Urinary Metal Concentrations and the Incidence of Hypertension Among Adult Residents Along the Yangtze River, China

Assessment of EMR ML Mining Methods for Measuring Association between Metal Mixture and Mortality for Hypertension

INTRODUCTION

There are limited data available regarding the connection between heavy metal exposure and mortality among hypertension patients.

AIM

We intend to establish an interpretable machine learning (ML) model with high efficiency and robustness that monitors mortality based on heavy metal exposure among hypertension patients.

METHODS

Our datasets were obtained from the US National Health and Nutrition Examination Survey (NHANES, 2013-2018). We developed 5 ML models for mortality prediction among hypertension patients by heavy metal exposure, and tested them by 10 discrimination characteristics. Further, we chose the optimally performing model after parameter adjustment by genetic algorithm (GA) for prediction. Finally, in order to visualize the model's ability to make decisions, we used SHapley Additive exPlanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) algorithm to illustrate the features. The study included 2347 participants in total.

RESULTS

A best-performing eXtreme Gradient Boosting (XGB) with GA for mortality prediction among hypertension patients by 13 heavy metals was selected (AUC 0.959; 95% CI 0.953-0.965; accuracy 96.8%). According to sum of SHAP values, cadmium (0.094), cobalt (2.048), lead (1.12), tungsten (0.129) in urine, and lead (2.026), mercury (1.703) in blood positively influenced the model, while barium (- 0.001), molybdenum (- 2.066), antimony (- 0.398), tin (- 0.498), thallium (- 2.297) in urine, and selenium (- 0.842), manganese (- 1.193) in blood negatively influenced the model.

CONCLUSIONS

Hypertension patients' mortality associated with heavy metal exposure was predicted by an efficient, robust, and interpretable GA-XGB model with SHAP and LIME. Cadmium, cobalt, lead, tungsten in urine, and mercury in blood are positively correlated with mortality, while barium, molybdenum, antimony, tin, thallium in urine, and lead, selenium, manganese in blood is negatively correlated with mortality.

Predictors of urinary heavy metal concentrations among pregnant women in Jinan, China

BACKGROUND

Toxic heavy metal exposure and insufficiency or excess of essential heavy metals may have negative effects on pregnant women's health and fetal growth. To date, the predictors of pregnant women's heavy metal exposure levels remain unclear and vary with different regions. The study intended to explore potential predictors of exposure to heavy metals individually and high co-exposure to heavy metal mixtures.

METHODS

We recruited 298 pregnant women in first trimester from prenatal clinics in Jinan, Shandong Province, China, and collected spot urine samples and questionnaire data on their demographic characteristics, lifestyle habits, consumption of food and dietary supplement, and residential environment. All urine samples were analyzed for seven heavy metals: cobalt (Co), molybdenum (Mo), strontium (Sr), arsenic (As), cadmium (Cd), lead (Pb) and mercury (Hg).

RESULTS

Factors associated with single heavy metal concentration were as follows: a) urinary As, Sr and Cd increased with women's age respectively; b) pregnant women with higher monthly household income per capita had lower Sr and Mo levels; c) pregnant women with intermittent folic acid supplementation and those not taking tap water as domestic drinking water had lower Sr concentrations; d) Cd was positively linked with consumption frequency of rice; e) Hg was adversely related to consumption frequency of egg and the women who took purified water as domestic drinking water had lower Hg exposure. In addition, pregnant women's age was positively associated with odds of high co-exposure to Co, As, Sr, Mo, Cd and Pb; while those with an educational level of college had lower odds of high exposure to such a metal mixture compared with those whose educational levels were lower than high school.

CONCLUSION

Predictors of single urinary heavy metal concentration included pregnant women's age (As, Sr and Cd), monthly household income per capita (Sr and Mo), folic acid supplementation (Sr), rice consumption frequency (Cd), egg consumption frequency (Hg) and the type of domestic drinking water (Sr and Hg). Pregnant women with older age, lower educational level tended to have high co-exposure to Co, As, Sr, Mo, Cd and Pb.

