Quantification of heavy metals and health risk assessment in processed fruits’ products

2.1 Sampling

Based on availability, processed fruits-based products (n = 345) of different national and international brands viz. National, Shan, Shangrilla, Mitchell’s, Fruit Tree and Nestle were collected from the local markets of Abbottabad, Haripur and Mansehra districts of Khyber Pakhtunkhwa. Pakistan. These samples were collected in triplicates and categorized in to different groups including “juices”, “jam”, “canned fruits”, “sauces”, “ketchup”, “pickles”, marmalades and tomato paste (102, 81, 39, 33, 33, 33, 15 and 9 samples, respectively) on the basis of product types.

2.2 Digestion and instrumentation

Processed fruit product samples were digested following the protocol as explained earlier by Parveen et al. (2020). Briefly, powder sample (∼2g) was added into digestion tubes containing digestion mixture (HNO₃ and HClO₄) at 2:1. The mixture was heated on digester Pelican Kelplus- KES 20LRAL-TS at 120 °C until clear solution was obtained. Afterwards, final volume was adjusted up to 50 mL with deionized water. Along with a batch of 6 samples a blank was also prepared in the same way. Quantification of selected metals: Cd, Cr, Co, Cu, Fe, Pb and Zn was done using the flame atomic absorption spectrometer (Perkin Elmer S#80156060702) under optimum analytical conditions. Working standards for tested metals were prepared from stock solutions (1000 mg/L). Optimum analytical conditions maintained on atomic absorption spectrophotometer ‘AAS’ to selected metals using air-acetylene flame (Perkin Elmer S#80156060702) are mentioned in Table S1.

2.3 Statistical analysis

Analytical data were evaluated by univeriate and multivariate approaches using STATISTICA (StatSoft Inc, 1999) and SPSS Statistics V21.0 (IBM, Chicago USA). Cluster analysis (CA) and principle component analysis (PCA) were performed using Ward’s and Varimex rotation methods to find compositional pattern of data set and sources of contamination. Graphical representations of data were done by using GraphPad Prism 8.0 (GraphPad Software, Inc. USA) and Sigma Plot 12.1 (Systat Software Inc. San Jose, California). Data were presented as mean ± SD for triplicate analysis.

2.4 Health risk assessment

Filed survey was conducted to document the information regarding the consumption of processed fruit-based products among the local communities of three districts: Haripur, Abbottabad and Mansehra of KP province of Pakistan. In total, 100 informants were participated in the field survey (Table S2). Data were collected using semi structured interviews and questionnaire comprising multiple choice questions. Information were gathered on family members, types of product consumed, and quantity consumed etc. (Table S3).

The health risk was calculated by finding the daily intake (mg/day/person) of selected metals through consumption of the processed fruit-based products and then relate them to the reference oral dose (Singh et al., 2010).

Estimated daily intake (EDI) was calculated following the procedure reported previously (Cherfi et al., 2014) using equation: $$\mathit{EDI}=C\times I$$ where; ‘C’ represents the metal concentration in fruit products (mg/kg), and ‘I’ denotes daily intake of the product.

Health risk index (HRI) is a proportion between the estimated daily intake (EDI) of the metal and reference oral dose (RfD) for each metal and body weight (BW) of the consumers (Cui et al., 2004) and the HRI less than 1 for any metal consider safe for consumer (Parveen et al., 2020). The HRI was calculated using following equation: $$\mathit{HRI}=\sum \frac{n\left(Cn\times Dn\right)}{\mathit{RfD}\times Bw}$$ where Cn is the mean metal concentration in specific fruit product on fresh weight basis (mg/kg); Dn is the average daily intake rate of a specific fruit product in a whole year; RfD showed safe level of exposure by oral intake for lifetime; and Bw is the average body weight.

Target hazard quotient (THQ) and Hazard index (HI) are used to evaluate the non-carcinogenic health risk to human (Yang et al., 2011). THQ was calculated using equation reported previously (USEPA, 2015). $$\mathit{THQ}=\frac{C\times I\times {10}^{-3}\times EFr\times EDtot}{\mathit{RfD}\times Bw\times ATn}$$ where C represents mean metals level in fruit product (mg/kg); I is the ingestion rate (g/day/person); Efr is the exposure frequency (days/year); EDtot is the total exposure duration (years); Bw is the average body weight adult (kg); and ATn is the averaging time, non-carcinogens (EDtot × 365 days/year).

