using metal kitchen tools 13,14 . In the case of foodstuffs stored in metal cans, the packaging coating may corrode, especially when stored foods have an acidic pH. Corroded metal packaging can become a source of migration of heavy metals, such as tin (Sn), Cd and Pb, to the stored product, that increase their content in food [15][16][17] . A human risk assessment can identify methods to minimize exposure to heavy metals, for example by reducing the weekly consumption of contaminated food products 18 .

Material and methods
The study material consisted of 370 samples containing 6 species of fruits: apples (Malus sylvestris (L.) Mill. var. domestica (Borkh.) Mansf.), pears (Pyrus communis L.), grapes (Vitis vinifera L.), raspberries (Rubus idaeus L.), strawberries (Fragaria × ananassa Duchesne), cranberries (Vaccinium macrocarpon Aiton) and 4 species of vegetables: beetroots (Beta vulgaris L. var. esculenta), celeries (Apium graveolens L. var. rapaceum), carrots (Daucus carota L. var. sativus) and tomatoes (Solanum lycopersicum L.). Almost all species of fruits and vegetables were available on the Polish markets in the form of fresh, frozen, dried or processed. Processed fruits and vegetables were in form of 100% juice, compote, jam, marmalade or stored in syrup or marinade. The exceptions were frozen fruits which were not available for sale in Poland: apples, cranberries, grapes, and pears. The most frequent group was fresh food and in the following order: processed, dried and frozen ( Table 1) Analytical methods. Fresh vegetables and fruits samples were prepared for chemical analysis in the same way that they are prepared for consumption (thorough washing, peeling, removing inedible parts). The cleared samples were shredded by a homogenizer T18 Digital Ultra-Turrax, IKA (Germany). The frozen samples were homogenized. The dried samples were grinded in a vibratory grinder Testchem PZS 01 (Poland). The processed products stored in syrup or marinade were separated and homogenized. The samples in the form of 100% juice, compote, jam and marmalade were thoroughly mixed. In the next phase, the samples were weighed using an analytical balance AS60/220/C/2, Radwag (Poland). The weight of each sample was varied and amounted to 2 g for fresh and frozen samples, 1 g for dried samples, 2 g for processed samples, except liquid samples-10 g. 8 mL of nitric acid (HNO 3 , 65%, Suprapur-Merck, Germany) was added to all food samples. Dried samples were Samples were subjected to a four-phases mineralization process in microwave mineralizer Magnum II, Ertec (Poland) ( Table 2). Cd and Pb concentrations were determined by atomic absorption spectrometry (AAS)-SavantAA Sigma with PAL3000 automatic sample feeder and graphite furnace GF3000 (GBC, Australia). Measurements were made with atomization in graphite furnace and background correction; instrumental parameters and measuring range of the spectrometer AAS are summarized in Table 3. To estimate the limit of detection (LOD) and the limit of quantification (LOQ), at least 20 blank samples should be analyzed; subsequently mean value and standard deviation (SD) were calculated for these measures. LOD is the lowest quantity or concentration of a component that can be reliably detected with a given analytical method. It is estimated by: LOQ is the lowest analyte concentration that can be quantitatively detected with a stated accuracy and precision. Its estimate is given by: LOD and LOQ values for Cd and Pb were summarized in Table 4. Standard solutions were used to create the calibration curve Certificate of Reference Material 1000 mg l -1 Lead Matrix: 2% HNO 3 SPEX CertiPrep and Certificate of Reference Material 1000 mg l -1 Cadmium Matrix: 2% HNO 3 SPEX CertiPrep standard solutions. Certified reference material (CRM)-Vegetable puree (TYG006RM) was used to confirm the correctness of analytical measurements in edible plant samples (Table 4).

Statistical methods.
