Background
Lead (Pb)-gasoline has been phased out in China since the 1980s, and was completely banned all over China since 2000 [
1]. Though laws were enforced rigorously to remove the use of Pb sources in commercial, and industrial sectors, in China, Pb exposure continues to be an important risk factor, affecting almost every organ, and system in the human body [
2]. The current mass epidemiological investigations mainly focused on occupational Pb poisoning, or childhood Pb exposure [
3,
4]. In recent years, medical big data has made it possible to identify the association of Pb exposure with disease profile based on the massive/large sample size of cross-sectional studies.
Pb is generally known as a widespread environmental pollutant, and causes a wide range of toxic effects, with no manifest biological roles. Traditional epidemiological studies, by using cross-sectional, case-control, cohort, or case report methods, have elucidated the relations between Pb, and certain kinds of diseases. Their research findings pointed to the association between Pb exposure and circulatory system diseases, e.g., hypertension, intracerebral hemorrhage, atherosclerosis and acute ischemic stroke [
5‐
7], and a great affinity of the cerebral endothelial cells to Pb suggested that Pb concentrations in endothelial cells were much higher than that in other brain cell types [
8]. Previous studies also uncovered that vasogenic edema, increased Blood-Brain Barrier (BBB) permeability as well as cerebellar hemorrhage were significant manifestations of Pb-induced damage of microvessels in developing brain [
9,
10]. Nephrotoxicity may occur as a result of Pb exposure because kidney was the main excretion route, only by which intracorporal Pb would be eliminated [
11]. Chronic Pb exposure at levels above 700–800 μg/L was an established indicator for chronic kidney disease (CKD) [
12]. Proximal tubular cells of the renal tubules accumulated Pb through specific Pb-binding proteins as the form of typical intracellular inclusions [
13,
14].
The research subjects of our study are outpatients and inpatients who were required to test blood Pb concentrations in military hospitals for various symptoms. International Statistical Classification of Diseases, and Related Health Problems (ICD-10) code was implemented by coders of Medical Records Management Division based on doctor’s clinical diagnosis [
15]. Thus, with this convenience brought by ICD in exchanging and sharing disease information across various hospitals, our study aimed to investigate (1) current status of blood Pb levels for patients who visited military hospitals; (2) the association of Pb exposure with the whole diseases’ profile, and corresponding medical burden; (3) associations between covariates and blood Pb levels. The results from this study may help us understand the underlying Pb-induced toxicity in human body system.
Methods
Study population
Study population comprises patients receiving whole blood Pb concentrations test in treatments. The present study was approved by the institutional review boards of Air Force Medical University (Registry no. KY20173346–1), and all studies were conducted in accordance with the ethical principles for medical research involving human subjects as defined in the Declaration of Helsinki. A total number of 294,402 outpatients/inpatients of all military hospitals between 2013 and 2017 were included as subjects by non-probability sampling using SQL expression of Oracle databases. Records lacking in any primary keys, including hospital information, patient’s identity, or visiting times, were excluded. Given that China is a country with 56 ethnic groups, we also used the self-report ethnic information to reveal in vivo Pb heterogeneity levels in different ethnic groups. Pb content measured by Atomic Absorption Spectroscopy (AAS) in whole blood was expressed as micrograms per liter (μg/L).
Data resource
This study aims to provide a comprehensive review of medical big data collected from military hospitals, which serves not only the military personnel, but also civilians, taking up a significantly high proportion of Chinese health services. The data was collected in the treatment, which includes patients’ demographic information, vital signs, diagnoses of diseases, medical orders, examination reports, lab tests results, drugs, medical supplies or even infections. Patients’ information was recorded in the processes of registration, ward management, medical order, auxiliary examination, and even in surgery. Besides, the information flow within departments of each hospital were documented by Admission-Discharge-Transfer (ADT), document entry, drug placement, or medical record cataloging. Medical costs were composed of bed, clinical use of drug, laboratory examination, X-ray, treatment, blood test, operation, nursing care, etc.