Hazard index (HI) is the sum of hazard quotients for trace metals and was calculated by formula as reported by earlier (USEPA, 2006; USEPA, 2015). $$\text{HI}={\text{THQ}}_1+{\text{THQ}}_2+\dots +\text{THQ}$$ where THQ1 − n is Target hazard quotients for 1 − n metals.

Target cancer risk (TCR) is intended to determine the cancer risk to consumers due to food consumption. The TCR in processed fruit products was calculated following the method reported previously (USEPA, 2006; USEPA, 2015) $$\mathit{TCR}=\frac{\mathit{Cb}\times 1\times {10}^{-3}\times CPS\times \times EFr\times EDtot}{\mathit{BWa}\times ATc}$$ where ATc is averaging time carcinogens; carcinogens potency slope oral (μg/g/day) is CPS; Efr is the exposure frequency (days/year); EDtot is the total exposure duration (years); BWa is the average body weight; and Cb is metal concentration.

3.1 Comparative assessment of HMs in fruits’ products

On the whole Fe was dominating in term of highest concentration in almost all brands of processed fruit-based products followed by Cd, Cr and Pb. Whereas, Co and Cu were usually at lowest levels (Table 1). Likewise, in many samples measured levels of HMs were below the detection limit.

As mentioned in Fig. 1a, in different brands of sauces Fe, Cd and Zn were highest in concentration (50.17 ± 15.1, 22.94 ± 6.92 and 4.100 ± 1.24 mg/kg, respectively). Relatively average concentration of Cd, Cu, Fe, Pb and Zn, in our samples were higher than previous reports from Bahrain (Musaiger, 2008) and Bangladesh (Haque et al., 2018). In ketchup samples, Fe, Cr and Pb were dominating in term of mean concentration levels (10.43 ± 3.5.725 ± 1.72 and 4.438 ± 1.33 mg/kg, respectively). Measured levels of all metals in ketchup were more than reported in tomato sauce from Nigerian (Adegbola, 2013) and in different brands of ketchup consumed by the inhabitants of Ghana (Boadi et al., 2012), Nigeria (Iwegbue et al., 2012) and Romania (David et al., 2008). In addition, average concentration of Pb (4.438 ± 1.33 mg/kg), was also higher than the recommended standard of Codex (2016) for ketchup (1.0 mg/kg). However, Fe concentration in our samples was relatively lower than reported earlier by Harmanescu et al., (2007) and David et al., (2008) from Romania.

Close

Fig. 1

Comparative assessment of heavy metals in processed fruits’ products.

Export to PPT

In juice samples of 34 brands average levels of HMs concentration in different brands of juice was in subsequent order: Fe ˃ Pb ˃ Co ˃ Cr ˃ Zn ˃ Cd ˃ Cu. Comparatively, lower mean concentrations of Fe, Zn, Cu, Co, and Cr were reported in the fruit juice from Saudi Arabia (Farid and Enani, 2010), but were higher than the fruit juice sold in the markets of Jordan (Massadeh and Al-massaedh, 2018), Pakistan (Anwar et al., 2014), Iraq (Al-Mayaly, 2013), Poland (Szymczycha-Madeja and Welna, 2013), Turkey (Hamurcu et al., 2010) and Italy (Coco, 2006). Threshold levels of Cd, Cu, Pb and Zn in fruit juice are 0.005, 0.01, 0.01 and 0.2 mg/kg, respectively (WHO, 1984). Conversely, measured levels of these metals in our case were far beyond the permissible limits, therefore may cause severe threat to consumers’ health. There was no significant difference in Fe and Cd levels in different brands of Jam (Table 1). Relatively, measured levels of Pb, Cd and Cr in our samples were significantly higher than fruit jam consumed in the markets of India and Rwanda (Asema and Parveen, 2018; Poornima et al., 2014; Mukantwali et al., 2014). Estimated levels of Cu, Fe, Pb, Zn and Cd levels in pineapple jam sold in the local markets of Rwanda were 3.29, 2.95, 0.77, 3.55, 0.01 mg/kg, respectively ().