Analysis of variance (ANOVA) was utilized to detect significant differences between average Cd and Pb concentrations in fresh, frozen, dried and processed vegetables and fruits. A p-value of less than 0.05 was used to indicate statistical significance. In the next stage, Outlier Analysis was performed. Boxplots were utilized to pictorially represent the location, dispersion and shape of the data distribution regarding the concentration of Cd and Pb in all types of analyzed food. Outliers show those values that largely differ from other values in the sample. They are typically either much higher or much lower than other points. Using the Post-hoc Multiple Comparison, attempts were made to determine significant differences between the mean values in the analyzed groups. Applying the Tukey's HSD (Honestly Significant Difference) test, all pairs were compared (fresh-dried, frozen-dried, processed-dried, frozen-fresh, processed-fresh, processed-frozen) in four variants: Cd concentration in fruits, Cd concentration in vegetables, Pb concentration in fruits and Pb concentration in vegetables. The significance level was adjusted for multiple tests using Bonferroni correction 22 .    3,4 . It was found that in 12 food samples, the Cd content exceeded the maximum acceptable level. Among the fruit samples, this result was observed in: frozen raspberries (n = 1; 122% of maximum level) and frozen strawberries (n = 1; 114% of maximum level). In the case of vegetables, this result was observed in: fresh beetroots (n = 2; 203% and 670% of maximum level), frozen carrot (n = 1; 113% of maximum level), fresh celery (n = 4; 130%, 150%, 345%, 356% of maximum level) and processed tomatoes (n = 3; 102%, 112%, 134% of maximum level). The maximum permissible Pb level was exceeded in 3 analyzed food samples: fresh beetroot (n = 1; 135% of maximum level), frozen carrot (n = 1; 117% of maximum level) and 1 sample of frozen tomatoes in which the Pb concentration was up to 1074% of the acceptable limit (Table 5). Tables 6 and 7 present the mean and SD, as well as the minimum and maximum values for the Cd and Pb contents in each of the analyzed fruits (Table 6) and vegetables (Table 7). Heavy metals concentrations were reported in mg/kg f.m. (fresh mass) in the fresh, frozen and processed products, while the content of Cd and Pb in dried products were presented in mg/kg d.w. (dry weight). Lack of a value in the tables means that the Cd or Pb value was below the LOQ for that particular sample.
The analysis of Cd and Pb contents in all food products is necessary due to the possibility of assessing the health risks associated with consumption of contaminated ready-to-eat different types of food. A review of the scientific literature showed that the issue of food contamination with heavy metals is discussed by several researchers. However, they mostly include only fresh fruits and vegetables. Additionally, there is a little data concerning the level of heavy metals contamination of vegetables and fruits cultivated in other European countries in the available literature. Consequently, the results presented in this paper may form the basis for further research on the scale of food contamination with heavy metals such as Pb and Cd.
Among fruits such as apples, pears, raspberries and strawberries, the highest average values of both Cd and Pb were observed in dried products (Cd: 0.023, 0.015, 0.116, 0.131 mg/kg d.w., respectively; Pb: 0.127, 0.036, 0.111, 0.161 mg/kg d.w., respectively). In cranberry samples, the highest levels of Cd were determined in fresh fruits (0.008 mg/kg f.m.), while Pb-in processed products (0.01 mg/kg f.m.). In the case of grape samples, the same average Cd concentration was recorded in both dried and fresh products (0.001 mg/kg), while the highest Pb content was observed in processed products (0.07 mg/kg f.m.). In most fruit samples the lowest average Cd concentrations were determined in processed products (grapes, pears, raspberries and strawberries-0.0004, 0.0008, 0.009, 0.003 mg/kg f.m., respectively), while Pb-in fresh fruits (cranberries, grapes, pears-0.004, 0.005, 0.008 mg/kg f.m.) or processed (raspberries and strawberries-0.011 and 0.006 mg/kg f.m.). In apple samples, the same average Pb value was recorded in both fresh fruit and processed products (0.009 mg/kg f.m.).
