Introduction

The persistence of heavy metals in the environment and the associated human health have become an environmental issue that has attracted much attention in recent years (Cai et al. 2012; Lei et al. 2013; Tang et al. 2010; Wu et al. 2016). Anthropogenic activities are a particularly important source of heavy metals, which can cause low-level poisoning through bioaccumulation in plants and animals (Duzgoren-Aydin 2007; Shen et al. 2020). As the process of urbanization continues to accelerate, soil pollution and the destruction of water habitats continue to occur (Islam et al. 2015; Xiao et al. 2015).

When heavy metals enter the human body via ingestion, inhalation, or dermal absorption, they may accumulate in certain organs and cause adverse effects (Sarvestani and Aghasi 2019). For example, arsenic (As), listed with supporting evidence as a group 1 carcinogen by the International Agency for Research on Cancer, can induce cancer and affect reproductive development (Abdul et al. 2015). Cadmium (Cd) may cause cardiovascular disease and osteoporosis (White et al. 2018). Lead (Pb) may interfere with physiological functions and have various adverse effects on the respiratory system (Boskabady et al. 2018).

Selenium (Se) has attracted attention due to its antioxidant and anticancer properties (Hatfield et al. 2014; Rayman 2012). The low-Se areas in China are mainly located in the north-eastern provinces, southwest regions and Yunnan-Guizhou Plateau, etc. It is well known that Se deficiency can cause human cancer, but excessive intake is also harmful (BaJaj et al. 2011). Seafood has been confirmed as rich in Se (Wang et al. 2021). Therefore, it is necessary to conduct research on whether the Se intake of people in the coastal city of Guangzhou is too high.

Guangzhou, a coastal mega city with a population of 18 million, is the fastest developing area of economy and urbanization in China. As one of the core areas driving the development of the Guangdong-Hong Kong-Macao Greater Bay Area, it is also one of the most important industrial centers in China. The main industries include electronic communication equipment, automobile manufacturing, and petrochemical industries. Heavy metals from automobile exhausts, incinerators, industrial waste, atmospheric deposition, and other activities continue to increase the burden of pollutants in the environment (Awasthi et al. 2018; Chen et al. 2010). Currently, heavy metals such as Cd, Pb, and copper (Cu) are polluting in the city significantly. In particular, the concentration of Pb in soils, according to the soil quality standards in China (GB15618-1995), has even reached the level of serious pollution (Gu et al. 2016; Lu et al. 2011). Increased levels of heavy metals in urban soils may pose health risks to the population.

In urban areas such as the city, the presence of excessive heavy metals is directly related to the health of tens of millions of people. The present study aimed to investigate the levels of nine heavy metals in the urine of the general population and estimate the human non-carcinogenic and carcinogenic health risks. Factors influencing the element concentrations, including region, gender, lifestyle habits, and environment, were studied, to provide useful information for predicting the levels of heavy metals in the human body.

Materials and methods

Materials and reagents

Nine heavy metal standards, chromium (Cr), manganese (Mn), nickel (Ni), antimony (Sb), As, Se, Cd, Pb, and Hg, were purchased from the National Institute of Metrology of China (Beijing, China). Three heavy metals, germanium (Ge), rhodium (Rh), and bismuth (Bi), used as internal standards, were purchased from China Nonferrous Metals and Electronic Materials Analysis and Testing Center (Beijing, China). Nitric acid (68%, ultra-pure analytical grade) was obtained from Merck (Darmstadt, Germany), and ultra-pure water (≥ 18.25 MΩ·cm) was prepared using a water purification system (Ewell Bio, Guangzhou, China). Microporous membranes (0.45 μm) were supplied by ANPEL (Shanghai, China).

Sample collection

In July 2018, a total of 480 urine samples (male: n = 236; female: n = 244) were collected from four regions, including two suburbs and two urban areas in Guangzhou. Each participant was requested to complete a questionnaire, to provide personal information including age, gender, weight, height, disease history, and alcohol and smoking habits. The demographic characteristics of the participants are listed in Table 1. Approximately 50 mL of first-voided morning urine was collected from each participant and sent to the laboratory in an ice-box, where it was stored at − 20 °C until use.