The descending order of HMs quantified in various brands of canned fruits was: Fe ˃ Zn ˃ Pb ˃ Co ˃ Cd ˃ Cr ˃ Cu. Fe was in highest concentration, followed by Zn and Pb (88.64 ± 23.1; 14.73 ± 3.93 and 11.02 ± 3.05 mg/kg, respectively). Source of raw material, mode of transportation, storage and packaging, pH, concentration of oxygen in space between product and can head and quality of coating material used (Iwegbue et al., 2009), and industrial processes i.e. kneading, cutting, rolling, sheeting, and chopping (Jothi and Uddin, 2014), contribute significantly in HMs contamination in canned fruits. Comparatively, measured values of HMs in our samples were higher than reported previously in canned fruits (Divis et al., 2017; Leao et al., 2016; Rafique et al., 2008). Furthermore, average Cd level in our samples was considerably higher than European Union standard “0.05 mg/kg” (EU, 2006b), which is alarming and may leads to serious adverse health effects.

In tomato paste, mean value of Fe was highest (22.06 ± 12.7 mg/kg), followed by Pb and Zn (8.603 ± 4.96 and 6.122 ± 3.53 mg/kg, respectively). Relatively, all HMs in tomato paste were higher than reported from Nigeria (Adegbola, 2013; Hadiani et al., 2014), Iran (Hadiani et al, 2014), Turkey (Kocak et al., 2005), and Romania (David et al., 2008)). Furthermore, mean values of Pb and Cd in tomato paste were above the threshold levels of these metals viz. 1.0 and 0.05 mg/kg, respectively (Codex, 2016). To best of our knowledge, HMs metals have rarely been investigated in marmalades. Present study revealed that Fe, Cd and Zn metals were highest (9.058 ± 4.96, 3.986 ± 1.78 and 3.856 ± 1.72 mg/kg, respectively) in different brands of marmalades, while Co was lowest in concentration (0.406 ± 0.181 mg/kg). In 11 different brands of pickles, Fe was highest with an average of (143.3 ± 43.2 mg/kg), followed by Cr and Pb at 12.97 ± 3.91 and 12.53 ± 3.77 mg/kg. Comparatively, measured levels HMs in our samples were higher than reported previously from Behrain and Turkey (Musaiger, 2008; Tuzen, 2007).

Fig. 1b, revealed that Fe concentration was reasonably higher in all brands of fruits’ produced, followed by Cd, Zn, Cr, Pb, Co and Cu. Cd metal ranged from 22.94 ± 6.91 to 2.193 ± 0.376 mg/kg in different brands of sauces and fruit juice, respectively. This indicates that even the lowest level of Cd in our samples was significantly higher than the permissible limit viz. 0.05 mg/kg (Codex, 2016), that may lead to serious health issues in consumers specifically associated with this metal (Baldwin and Marshall 1999). Municipal sewage, chemical industries, Ni-Cd batteries, metallic alloys, tobacco smoking, smelting, plating, and non-recycled Cd products contributing significantly in increasing levels of Cd in soil, water and food chain (Divis et al., 2017; Poornima et al., 2014; Chakraborty et al., 2013; Tiimub and Afua, 2013; ATSDR, 1999). Comparatively, Cr was highest in different brands of fruit juice, jams and ketchup (12.97 ± 3.93, 5.792 ± 1.72, 5.327 ± 1.47 mg/kg, respectively). And these values were relatively higher than the recommended level of Cr in canned or processed foods is 0.4 mg/kg (FAO/WHO, 2003). However, in other products i.e. canned fruits, sauces, tomato paste, marmalades and pickles measured levels of Cr were within safe limit.

Average concentration of Co was highest in pickles, followed by canned fruits, jams, juices, marmalades, ketchup, sauces, and tomato paste. Measured levels of Co in fruit juice were relatively higher than reported previously from Saudi Arabia (Farid and Enani, 2010) and Ghana (Ackah et al., 2014). In processed foods, Co concentration should not be more than 0.05 mg/kg (EU, 2006a), but in except tomato paste this metal was significantly higher than threshold level, which may impose adverse health effects on consumers Use of copper vessels in jams ensures equal heat distribution, lowers cooking time, and preserves fruit colors. In studied samples, Cu concentration was ranged 0.623 ± 0.18 mg/kg (sauces) to 3.907 ± 2.23 mg/kg (tomato paste). Permissible limit for Cu in fruits’ juice is 5 mg/kg, whereas in tomato ketchup, sauces and tomato paste is 50 mg/kg (FAO/WHO, 2003). Therefore, in our samples Cu concentration lies within permissible limits.