The     The high contamination found in vegetables might be closely related to the pollutants in irrigation water, farm soil, fertilizers and also industrial and low pollution household emissions. Differences in levels of contamination between fruits and vegetables may result from the specificity of the geographical area from which they are collected, their diverse capacity to accumulate heavy metals, as well as the way they are processed. It should be pointed out that in polluted environments (soil, water, and air), the presence of toxic metals in elevated concentrations is not uncommon. Due to the structure of consumption of various groups of food products both in Poland and other countries, a significant risk of exposure to heavy metals is associated with the consumption of fruits and vegetables, which are one of the main elements of the diet. Unfortunately, complete elimination of elements such as Cd or Pb from these products is impossible, and the technological processes used in food production can only remove a small part of the impurities from selected products or even contribute to their increased contamination. Thus, there is a need for regular monitoring of heavy metals on every kind of foodstuff, not only in fresh products, in order to estimate the health risk from heavy metals in the human food chain.
Statistical analysis. ANOVA. For the purpose of ANOVA carried out to detect significant differences in the heavy metal concentrations of the four types of food (fresh, dried, frozen, and processed), samples with concentration value below the LOQ were removed from the analysis. In the case of Cd concentration, the value of F statistic was 11.15 for fruits and 4.049 for vegetables, leading to significant results with p-values below 0.001 and 0.01 respectively. For the of Pb concentration, the ANOVA results were even more extreme with F values of 56.59 for fruits and 7.13 for vegetables with associated p-values being below 0.001 in both cases. These results show that there is strong evidence to believe that mean Cd and Pb contents in the four types of fruits and vegetables are not equal (Table 8).
Outlier analysis. The boxplots depicted in Fig. 1 were used to illustrate the outlier analysis for Cd and Pb. Each plot shows the median of the observations along with the lower quartile (Q1) and the upper quartile (Q3). The highest and the lowest observations are shown by the whiskers. From Fig. 1a, there appears to be two outliers in the dried fruits with values 0.277 and 0.210. From Fig. 1b, there seems to be six outliers in the fresh vegetables with values of 0.203, 0.670, 0.260, 0.690, 0.300 and 0.712. In Fig. 1c, we see two outliers in the processed fruits with values of 0.127 and 0.047. Finally, Fig. 1d shows that there is one one outlier in the frozen vegetable category with the value of 0.537.
Outliers associated with high Cd and Pb values in fruit and vegetable samples may be the result of sample contamination during technological processes or vegetables/fruits cultivation in a polluted agricultural area.   Fig. 2. For the Cd concentration, comparison of all pairs of means indicated that the content of Cd in dried fruits is significantly different from mean concentrations of other types of food namely fresh, frozen, and processed fruits, see Fig. 2a. In the case of vegetables, the mean Cd contents of fresh and processed vegetables are different, see Fig. 2b, although mean Cd content of frozen and fresh vegetables are also significantly different if a significance level of 10% is used. Upon analyzing the mean concentrations of Pb in fruits, we found that the mean content of dried fruits was significantly different from the other three types, namely fresh, frozen and processed, see Fig. 2c. For the Pb concentrations in vegetables, a highly significant difference was detected between the means of processed and dried vegetables. In addition, mean Pb concentrations of fresh versus dried and processed versus frozen vegetables were significantly different, see Fig. 2d.

Conclusions
The research showed that the values of Cd and Pb in all types of tested fruit and vegetable samples: fresh as well as processed, frozen and dried, are very diverse. The highest levels were noted in dried products. It may be the result of the elimination of water which increases the concentration of dry matter content. In addition, heavy metal contamination of dried fruits and vegetables may be the result of technological processes used in food production. The maximum allowable concentration of the toxic metals was exceeded in several samples: Cd in 2 fruit and 10 vegetable samples, and Pb in 3 vegetable samples. Despite the fact that the exceedance of the Cd limit values concerned only 1% of the analyzed fruits and 9.3% of vegetables, and Pb-only 2.8% of all tested vegetables, the contamination of these groups of food products can be a significant source of consumer exposure to heavy metals, because they are an important part of the diet of most people.
The State Sanitary Inspectorate should thoroughly control those products in which the Cd and Pb limit values have been exceeded. It should be stated whether the exceedances occurred in only one batch of those products or if a larger number of them was contaminated. These products should be immediately withdrawn from sale and their producer should be obliged to indicate the source of their contamination.

Data availability
The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.