Table 1 Socio-demographic characteristics of the study participants
Full size table

Sample treatment and instrumental analysis

After thawing to room temperature, a 1 mL urine sample was diluted to 10 mL with nitric acid (1%) and vortex mixed. The urine sample was then filtered through a 0.45-μm microporous membrane and analyzed using inductively coupled plasma mass spectrometry (ICP-MS) (ICPQ, Thermo Inc, USA). The instrument parameters are described as follows: radio frequency power was set at 1550 W; plasma gas, helium collision gas, and carrier gas flows were 15, 4.3, and 1.05 L/min, respectively; and mass spectrometry resolution was 0.69–0.78 amu. At the same time, we used peristaltic pump to inject the internal standard solution containing 10 μg/L. The mass of each element and its corresponding internal standard are additionally listed in Table S1.

Quality assurance and quality control

The standard deviation value obtained from the blank determination 11 times is 3 times, and the corresponding concentration is the detection limit of each element. Except for Pb, the elements were below the limit of detection (LOD) in blank samples. The LOD and limit of quantification (LOQ) were 0.02–1.2 and 0.07–4 μg/L (Table S2), respectively. The regression coefficients of standard calibration curves used in this study were all < 0.99. The recoveries of heavy metals were generally satisfactory and ranged from 82.5 to 118.5%.

Health risk assessments

To assess the health risks of human exposure to the heavy metals, non-carcinogenic risks and carcinogenic risks (CR) were calculated. Specifically for non-carcinogenic risks, hazard quotient (HQ) represented the risk brought by an element, while the HQ sum, i.e., the hazard index (HI), represented the cumulative risk from all the elements, considering the additional action of the elements. The formulas used for health risk assessment are as follows (Chen et al. 2018; Liu et al. 2013; Peng et al. 2020):

$$TEDI=\frac{C\times V}{BW\times f}\times 0.001$$
$$HQ=\frac{TEDI}{RfD}$$
$$HI=\sum_{i}^{n}{HQ}_{i}$$
$$CR=TEDI\times CSF$$

where TEDI (μg/kg-bw/day) is the total estimated daily intake of a heavy metal; C (μg/L) is the urinary concentration of the element; V (L) is the volume of urine excreted per day, with 2 L/day used for adults (Chen et al. 2019) and the empirical urine volume for children calculated as V (L) = ((age − 1) × 100 + 400)/1000; BW (kg) is body weight; f (dimensionless) represents the urinary excretion factor of an element; RfD (mg/kg/day) is the reference dose of the element considering non-carcinogenic toxic endpoint; and CSF (mg/kg/day)−1 is the cancer slope factor of the element considering carcinogenic toxic endpoint. If an HQ > 1, it indicated potential non-carcinogenic risk for humans, and the same applied for HI. When CR value was larger than 1 × 10−6, the potential carcinogenic risk was considered to be unacceptable (USEPA 2001). The calculation parameters used in the study are listed in Tables S3–S5.

The Monte Carlo simulation (MCS) method was adopted to improve the accuracy of the risk assessment. The input probability distribution was based on Anderson-Dahlin, comparing each empirical statistical distribution with the sample probability distribution to obtain related parameters such as urine concentration, urine volume, and human body weight. The 95th confidence interval for health risk was calculated via Crystal Ball© (Oracle, Redwood City, USA) using 10,000 trials. Detailed information on the statistical distributions of the various parameters is shown in Table S6.

Statistical analysis

SPSS 13 (IBM, USA) software was used for the statistical analysis. The heavy metal concentrations used for descriptive statistics were expressed as μg/L. Because of the background interference of Pb, the blank values were subtracted. The significant difference was checked via Student’s t-test, with the confidence value p set at < 0.05. The correlation between any two variables was analyzed using Spearman correlation.

Results and discussion

Urinary concentrations of the heavy metals

The detection frequencies and concentrations of the heavy metals in urine are summarized in Table 2. Ni, As, Se, Cd, and Pb were frequently detected in the 480 samples, with detection frequencies of 94.4–100%. Comparatively, Cr, Sb, and Hg were only detectable in 38.5%, 33.5%, and 25.2% of samples, respectively. These high detection frequencies suggest that residents in the city are widely exposed to these heavy metals.

