| Coke oven emissions (COEs) are a mixture of complex components produced during destructive distillation of coal, containing particulate matters (PM) and volatile organic compounds, such as carbon monoxide, sulphur dioxide, benzene soluble matters, polycyclic aromatic hydrocarbons (PAHs), and heavy metals. PAHs and heavy metals are the main components in COEs. It was reported that 22% of the total global PAH emissions in 2004 were from China. The respiratory tract and digestive tract are the main pathways for PAHs into human body. The majority of PAHs in the atmosphere exist on the PM with aerodynamic diameter<5?m, and enter the deep parts of lung when breathing. Inhalation of PAHs-polluted air was demonstrated to be associated with elevated risk of pulmonary diseases, including asthma, chronic obstructive lung diseases (COPD), and lung cancer, contributing to 1.6% of the lung cancer morbidity rate in China. Heavy metals pollution is also one of the major public health problems worldwide, especially in the developing countries like China. The heavy metals, including arsenic (As), cadmium (Cd), chromium (Cr), nickel (Ni), and lead (Pb), can be absorbed into the human body mainly through inhalation, food, and drinking water. After entering into the human circulation system, heavy metals are widely distributed in bones, tissues, and organs, and they participate in the physiological and biochemical reactions or produce certain toxic effects. Most of the heavy metals absorbed were excreted through urine in a short time. It was reported that exposed to heavy metals could contribute to elevated oxidative stress and inflammation in the lung, generating tissue destruction, and increasing the risk of obstructive pulmonary diseases.Lung function parameters are the key indices to diagnose the lung diseases of COPD and asthma. The widely used lung function parameters include the forced vital capacity (FVC), forced expiratory volume in one second (FEV1), the percentage of predicted FVC (FVC%), the percentage of predicted FEV1 (FEV1%), FEV1/FVC ratio, and forced expiratory flow between 25% and 75% of vital capacity (FEF25.75). In accordance with the recommendations of American Thoracic Society (ATS) and European Respiratory Society (ERS), those who with FEV1/FVC<70% after inhaled bronchodilators are diagnosed as COPD. Epidemiological and laboratory studies showed that exposure to PAHs and heavy metals were associated with the occurrence and development of respiratory diseases. However, epidemiological studies, especially the longitudinal study, on the relationships of PAHs and heavy metals exposure with lung function decline are scarcely reported, and the underlying mechanisms on how PAHs and heavy metals affect the lung function remain unclear.Coke-oven workers are a typical occupational population exposed to high concentrations of PAHs and heavy metals. In the present study, we conducted a follow-up study of coke-oven workers. In the first part, we discussed the associations between the baseline urinary levels of twelve PAH metabolites [1-hydroxynaphthalene (1-OHNa), 2-OHNa,2-hydroxyfluorene (2-OHFlu),9-OHFlu,l-hydroxyphenanthrene (1-OHPh), 2-OHPh,3-OHPh,4-OHPh,9-OHPh,1-OHP,6-hydroxychrysene (6-OHChr), and 3-hydroxybenzo[a]pyrene (3-OHBaP)] and lung function decline. In the second part, we evaluated the associations between the baseline urinary levels of 5 heavy metals [arsenic (As), cadmium (Cd), chromium (Cr), nickel (Ni), and lead (Pb)] and lung function decline. The study aimed to clarify the dose-response relationships between PAHs and heavy metals exposure and the declines of lung function measurements among coke-oven workers, and also provided epidemiological support and effective clues for occupational protection of respiratory diseases among coke-oven workers.Part I The associations between polycyclic aromatic hydrocarbons exposure and lung function decline among coke-oven workersObjectives:This study aimed to investigate dose-response relationships of urinary PAH metabolites with lung function declines among coke-oven workers.Methods:We conducted healthy physical examinations among workers in coke-oven plant of Wuhan Iron and Steel Corporation in October 2010. A total of 1628 workers were participated in the study and signed the informed consent. A standardized questionnaire was administered to all subjects to obtain data on their demographic information, smoking status, drinking status, occupational history and history of diseases. The biological samples of the urine and blood samples were also collected. The follow-up survey of the population was conducted in October 2014, and 1243 of 1628 workers (follow-up rate of 76.4%) were successfully followed up. We detected