Maximum permissible limit for Fe in processed foods is 50 mg/kg (Health Ministry, 1988). Pickles, canned fruits and sauces were among the top three groups having the highest levels of Fe at 143.3 ± 43.2, 83.64 ± 23.1 and 50.16 ± 15.1 mg/kg, respectively. Extensive use of Fe in machinery or steel containers in food processing industries, use of iron rich spices in sauces and pickles may increase its concentration processed foods. Likewise, presence of acids in different fruits may cause leeching of iron, particularly if such fruits are stored or packed in iron or steel containers (Ogidi et al., 2017). Pb concentration was significantly higher in pickle samples, canned fruits and tomato paste (12.53 ± 3.77, 11.02 ± 3.05 and 8.602 ± 4.96 mg/kg, respectively) compared to its permissible limits viz. 1 mg/kg for canned fruits, jams and preserved tomatoes and 0.03 mg/kg for juices (Codex, 2016). Pb concentration in our samples was in agreement with previous reports from Bangladesh (Tasnim et al., 2010) and Iran (Fathabada et al., 2018b). High concentration of Pb in canned fruits and paste may be attributed to the soldering process (Divis et al., 2017), while Pb deposition in soil and atmosphere, use of pesticides and fertilizers, harvesting techniques, storage conditions, transportation, processing machinery, water pipes and containers, gasoline and burning paints may involve in Pb contamination in processed foods (Jothi and Uddin, 2014). It has been reported that about 90% daily exposure to Pb in human comes from the food (Krejpcio et al., 2005), which is lethal to human and causes various types of cancer (Shariatifar et al., 2017). Average concentration of Zn in all products ranged 2.771 ± 0.47 (in juices) to 14.17 ± 3.93 mg/kg (in canned fruits). Measured level of Zn was much higher in canned fruits than permissible limit that is 5 mg/kg (FAO/WHO, 2003). Maximum level of zinc in canned fruit might be due to its application in metal alloys used in making cans.

3.2 Correlations

Interrelationship among the measured levels of mean HMs concentration were investigated in terms of correlation varimax. Results of Pearson’s correlation coefficient analysis of HMs’ concentration in processed fruit-based products are mentioned in Table 2. Except Cd, all metals showed associations with each other. This might be due the fact that Cd has different origin than rest of the metals quantified in our samples. Highly significant positive associations were observed between Co-Pb and Cr-Fe at r = 0.956 and 0.843 (p < 0.01), respectively. Whereas, Co-Zn, Pb-Zn, Fe-Zn and Fe-Pb also depicted strong positive relationships with r = 0.841, r = 0.813, r = 0.781 and r = 0.746, respectively (p < 0.05). Such a strong association between these metals suggests their common sources of contamination in processed foods such as various activities involve in food processing industries. However, inverse relationships showed depletion or enrichment of specific elements at the cost of others as seen in case of negative correlations of Cd with other metals (Table 2). Furthermore, negative associations suggest that Cd content are not controlled by single source factor and may be synergistic effect of both natural and anthropogenic activities (Suresh et al., 2012). The quartile distribution of HMs in different categories of processed fruit products was assessed as shown in box and whisker plot (Fig. 2). All targeted metals demonstrated broad range spread over various orders of magnitude. Cd, Cr, Pd, Zn and Fe showed broad and symmetrical distribution, whereas Co and Cu depicted broad ranged asymmetrical spread in the quartile values. Narrow range distribution was observed in the case of Mg, Zn, Cu, and Cd.

Close

Fig. 2

Quartile distribution of heavy metals in studied samples.

Export to PPT

3.3 Multivariate analysis

Multivariate analysis was performed using cluster analysis (CA) and principal component analysis (PCA) to identify the sources of heavy metals in selected samples. BioVinci software was used for hierarchal cluster analysis of HMs in processed fruit samples (Fig. 3), depicted three main clusters. Based on average distribution pattern of HMs, two main clusters were identified for heavy metals. Among HMs, Cd-Zn were placed in first cluster, whereas Fe, Cr, Pb, Co and Cu were grouped in second cluster. The second cluster comprises two subgroups viz. Fe-Cr and Pb-Co, while Cu was in separate group. Likewise, based on distribution of HMs, different categories of processed fruit-based products were also in two main clusters, which were composed of sub-clusters. In first cluster, pickles and canned fruits were closely associated, accompany by tomato paste. Marmalades, ketchup, juice and jams were grouped in second cluster, whereby ketchup and juice were closely linked along with marmalades and jams. However, sauces were placed in separate group. Hierarchal cluster analysis specifies close association of HMs metals, which might be due to similarities in distribution pattern and source of contamination in the studied samples. Though, Cd is a carcinogenic metal, but was bunched with Zn that is an essential metal and within the permissible limit.