Table 2 Urinary concentrations (µg/L) of heavy metals in the general population of the city
Full size table

Although the heavy metals were widely detected, there were large concentration ranges from not detected to dozens of μg/L. The element As had the highest concentration (1.2 × 103 μg/L), with a median of 41.0 μg/L, which was higher than that (median = 19.0 μg/L) reported by China National Human Biomonitoring (CNHBM), comparable with levels found in Wuhan (median = 39.6 μg/L) and Chengdu (median = 40.3 μg/L) (Cao et al. 2021; Liu et al. 2020; Zeng et al. 2019), but lower than that found in Hainan (median = 73.2 μg/L) and the Yangtze River Basin (median = 73.6 μg/L) (Cui et al. 2017; Inoue et al. 2014).

For the other elements, the median concentration of Se was as high as 35.4 μg/L, followed by other elements in the sequence of Ni (2.25 μg/L), Cd (1.50 μg/L), Pb (0.96 μg/L), and Mn (0.28 μg/L). Concentrations of the three elements Cr, Sb, and Hg were generally lower than their LODs. Se concentrations (median = 35.4 μg/L) were lower than those in Chengdu (median = 323.1 μg/L) and equivalent to levels in Hainan (median = 36.5 μg/L) and Wuhan (median = 36.5 μg/L) (Inoue et al. 2014; Liu et al. 2020; Zeng et al. 2019). Urinary Ni levels were relatively high (median = 2.25 μg/L), although were lower than those of children in Hunan (median = 44.4 μg/L) but higher than those of the Chengdu and Wuhan-Zhuhai cohorts and in the Yangtze River Basin population (Cui et al. 2017; Liu et al. 2020; Tan et al. 2020). Although the median concentration of urinary Cd in the present study (1.05 μg/L) was higher than that of children in Hunan Province (median = 0.18 μg/L) and the CNHBM data (median = 0.34 μg/L), it was in line with those from other regions, such as Wuhan, Hainan, and Zhuhai cities in China (Cao et al. 2021; Cui et al. 2017; Inoue et al. 2014; Liu et al. 2020; Zeng et al. 2019). Pb levels were comparable to those of the CNHBM (median = 0.89 μg/L), but lower than those from Chengdu, the Wuhan-Zhuhai cohort, the Yangtze River Basin, and eight provinces in China (Cui et al. 2017; Ding et al. 2014; Liu et al. 2020; Tan et al. 2020). Relative to various other areas of China, the concentrations of the heavy metals in the urine of the residents were at the middle level. However, globally, the urinary concentrations of heavy metals in the city are at a high level, although much higher urinary As in Japan has been observed (Table S7).

The high As concentrations might be because this element is widespread in the environment, leading to human exposure through diet (including food and water) and inhalation. A previous study has shown that people who eat seafood have a higher urinary As concentration than people who do not eat seafood (Wang et al. 2018). According to the report by Huang et al. (2019), the per capita intake of aquatic products in Guangdong Province is up to 60.4 g/day. In addition, the desorption of As from (hydrogen) oxides and the oxidation or reductive dissolution of arsenic-containing minerals are the geological sources of water (Zhao et al. 2020). Se is an essential nutrient element for humans, with important biological functions. However, excessive intake of Se may cause a number of health problems such as skin cancer, deterioration in liver and kidney function, and pneumonia (Brown and Arthur 2001). The Se content of the human body depends on the Se content of the geographical environment and on eating habits. The average soil Se content in Guangdong Province is equivalent to the national average (Yu et al. 2007), and there is no correlation between soil Se deficiency and human Se deficiency in the city. However, it has been confirmed that seafood is rich in Se (Yamashita et al. 2013), and the city is an area with relatively high Se intake (Wei et al. 2015). Therefore, it is speculating that excessive consumption of seafood may have led to an increase in urine Se content in the present study. In addition, reports have shown that metal dust containing Se is discharged or enters the environment in the form of Se compounds during industrial production and the application of pigments, dyes, and rubber, causing pollution (Tan et al. 2016; Xiao et al. 2019). As a highly developed area, Guangzhou has large output and large products. It is close to the two major manufacturing cities of Dongguan and Foshan, which may partly explain the cause of the high urine Se concentrations in the population in Guangzhou.