Relationship between Occupational Metal Exposure and Hypertension Risk Based on Conditional Logistic Regression Analysis

Occupational exposure is a significant source of metal contact; previous studies have been limited regarding the effect of occupational metal exposure on the development of hypertension. This study was conducted to assess the levels of exposure of certain metals (chromium (Cr), iron (Fe), manganese (Mn), and nickel (Ni)) in hypertensive and non-hypertensive workers and to assess the relationship between the risk of hypertension and metal exposure level. Our study included 138 hypertensive patients as case groups and 138 non-hypertensive participants as controls. The exposure risk level was divided according to the limit value after collecting and testing the metal dust in the workshop. Considering the influence of single- and poly-metal, single factor analysis and conditional logistic regression analysis of poly-metal were carried out. The results of the model indicated that the incidence of hypertension increased with an increase in Cr exposure level, and the risk of hypertension was 1.85 times higher in the highest exposure than in the lowest exposure (95% CI: 1.20−2.86, p < 0.05). Mn has the same effect as Cr. There was no significant correlation between Fe or Ni and hypertension. Our findings suggested that Cr and Mn exposure in the work environment might increase the risk of hypertension, while no effect of Fe and Ni on blood pressure was found. Prospective study designs in larger populations are needed to confirm our findings.

Co-exposure to priority-controlled metals mixture and blood pressure in Chinese children from two panel studies

Metals may affect adversely cardiovascular system, but epidemiological evidence on the associations of priority-controlled metals including antimony (Sb), arsenic (As), cadmium, lead, and thallium with children's blood pressure (BP) was scarce and inconsistent. We conducted two panel studies with 3 surveys across 3 seasons among 144 and 142 children aged 4-12 years in Guangzhou and Weinan, respectively. During each seasonal survey, urine samples were collected for 4 consecutive days and BP was measured on the 4th day. We obtained 786 BP values and urinary metals measurements at least once within 4 days, while 773, 596, 612, and 754 urinary metals measurements were effective on the health examination day (Lag 0), and the 1^st, 2^nd, and 3^rd day preceding BP measurement (Lag 1, lag 2 and lag 3), respectively. We used linear mixed-effect models, generalized estimating equations and multiple informant models to assess the associations of individual metal at each lag day and accumulated lag day (4 days averaged, lag 0-3) with BP and hypertension, and Bayesian Kernel Machine Regression to evaluate the relations of metals mixture at lag 0-3 and BP outcomes. We found Sb was positively and consistently related to systolic BP (SBP), mean arterial pressure (MAP), and odds of having hypertension within 4 days, which were the strongest at lag 0 and declined over time. And such relationships at lag 0-3 showed in a dose-response manner. Meanwhile, Sb was the only contributor to the relations of mixture with SBP, MAP, and odds of having hypertension. Also, synergistic interaction between Sb and As was significant. In addition, modification effect of passive smoking status on the association of Sb and SBP was more evident in passive smokers. Accordingly, urinary Sb was consistently and dose-responsively associated with increased BP and hypertension, of which Sb was the major contributor among children.

The Association Between Life Events and Incidence of Hypertension Among Government Employees in China: A Prospective Cohort Study

Background

Hypertension (HTN) is a global public health concern. However, the association between life events (LEs) and HTN is complex. Thus, we conducted a prospective cohort study to explore this complex association.

Methods

A total of 8,077 government employees without HTN were recruited through cluster sampling between 2018 and 2019 in Hunan Province, China. At baseline, information regarding sociodemographic characteristics, LEs, and behavioral factors was collected. After the 1-year follow-up, the participants were revisited to obtain the HTN diagnosis. Crude and adjusted Poisson regression models were constructed to calculate the incidence rate ratios (IRRs) and 95% confidence intervals (CIs). Cubic regression spline models were used to visualize the trends between LEs and HTN IRRs. Interactive and subgroup analyses were also performed.

Results

The 1-year HTN incidence rate among government employees in Hunan province was 4.30% (95% CI: 3.86-4.74%). LEs were associated with a higher HTN risk (IRR, 1.02; 95% CI, 1.00-1.04). When calculating positive and negative LEs scores separately, only the latter was a risk factor for HTN incidence (IRR, 1.04; 95% CI, 1.03-1.06); conversely, positive LEs reduced the risk (IRR, 0.90; 95% CI, 0.85-0.96). Compared with patients in the lowest quartile of LEs score, those in quartiles two (IRR, 1.28; 95% CI, 0.96-1.71), three (IRR, 1.43; 95% CI, 1.04-1.96), and four (IRR, 1.73; 95% CI, 1.26-2.37) were at progressively higher risk. In restricted spline curves, a non-linear association was noted between LEs and HTN risk. Regarding the subcategories of LEs, work-related LEs, personal LEs, and all subcategories of negative LEs were associated with an increased risk of HTN. However, among positive LEs, only the family-related cases were associated with a lower risk of HTN.

Conclusion

LEs had a non-linear association with an increased risk of HTN. Negative LEs were risk factors for HTN incidence, whereas positive LEs reduced the risk of HTN. Thus, the importance of LEs should be highlighted in the development of HTN prevention strategies and initiatives.

