According to our knowledge, this is the first study to use advanced multipollutant models to assess the individual and joint associations of 12 urinary metals with four major pulmonary function parameters (FEV1, FVC, FEF25 − 75%, and PEF) in the pediatric population (6–17 years old). Overall, we observed a trend of declining pulmonary function with increasing concentrations of the metal mixture. And Pb had the highest weight to the negative associations, and revealed an almost "L"-shaped association with FEV1, FVC, and PEF. Additionally, there was suggestive antagonistic interactions between Pb and Cd. And Ba was positively associated with lung function metrics.
Broadly, we found that Pb was the most highly weighted of the 12 metals studied, and it was inversely associated with all four lung function measures. Our findings are congruent with those of earlier cross-sectional researches in healthy Chinese youngsters and American children and teens(Feng et al., 2022; Pan et al., 2020). The association between blood Pb, Hg, and Cd and pulmonary parameters was investigated in 221 healthy children, and only Pb levels were significantly associated with the FEV1 and FVC impairment (Pan et al., 2020). And blood Pb was also found to be the most damaging on FEV1 and FVC decline when the combined effects of polycyclic aromatic hydrocarbons (PAHs) and blood metals (Pb, Cd, and Hg) on health pulmonary function indices in 6-19-year-old children from the NHANES survey were analyzed (Feng et al., 2022). And unlike most previous studies merely focused on large airway function (FEV1 and FVC), our study examined the individual and joint effects of 12 urinary metals, adding evidence on small airway function metrics (FEF25 − 75% and PEF). PEF is a flow-limited and effort-independent metric that describes lung and airway mechanics (Tantucci et al., 2002); and FEF25 − 75% represented small airway function and was demonstrated to be more sensitive than FEV1% in reflecting airway hyperresponsiveness, lower airway inflammation, and allergy sensitization (Ciprandi & Cirillo, 2019; Qin et al., 2021).
Interestingly, discrepancies were found when comparing our study to a prior one that included 1,234 children aged 6 to 17 and used data from the NHANES 2011–2012 survey cycle. With multivariate linear regression, the negative effects of urine Pb were only identified in FEF25 − 75% (Madrigal et al., 2018). Divergent methodologies and sample sizes may be responsible for the inconsistencies. A study with 206 preschoolers revealed no association between Pb and lung function levels (Zeng et al., 2017). And exposure scenarios could also induce divergence; for instance, associations between 17 toenail metal concentrations and pulmonary function parameters among children with asthma (n = 39) in Chicago were assessed, and no association of toenail Pb with pulmonary function was found (Madrigal et al., 2021). Aside from the limited sample size, the reliability of many toenail metals has not yet been fully established (White et al., 2018), and earlier study suggested no correlation between levels of Pb in toenails and urine (Kuiper et al., 2014). The disparity of biomarkers in urine and toenails could reflect the varying exposure timings and accumulations, as well as lead exposure from various sources. Evidence for this is, however, limited, and further research will need to clarify this.
Additionally, we discovered that Cd was negatively associated with FEV1 and FVC and made a significant contribution to the combination effect; while the null association was seen with FEF25 − 75% and PEF when considering the effects of several other metals. Cd is a known lung toxicant that can accumulate in tissues and organs via blood circulation, causing respiratory diseases when inhaled, ingested, or absorbed via the skin. Several rat models also provided supportive evidence that exposure to Cd could alter lung structure and function(Lai & Diamond, 1992; Surolia et al., 2015). Epidemiology researches have frequently indicated that Cd exposure was linked with impaired pulmonary function in occupational workers (Moitra et al., 2013; Moitra et al., 2015), general adults (Sobel et al., 2022; Yoon et al., 2015), and smokers (Mannino et al., 2004; Rokadia & Agarwal, 2013), but there have been few studies on children. Toenail Cd was found to have a significant negative association with FEV1 in Chicago school-aged asthmatic children (n = 39), but not with FVC (Madrigal et al., 2021). Interestingly, two similar studies with NHANES data, one in children aged 6–17 demonstrated no association of both blood (n = 1234) and urinary (n = 408) Cd with spirometry metrics without taking other metals’ effects into consideration(Madrigal et al., 2018); while another study involving 6–19 years children (n = 1734) suggested an inverse relationship with FEV1 (β=-89.8, 95%CI, -155.3, -24.3, p = 0.007), but not with FVC (β=-71.2, 95%CI, -145.9, 3.5, p = 0.062) when considering the influence of polycyclic aromatic hydrocarbons (PAHs) and other metals, therein Pb and Hg, in multivariate regressions (Feng et al., 2022). Therefore, the methodology differences may explain some of our discrepant results. Additionally, three cross-sectional studies based on the Chinese population reported no association of Cd with FEV1 and FVC (Leung et al., 2013; Pan et al., 2020; Zeng et al., 2017). Of those, one among children with a mean age of 12.5 years who without diagnosis of heavy metal exposure (n = 221) (Pan et al., 2020), one in preschoolers living in areas with electronic waste recycling facilities (Zeng et al., 2017). And another study with preschool children (n = 893) further indicated no association of Cd with FEF25–75% and PEF (Leung et al., 2013). Differences in sample size, age, environment, and statistical models may all account for the discrepant results between previous studies and ours.