In terms of the relatively high concentrations of Cd and Ni, a report suggests that the impact of eating seafood on the human body may not only increase urinary As concentration, but also the concentrations of Cd and Ni in the urine (Vimercati et al. 2017). A high level of Cd and Ni was often involved in industrial effluents discharged from metal processing which were prevalent in Guangzhou (Shang et al. 2017; Zhao et al. 2020). In addition, rice is one of the staple foods of the population. If the rice grown in Guangdong Province is continuously consumed, it may pose potential non-carcinogenic and carcinogenic risks to local residents because Cd pollution is widespread in the province’s paddy soil (Ma et al. 2017). Furthermore, inhalation is an important pathway in the human body’s exposure to Cd and Pb because of traffic factors, such as tire wear (Men et al. 2018) and urban traffic congestion (Mortada et al. 2002). Traffic congestion in the city is well known for its severity, especially during commuting hours (Men et al. 2018), which might contribute to the burden of heavy metals in the city’s residents.

In general, heavy metals in the study area can be mainly influenced by anthropogenic activities. Industry is another important source of pollution. Due to the economic structure and geographical location of Guangzhou (being very close to manufacturing cities such as Dongguan and Foshan), the heavy metals in the environment may come from these manufacturing industries, and this hypothesis is supported by other studies (Gu et al. 2014; Gu and Gao 2017). These may lead to an increase in the concentration of heavy metals in the urine.

Factors influencing urinary heavy metal levels

Many factors potentially influence the concentrations of heavy metals in humans. An understanding of these factors and how they influence the urinary concentrations of these elements can provide useful information in predicting the human exposure burden of heavy metals. In this study, the influence of gender, age, and region on urinary heavy metal levels was analyzed and is shown in Fig. 1. There were no significant differences in the mean concentrations of Cr, Mn, Ni, Sb, Pb, As, and Se between males and females. Nevertheless, average concentration of Cd in the urine of females was significantly higher than that of males (p < 0.05). Reported in the literature, smoking is one of the main sources of Cd in the general population (Järup and Åkesson 2009; Virani et al. 2016). In our study (Table 3), 12.1% of the participants had smoked in the 24 h before sampling, and their urinary Cd concentration was significantly higher than that of non-smokers (p < 0.01). Studies have shown that women are more likely to be exposed to Cd in their daily lives because they are more likely to be iron (Fe) deficient, and low stored Fe levels in the body may lead to increased Cd absorption (Gallagher et al. 2011; Satarug and Moore 2004). However, this may be of less influence than smoking in the present case.

Fig. 1
figure 1

Concentrations of heavy metals in the urine of general populations in city (a gender; b age; c area). Horizontal lines within the box and the lower and upper limits of the bar indicate median, 25%, and 75% values, and the squares represent mean values. The whisker extends to the last observation within 1.5 times the interquartile range. The asterisks outside the whiskers represent the outliers

Full size image
Table 3 Effects of demographic characteristics, dietary characteristics, and health status on urinary concentrations of heavy metals
Full size table

The different living habits and environment of the age groups may be factors affecting the urinary concentrations of heavy metals. In the present study, the subjects were divided into two age groups, i.e., < 18 years (teenagers) and ≥ 18 years (adults). The urinary concentrations of the elements for teenagers and adults are shown in Fig. 1b, which shows that there were significant differences (p < 0.01) between them for Ni, Cd, and Pb. The average concentrations of Cd and Pb in the urine of adults were significantly higher than those of teenagers, whereas the opposite result was observed for Ni, which may be related to the different lifestyle habits of the groups (Table 3). As previously mentioned, smoking is an important cause of higher urine Cd concentration. In this study, all the smokers were adults; thus, higher urinary Cd concentrations were expected. Similarly, Mortada et al. (2002) showed that smoking was related to an increase in urine Pb, and Ding et al. (2014) observed similar results to our study. Meanwhile, the results also found that the urinary Ni concentration of those taking pills was significantly lower than that of those not taking pills. 90.5% of the participants taking pills were adults. It is speculated that some ingredients in the drug inhibit the excretion of Ni, which leads to age differences.