Data mining
Data processing is done by Structured Query Language (SQL) in Oracle database 10 g. Each medical record consists of information including patient’s demographic characteristics, vital signs, diagnoses, ADT, surgeries, medical orders, examination reports, lab tests, drugs, medical supplies, medical costs, nursing care, infections, and treatment outcomes. First, we extracted inpatients from this data set, those who had received blood Pb concentration test in hospitals. The primary key of each patient’s record includes hospital code, patient ID, and visit times. Medical costs as well as diagnosis information were included during the analytical process. With a reference to the above indicators, the main diagnostic results, and other important information were retrieved from this relational database for further correlation analysis of Pb exposure.
Study setting
Data sets from more than 200 military hospitals between 2013 and 2017 were obtained, with each hospital equipped with a Hospital Information System (HIS) for uniformed services as well as similar data dictionary. People’s Republic of China (PRC) covers an area of approximately 9,600,000 km2, governing 23 provinces, five autonomous regions, four direct-controlled municipalities (Beijing, Tianjin, Shanghai, and Chongqing), and the special administrative regions of Hong Kong, and Macau. The hospitals in this study are mostly located in the capital cities of these provinces. Patients were labeled with two geographic tags, their birthplaces implicated in their ID card, and the workplace implicated in the postal code of China. We intended to outline Pb spatial distribution in patients by these two geographic tags.
Statistical analysis
Our analysis is done on the basis of patients’ disease profile, and medical burden of Pb exposure in more than 100 cities, which are either provincial capitals or cities in key regions, accounting for about 30% of the major cities in China. Personal profiles were compiled based on demographic factors, including gender, age, or blood types of patients receiving blood Pb test. We counted Pb values every single time when patients got Pb tested for multiple times. Kolmogorov-Smirnov test was employed to identify the distribution of whole blood Pb concentration, with Q-Q plot applied to visualize the results. To analyze our data by a cross-sectional approach, we evaluated the differences between two independent samples by Wilcoxon rank sum test, or multiple independent samples by Kruskal-Wallis H test. Medical cost structure was analyzed based on all details of medical expenses. Data were analyzed, or illustrated using GraphPad Prism 6, and Python (version 5.1.0; Anaconda3). Values were considered statistically significant at p < 0.05.
Discussion
In the present study, we investigated the blood Pb levels among the a massive number of patients with various diseases, and found that: (1) the average blood Pb concentrations of Chinese patients was 28.36 μg/L; (2) obvious geographical differences existed in blood Pb levels; (3) Han Chinese patients’ Pb levels were significantly lower than other minorities groups; (4) Pb exposure may be an important risk factor for malignant neoplasm of lung, considering the significantly higher Pb levels in lung cancer than other respiratory, and intrathoracic organs.
Compared with the blood Pb levels reported in previous research targeting occupational populations [
4], children [
3], or adults [
16], Chinese patients as a whole, reported the lowest Pb levels, which to some extent represents conditions of residence throughout China. The findings may also be attributed to differences in the age structures of all the subjects. About 30% patients were over 60 years old, and their Pb concentration was the lowest compared with the rest three groups, 16.00 μg/L.
This study revealed the geographical differences of blood Pb levels in patients who visited hospitals, which could partially represent general population in China [
1,
17]. Systematic review of published papers illustrated that, of 24 provinces, or regions, only 4 provinces (Hunan, Guangdong, Gansu, Jiangxi) showed a trend of increased mean blood Pb levels, and the prevalence of Pb poisoning [
18]. However, two adjacent provinces, namely, Yunnan, and Guangxi Zhuang Autonomous Region, showed opposite trend in our survey, with Yunnan being the highest, and Guangxi the lowest. Similar results were observed not only in birthplace, but also in workplace, suggesting career as a factor related to Pb exposure. The fact that Yunnan has long been taking the first place in deposits of heavy metals may account for the Pb accumulation of patients born or worked in Yunnan [
19]. Meanwhile, patients from Guangxi mainly dwelled in Guilin, a prefecture in the far south of China. This huge contrast of patients’ Pb levels between two regions may result from the heterogeneity of their living environment.