Family History of Hypertension and Cobalt Exposure Synergistically Promote the Prevalence of Hypertension

It has been previously reported that family history of hypertension (FHH) and exposure to metals are each independent risk factor for hypertension, but the interaction between the two in relation to hypertension risk has been poorly studied. The object of this study is Dong ethnic group in Guizhou, China. The impacts of exposure to metals and FHH on hypertension incidence were examined by using the restrictive cubic spline (RCS) model as well as the multivariate logistic regression model. As a result, FHH, together with cobalt and lead exposure, was identified to show independent significant correlation with hypertension incidence (P < 0.05). The risk of hypertension increased with the increase in lead and cobalt exposure quartiles. Typically, the RCS model revealed such dose-response relation. To further confirm the association of cobalt, lead, and FHH with the risk of hypertension, multiplication and addition models were used to analyze the influence of the interactions between these variables on the risk of hypertension. The results showed that there was a multiplying interaction between the influence of the FHH and cobalt on the risk of hypertension. As for the additive interaction between cobalt and FHH, the relative excess risk due to interaction (RERI) was determined to be 0.596 (95% Cl: 0.001-1.191), the attributable proportion due to interaction (AP) was calculated as 0.256 (95% Cl: 0.075-0.437), whereas the synergy index (S) was identified to be 1.814 (95% Cl: 1.080-3.047). Our study provides some limited evidence that a FHH and cobalt exposure synergistically promote the prevalence of hypertension.

Multiple metal exposure and obesity: A prospective cohort study of adults living along the Yangtze River, China

Association between long-term exposure to multiple metals and obesity remains inconclusive, and prospective evidence on the region along the Yangtze River was limited. Thus, our study aimed to examine the association of multiple metal exposure and obesity. We measured baseline urine levels of 22 metals of 982 adults living along the Yangtze River, incidence of obesity was calculated from body mass index (BMI) and waist circumference (WC) measured at follow-up survey. Cox proportional hazards models were used to examine the hazard ratios (HR) and 95% confidence interval (CI) for the association between urinary metals and obesity, and the mixing effect of metals on obesity was estimated by using quantile g-computation. In multiple-metal models, arsenic was significantly associated with BMI/obesity, with the HR in the highest quartiles of 0.33 (95% CI: 0.16, 0.69; p-trend = 0.004). The HRs for WC/obesity of arsenic and molybdenum were 0.49 (95% CI: 0.32, 0.75 for the fourth vs. first quartile; p-trend = 0.002) and 1.83 (95% CI: 1.25, 2.70; p-trend = 0.001), respectively. Quantile g-computation mixtures approach showed a significantly negative joint effect of multiple metals on WC/obesity, with the HR of 0.26 (95% CI: 0.14, 0.47; p < 0.001) when increasing all seventeen metals by one quartile. Our study suggests that all seventeen metal mixed exposure may be negatively associated with obesity. Further cohort studies are needed to confirm these findings and clarify the underlying biological mechanisms.

Bird's feather as an effective bioindicator for detection of trace elements in polymetallic contaminated areas in Anhui Province, China

Environmental pollution, especially because of trace metals, seriously affects ecological safety, and bird feathers are often used as bioindicators to monitor this risk in various environments. However, the feasibility of feathers as bioindicators for trace metals in polymetallic contaminated areas has not been extensively studied. In this study, we used inductively coupled plasma mass spectrometry (ICP-MS) to quantify and compare the contents of nine trace metal(loid)s (V, Cr, Mn, Co, Cu, Zn, As, Cd and Pb) among soil, plants, insects and birds (feathers and internal tissues) sampled in the mining area of Tongling, a polymetallic contaminated area in Anhui Province, eastern China. We detected significant trace metal pollution in the abiotic and biotic materials. The contents of Cr, Cu, Zn, As and Pb in feathers differed among bird species and among sampling sites, with higher contents often recorded in tree sparrows (Passer montanus). The metal(loid)s V, Mn, Co, Zn, and As had higher contents in feathers than in internal tissues including heart, liver, kidneys, muscles and bones. The contents of some elements in feathers were positively correlated with those in internal tissues, for example, Co, As, and Cd in the heart, V and Co in the kidneys, Cd in the liver, Pb in bones, and As in muscles. Furthermore, the contents of V, Cr, As and Pb in feathers were higher than those in other biomaterials, implying an increasing trend from plants, insects, and feathers. Our study indicates that bird feathers can be used as effective, non-destructive bioindicators to monitor trace metal(loid) pollution, especially for V, Co, As, Cd and Pb, in polymetallic contaminated areas, providing reliable information for ecological assessment.