Significant differences were observed in urinary Cr, Mn, Ni, Se, Cd, and Sb between suburban and urban areas (Fig. 1c) with the former higher than the latter except for Sb. The higher urinary concentrations in suburban areas may be related to the location of industries in the region. One of the largest textile production bases and manufacturer of jeans in China is situated in a suburban area of the city. As reported by Liang et al. (2013), textile dye wastewater contains, among other heavy metals, Pb, Ni, and Cr, leading to human exposure to these toxic elements. We also observed that the number of smokers in suburban areas was slightly higher than in urban areas, accounting for 56.9% of smokers, which might have contributed to the higher toxic element levels in the suburban group.

To further understand the influencing factors, genetic diseases, smoking, and other lifestyle habits were analyzed (Table 3). The presence of genetic disease did not significantly affect the urinary concentration of heavy metals (p = 0.061–0.987). Smoking is an important exposure route for Cd; however, we also explored whether Cd concentration had a significant relationship with taking medication. The Cd level of participants who took medication was significantly higher than that of those who did not take it (p < 0.01), while the opposite result was observed for Ni. Since the chemical composition of the medicine is relatively complex, the specific mechanisms of the effect of medication on Ni and Cd levels need to be further studied. In addition, the results showed that both smoking and alcohol consumption had significant effects on urinary Pb concentration (p < 0.01). Body mass index (BMI) can also influence Cd, Se, and Ni concentrations (p < 0.01). Our results saw urinary Cd and Se concentrations increasing with increasing BMI, whereas urinary Ni concentrations decreased. Among the participants, 88.9% of adults had BMI > 25 and 73.1% of children were of BMI < 18.5. The results indicate that Cd and Se levels require more attention in adults, while Ni should be the focus in children.

Human exposure and health risk assessment

The trace elements are excreted from the human body via urine, with half-lives of approximately 4–10 days (Marchiset-Ferlay et al. 2012; Ronteltap et al. 2007). Thus, heavy metal concentrations in urine can reflect recent human exposure to toxicants. The many forms of heavy metals need to be considered when estimating human internal exposure and the associated health risks. They may be metabolized by the body to produce several substances, including inorganic As (In-As) and organic As. The main elimination pathway of As is urinary excretion (Brima et al. 2006). In-As accounts for about 10–30% of the total arsenic in urine (Todorov et al. 2005), and As toxicity is calculated based on In-As. The content of In-As in urine is considered as 15% in the present study. In addition, Cr (VI) is more toxic than Cr (III), and the toxicity of As (III) is 2–10 times that of As (V) (Tchounwou et al. 2003). The chemical form of heavy metals determines their toxicity; thus, we based this study on the safest considerations and used As (III) and Cr (VI) for the calculations. For Hg, it should be noted that there are three main forms including metallic mercury (mercury vapor), and inorganic and organic forms. However, methylmercury is not mainly excreted through urine (Clarkson 1997). Urine is one of the important excretion pathways of Hg (II); therefore, the calculation of the risk assessment in the present study is based on the toxicity data for Hg (II).

TEDI was used in this study to estimate the human internal exposure per day, and HQ (HI) and CR were used to evaluate the associated health risks considering the non-carcinogenic and carcinogenic endpoints of the heavy metals, respectively (Table 4). At the same time, the two-dimensional (2-D) MCS method was used to provide accurate health risks (Figs. 2c and d). Among the nine target heavy metals, Se showed the highest median TEDI value of 1.60 × 10−3 μg/kg-bw/day, followed by As with 1.50 × 10−3 μg/kg-bw/day. The other elements had TEDI values of lower than 10−3 μg/kg-bw/day.

Table 4 Total estimated daily intake (TEDI) based on concentrations of heavy metals in the urine
Full size table
Fig. 2
figure 2

Human health risks estimated from urinary heavy metals for the general population of the city, China (a non-carcinogenic risks; b carcinogenic risks; c Monte Carlo simulation of hazard quotient [HQ]; d Monte Carlo simulation of carcinogenic risk [CR])