Close

Fig. 3

Hierarchal clustering of heavy metals in processed fruits' products.

Export to PPT

Principle component analysis (PCA) is extensively used to estimate the contribution of natural and anthropogenic sources of HMs contamination in water, soil, air and plant samples (Liu et al., 2015). Mean values of HMs in processed fruits’ products were treated by PCA using Varimax rotation with Kaiser Normalization to find out their compositional pattern and sources of contamination. Two principle components viz. PCA1 and PCA2 were extracted based on eigenvalue (˃1) and accumulated variance explained 81.88% of the total variations (Table 3). First component contributes 57.96% variation of the total variation with maximum loading of four metals (Pb, Fe, Cr and Co), with percentage variance of 0.937, 0.929, 0.896 and 0.877, respectively. The contribution rate of these metals in PC1 revealed that main sources of contamination of these metals are related to each other. Such as natural phenomenon including mineral weathering, biogenic and forest fires, erosion and volcanic activities, and anthropogenic activities like industrial processing, use of chemical fertilizers and pesticides, municipal water and mining contributing considerably Co, Cr, Fe and Pb contamination in food chain and processed foods as well (Singh et al., 2018). The contribution of second component was 23.92% of the cumulative variation and had maximum loading of Zn (0.836%) and Cd (0.811%). This indicates that Zn and Cd had common origin in processed fruits’ products. Fig. 4a, indicates that Pb, Fe, Cr and Co metals shares come sources of contamination in processed fruits’ products shares, while Cd and Zn come from other sources. In Fig. 4b, solid symbols represent the mean concentration of HMs in each food type. The position of black circles in a plot relative to the direction of green lines approximates correlations between food and the gradient of element concentrations. The lengths of green lines indicate the overall contribution of each food to the analysis. The directions of the green lines indicate the correlation with each axis i.e. vector lines parallel to an axis are highly correlated with that axis, while angles between the vector lines show correlations between food types.

Close

Fig. 4

Principle component analysis of heavy metals' concentration in samples analyzed.

Export to PPT

3.4 Health risk assessment

Health risk assessment associated with HMs’ contamination in processed fruits’ products was calculated by estimating health risk index (HRI), target hazard quotient (THQ), hazard index (HI) and target cancer risk (TCR). Heat map of HRI associated with Fe, Zn, Cu, Co, Cr, Cd and Pd concentration in 345 samples of processed fruits’ products categorized into eight different groups was generated using GraphPad Prism version 8.0 (Fig. 5a&b). For each group, HMs concentration was presented in each rectangle of the heat map, whereby red colors represent high metal concentrations and blue color indicates low concentrations. Measured levels of HRI for Co, Cu, Fe and Zn (Fig. 5a), in all categories of processed fruits’ products were within the safe limit (<1.0). However, HRI values for Cd, Cr and Pb were greater than unity in all samples and could be health risk to consumers (Parveen et al., 2020). The HRI values for Cd were higher in almost all groups of fruits’ products, particularly in sauces, jam, and canned fruit. Likewise, HRI associated with Pb except sauces and marmalades and Cr except sauces, jams, tomato paste and marmalades was more than unity. Health protection standard of lifetime risk for THQ and HI is 1.0 (USEPA, 2006).

Close

Fig. 5

HRI, THQ and HI associated with HMs contamination in processed fruits’ products.

Export to PPT

In the present study, THQ and HI values of HMs concentrations in all different groups of processed fruits’ products (Fig. 5b&c) were within the safe limit (<1.0). Consequently, our findings demonstrated that ingestion of these products pertaining to these metals is safe and will not cause non-carcinogenic risk to consumers. The THQ and HI values of Cd, Cr and Pb were relatively higher than rest of the studied metals, therefore further assessed for TCR. As demonstrated by USEPA (2006), acceptable cancer risk limit for carcinogenic metals i.e. Cd, Cr and Pb is ranged 1 × 10⁻⁴ to 1 × 10⁻⁶. Among, the studied samples TCR values for Cd, Cr and Pb were found within the acceptable limit (Fig. 6). However, TCR values of Cd was alarming in sauces, jams and canned fruits and may cause cancer to consumers over long-time exposure. Likewise, Cr and Pb concentration was alarming in pickles and canned fruits.

Close

Fig. 6

Target cancer risk associated with heavy metals' concentration in processed fruits.

Export to PPT