China is a multi-ethnic country. Han Chinese, as the majority, showed a significantly lower blood Pb levels than other minorities. The difference between Han group, and other groups has been pointed out in previous study, reporting a mean blood Pb levels of 55.70 μg/L, 53.00 μg/L in Uyger, and Han children, respectively [
20]. This difference in Pb levels may be attributed to the varied living, and eating habits of ethnic groups. Remarkably, patients with Rh negative blood type seemed to accumulate more Pb than Rh positive. Blood serves as an important medium for Pb transport. More than 99% of blood Pb stays with red cells, while the rest is distributed in plasma [
21]. Rh proteins form a core complex which is critical to the structure of the erythrocyte membrane [
22], and might affect the sequestration, or excretion of Pb between red cells, and blood. Interestingly, Pb levels became lower when patients were hospitalized for a longer time. This may be attributed to medical interventions, professional health education, and guidance on behavior or nutrition from doctors during treatment.
In 2016, Pb exposure accounted for 540, 000 deaths and 13.9 million years of healthy life lost (disability adjusted life years, DALYs) worldwide due to long-term effects on health [
23]. An environmentally attributable fraction model was used to estimate Pb-attributable total economic costs, $ 227.7 billion in Eastern Asia (China, Mongolia), represented by each 1-year cohort of children < 5 years of age [
24]. DALYs are particularly useful for prioritizing public health interventions, meanwhile, total economic cost is used as a tool to assess medical burdens of a country or an area. Our study discovered that the average medical cost due to Pb poisoning was 6888 CNY, of which drug cost took the largest proportion, 4670 CNY. These findings suggested that the treatment of Pb poisoning, especially drug treatment, may be the major resource of medical burden for patients under Pb exposure, and average medical cost needs to be considered when assessing personal economic burden of residents, especially those in developing countries or areas.
Almost all the previous studies on the association between Pb exposure and lung cancer had Pb workers (among battery, or smelter workers) as their subjects. Besides, some studies reported confusing and even contradictory results on the associations. The estimated risk for lung cancer, as is commonly agreed, were elevated in Pb-only workers compared with other workers [
25]. However, in two population-based case-control studies, investigators observed no increased risk of lung cancer with exposure to Pb compounds [
26]. The present study further confirmed the association of Pb exposure with lung cancer among residents, in addition to occupational workers. However, the findings are limited because other possible factors affecting lung cancer risks were not considered.
This study reported a blood Pb concentration of 24.22 μg/L in patients under 18 years old. Teenagers, especially children, in their growth period, are sensitive to Pb exposure, which impairs their cognitive function, mental behavior, infant growth, or even influences future socioeconomic status [
16,
27‐
29]. Guide issued by Ministry of Health to preventive measures against child-related high blood Pb levels and Pb poisoning set upper limit for children at 100 μg/L in China. However, Centers for Disease Control and Prevention (CDC) in America has set the upper limit for children at 50 μg/L in 2012. Therefore, we highly recommend that the critical value of blood Pb poisoning, especially for children, should be restricted to 50 μg/L in China.
Admittedly, this study has some limitations. Although we intended to investigate the current status of blood Pb levels all over China, there, to some extent, still existed flaws or loopholes in population selection. Those who have never tested their blood Pb concentration were not included in this study. The conclusion of the study representing whole population needs further verification. Furthermore, this study is not a case-control or cohort study in terms of methodology. Some conclusions, such as Pb as risk factor for lung cancer, need to be further verified in cohort study, or basic research. Finally, all data were limited in the HIS databases in the hospitals. Connections with patients’ other datasets outside hospitals will facilitate, and extend research of Pb exposure on human bodies.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.