Full size image

The HQ values of the heavy metals, based on RfD, ranged from 8.3 × 10−4 to 34.3 (Fig. 2a). The HQ values were far less than one unit for Cr, Mn, Ni, Sb, and Pb, which indicates that exposure to these elements does not pose obvious health risks for the residents of the city. However, the 2-D MCS results (Fig. 2c) demonstrated that 8.83%, 8.31%, and 3.38% people suffered non-carcinogenic risks from the elements Cd, Se, and Hg (lg (HQ) > 0), respectively. Particularly, the HQ values of As from 38.2% of subjects exceeded one unit. In total, the HQ values of the elements followed a descending order of As > Cd > Se > Hg > Ni > Sb > Pb > Mn > Cr. Humans can be exposed to multiple heavy metals simultaneously, and HI as a measure of cumulative risk reflects this (Fig. 2a). The results showed that 69.8% of the subjects had HI values higher than one unit, indicating that heavy metals had potential non-carcinogenic risk for most of the city’s residents. Among the target elements, As was the main contributor to HI, with a mean of 50.0% of the total HI, indicating that non-carcinogenic risk from As should be given attention.

According to the Integrated Risk Information System (IRIS) of the US Environmental Protection Agency (http://www.eap.gov/IRIS), Cr, Pb, As, and Cd are classified as carcinogens. Thus, the carcinogenic risks for these four heavy metals were calculated and are shown in Fig. 2b. This risk ranked as As > Cd > Cr > Pb in our sample. The 2-D MCS results showed that 78.0% and 13.3% of the subjects exceeded the risk level of lg (CR) >  − 6 from Cr and Pb, respectively (Fig. 2d), indicating that most residents are subject to lifetime carcinogenic risk from Cr. More serious is that As and Cd both surpassed the acceptable risk level, with 90.1% and 35.4% of subjects exceeding lg (CR) >  − 4, respectively, for these elements, an unacceptable risk level. Considering the highly toxic properties of As and Cd, this unacceptable risk needs urgent attention.

Uncertainty analysis

The health risk of pollutants reflects their potential harm to human health. Many factors influence the assessed health risks of heavy metals, which may lead to an overestimate or underestimate of those risks. Considering the higher percentage of the subjects exceeded the carcinogenic risk level than that of non-carcinogenic risks of the heavy metals, the uncertainty of the estimation of carcinogenic risks of the present results was further discussed.

First, the TEDIs in this study were calculated from urinary concentrations of the elements. Various values for element excretion factor (f) are found in the literature, and we used generally suggested values for f. This would have affected the TEDI values and thus the CR values. Second, the TEDIs reflected total human intake by all pathways, including oral ingestion, inhalation, and dermal contact. However, the CSF value of a chemical is based on only oral intake of the element. To our best knowledge, pollutants entering human body through various pathways may result in different health hazards. In other words, the same concentration detected in the human body may have a different health risk depending on whether the pollutants entered through oral ingestion or inhalation. In this study, only CSF values were used and the oral ingestion pathway considered, potentially leading to an underestimate of the CR of the elements. Third, there is considerable uncertainty in the order of magnitude of the reference data. For example, a CSF value of 1.5 (mg/kg/day)−1 was used for As in our study; however, a CSF value of 25.7 (mg/kg/day)−1 was found in one toxicological review (Lei et al. 2015; EPA/635/R-10/0012010), which is more than 17 times higher than our study value. Finally, this was identified the human exposure pathways without external exposure data combined with data analysis. The next step is to combine internal and external exposures with environmental samples. These uncertainties may have led to the overestimation or underestimation of health risks. Despite these uncertainties, our study results still provide an understanding of human exposure to heavy metals and the associated health risks to those exposed.

Conclusions

This study measured the concentrations of nine heavy metals in the urine of residents in Guangzhou in southern China. As, Se, Cd, and Pb are ubiquitous in the environment with detection frequency of up to 98%, of which As is the dominant contaminant. Gender, age, and regional significant differences in element concentrations were observed. Females had higher urine Cd concentrations than males. Cd and Pb concentrations were higher in the urine of adults than that of teenagers. In addition, Cr, Mn, Ni, Se, Cd, and Sb in the suburban samples were significantly higher than in the urban samples. Lifestyle and habits such as drinking alcohol and smoking can also significantly affect the urinary concentrations of heavy metals. According to the health risk assessment, 38.2%, 8.83%, 8.31%, and 3.38% of subjects were exposed to non-carcinogenic risks from As, Cd, Se, and Hg. The carcinogenic risk level of 10−6 was surpassed for As and Cd in the subjects and 90.1% of subject surpassed the acceptable risk level (10−4), which requires particular attention.