It is noteworthy that, in our study, Pb and Cd had antagonistic interactions. In contrast to our findings, a newly published cross-sectional study revealed a slight synergistic effect of Pb and Cd on children's lung function (Pan et al., 2020), and another reported an absence interaction(Zeng et al., 2017). The antagonized toxic effect of Pb and Cd co-exposure revealed in the current study could be due to possible competition between Pb and Cd (Pandya et al., 2010). More antagonistic interactions were found in which the presence of Cd mediated Pb uptake and the development of Cd had a significant impact on Pb in mouse kidneys (Smith et al., 2012). Moreover, the urinary half-lives of the metals studied varied from 1 to 4 days for Pb (Nordberg et al., 2014), and up to around 14 years for Cd (Suwazono et al., 2009). Thus, Cd and levels imply chronic exposure, whereas Pb may indicate acute exposures, i.e., within the last few weeks. Our findings may suggest that acute metal exposure may be a stronger predictor of pulmonary function decline. Still, few other studies have addressed the precise nature of the interacting contributions, and additional studies using different samples or other methods are needed to generate empirical evidence.
Tu was first shown to be strongly linked with decreased lung function parameters among children. Since the effect of Tu has historically been assessed in factory workers through welding processes, which involves the potential hazards for inhalation exposures that may lead to acute or chronic respiratory diseases (Meo & Al-Khlaiwi, 2003). To our best knowledge, only one recent study has taken Tu in a general population of American Indian adults, in both models of single metals and metal-mixtures, associations for Tu were mostly null, with no consistent pattern of relationship for Tu across the different spirometry endpoints (Sobel et al., 2022). Currently, there is no consensus regarding the association of Tu with lung function parameters, and this is the first study to be conducted in a general population of children, more epidemiological and toxicological evidences are warranted.
Unexpectedly, Ba was found to be positively associated with lung function metrics. The respiratory health impact of Ba was limited and discordant. Similar to our findings, a case-control study in Chinese adults (n = 1102) suggested that urinary barium was likely to have a protective effect on asthma (OR,0.44, 95%CI, 0.27, 0.71) (Huang et al., 2016). Whereas the null finding was reported by a study among college students (n = 21) regarding Ba as one of the chemical constituents of particulate matter in relation to pulmonary function (Wu et al., 2013). A possible explanation for this difference may be explained by their limited sample size. Two animal models have shown the negative effect of Ba on bronchopulmonary reactivity (Hicks et al., 1986), and inflammatory cell infiltration in the lung (Ueha et al., 2020). As Ba does not lead to bioaccumulation in human tissues, but if insoluble compounds are inhaled, their levels can build up in the lungs to cause a benign condition known as “Baritosis” (Peana et al., 2021). However, there is most likely a good safety margin between even the highest natural barium intake and the threshold for toxic effects, and further research is needed to explore the association.
Although the specific underlying molecular mechanisms are yet unknown, results from epidemiological and experimental studies suggest that the relationship between multiple metals and lung function was primarily due to the formation of reactive oxygen species (ROS) and oxidative stress brought on by metal exposures. For example, the longitudinal study of 1243 workers with a 4-year follow-up has similar results with the current study; in addition to the effects of Pb on the reduction of FEV1, they further suggested that high Pb exposure was associated with an increase in 8-iso-PGF2α, supporting the effect of Pb exposure on elevated lipid peroxidation and redox imbalance, which could eventually lead to morphological and functional changes in lung epithelial cells (Wei et al., 2020). Oxidative stress is the well-documented mechanism mediating the association between Pb and Cd exposure and lung function impairment (Atapaththu et al., 2016; Moitra et al., 2015; Wei et al., 2020). And heavy metals were suggested to increased 8-iso-PGF2α and malonaldehyde (MDA) generation (Ashrap et al., 2021; Jomova & Valko, 2011), these two biomarkers were commonly used as oxidative stress indicators signaling lipid peroxidation. And sustained exposure to ROS may cause DNA damage in epithelial cells, initiating apoptotic pathways via the overexpression of p53 and TGF, as well as the secretion of numerous factors from fibroblasts (Plataki et al., 2005). Increased apoptosis of pulmonary epithelial cells may result in a loss of cell turnover equilibrium, and fibroblast-induced factors, leading to aberrant reepithelialization (Kuwano, 2007). The above series of mechanistic pathways would result in decreased lung compliance and restricted lung function (Gutteridge & Halliwell, 2000). However, there is still much to learn about the complete picture of putative mechanistic pathways for metal mixtures on pediatric lung function.
Our study findings need to be interpreted in the context of certain limitations. First, after excluding the population that does not fulfill the research's inclusion requirements, the population representativeness and the extensibility of results may be influenced. Second, substituting values below the lower LOD with LOD divided by the square root of 2 is likely to introduce bias(Nie et al., 2010). Third, we made multiple comparisons in both the linear regression models and the BKMR models, which increased the possibility of false positive results. Fourth, because there was only one measurement of the metal concentration per individual at a single point in time, urinary metals could only reflect short-term exposure to metals from all sources, which could raise some potential concerns about exposure misclassification. Multiple samples taken from people at various times can enhance the accuracy of exposure assessment. Fifth, although several significant confounders were controlled for, the confounding effects of other factors must not be neglected. Finally, because our study was cross-sectional, it was not possible to draw conclusions about causality. Accordingly, large-scale experimental or prospective design investigations should be conducted to confirm the causal links between metal mixtures and lung functioning.