the baseline (year 2010) urinary concentrations of 12 PAH metabolites (1-OHNa,2-OHNa,2-OHFlu,9-OHFlu,1-OHPh,2-OHPh,3-OHPh, 4-OHPh,9-OHPh,1-OHP,6-OHChr,3-OHBaP) by using Gas Chromatography-Mass Spectrometer (GC-MS). Their lung function measurements, including FVC, FEV1, FVC%, FEV1%, FEV1/FVC ratio, and FEF25-75 were also detected by using Chestgraph HI-101 in both baseline (2010) and follow-up study (2014). The dose-response relationships between single urinary PAH metabolite and declines in above lung function parameters were conducted using multivariate linear regression models, with adjustment for gender, working years, body mass index (BMI), waist-to-hip ratio, smoking status, drinking status, and physical activity. The false discovery rate (FDR) method was used to control for multiple comparisons and a FDR-adjusted P<0.\0 was considered noteworthy. We also classified the workers by the quartiles (25th,50th, and 75th percentiles) of each PAH metabolite, and workers belonged to the first quartile of each urinary PAH metabolite (?25th percentiles, Q1 subgroup) were used as the reference group. The multivariate linear regression models with adjustment for above confounders were used to investigate the associations. To estimate the most significant urinary OH-PAH on declines of FEV1/FVC ratio and FEF25-75, we included all ten urinary PAH metabolites in the linear regression models, and used the backward elimination procedure to retain PAH metabolites that were significantly associated with declines in lung function parameters, with adjustment for above confounders.Results:Compared to the lung function levels measured in the baseline study, among all subjects, office workers and coke-oven wokers, the FVC, FEV1, FVC%, FEV1% and FEF25.75 were all significantly decreased 4 years later (all P<0.05). No significant differences were found in FEV1/FVC ratio between the baseline and follow-up tests (all P>0.05). When using single OH-PAH and ?OH-PAHs in multivariate linear regression models, it was shown that each 10-fold increase in 1-OHNa and 2-OHNa generated 2.37% (95%CI:0.78%-3.96%) and 2.03%(95%CI:0.37%-3.66%) increased decline in FEV1/FVC, respectively; each 10-fold increase in 2-OHFlu and 9-OHFlu were associated with 2.28% (95%CI:0.85%-3.68%) and 1.24%(95%CI:0.55%-1.96%) elevated decline in FEV1/FVC, respectively; each 10-fold increase in 1-OHPh,2-OHPh, and EOH-PAHs generated 1.61% (95%CI:0.37%-2.86%),1.61%(95%CI:0.23%-2.97%), and 2.90%(95%CI:0.94%-4.84%) accelerated decline in FEV1/FVC, respectively. We did not observe urinary 3-OHPh, 4-OHPh,9-OHPh, and 1-OHP were associated with FEV1/FVC decline. None of these urinary PAH metabolites had a linear dose-dependent manner with declines in FEV1, FVC%, and FEV1%(all FDR P>0.10). Each 10-fold increase in 1-OHNa was associated with 214.42mL (95%CI:7.21-421.63mL) increased decline in FEF25-75; each 10-fold increase in 1-OHPh,2-OHPh, and 9-OHPh generated 264.94mL (95%CI: 103.85-426.00mL),205.05mL (95%CI:27.22-382.90mL), and 264.43mL (95%CI: 96.64-432.22mL) accelerated decline in FEF25-75, respectively; each 10-fold increase in 1-OHP and ?OH-PAHs generated 379.79mL (95%CI:172.53-587.04mL) and 292.91mL (95%CI:39.60-546.22mL) elevated decline in FEF25-75, respectively.We further classified the subjects by the quartiles of each PAH metabolite. After adjustment for potential confounders, workers within the Q4 subgroups of 1-OHNa (>3.02 ?g/mmol creatinine),2-OHNa (>2.77?g/mmol creatinine),2-OHFlu(>1.56?g/mmol creatinine),1-OHPh (>1.57?g/mmol creatinine),3-OHPh (>0.67?g/mmol creatinine), 9-OHPh (>1.35?g/mmol creatinine), and ?OH-PAHs (>19.31?g/mmol creatinine) had significantly greater declines in FEV1/FVC ratio than those within the Q1 subgroups of each PAH metabolite, respectively (all P<0.05 for Q4 vs. Q1). Additionally, workers with urinary 9-OHFlu>0.58?g/mmol creatinine had significantly higher decline in FEV1/FVC ratio than those with urinary 9-OHFlu?0.23?g/mmol creatinine (Q3 and Q4 vs. Q1 subgroup, both P=0.002). Significantly elevated trend in increasing decline in FEV1/FVC ratio was found from Q1 to Q4 subgroups of ?OH-PAHs and each PAH metabolite (all Ptrend<0.05), with exception of 1-OHP (Ptrend=0.182). In addition, workers with urinary 1-OHPh>0.87?g/mmol creatinine (Q3 and Q4) had significantly higher decline in FEF25-75 than those with urinary 1-OHPh<0.42?g/mmol creatinine (Q1)[?(95%CI)=292.68mL (75.34-510.02mL) and 314.74mL (98.88-530.60mL), respectively, P=0.008 for Q3 vs. Q1 and P=0.004 for Q4 vs. Q1]. Workers within the Q4 subgroup of 9-OHPh (>1.35?g/mmol creatinine) had significantly greater decline in FEF25.75 than those within the Q1 subgroup [?(95%CI)=415.89mL (200.00-631.79mL), P<0.001 for Q4 vs. Q1].Compared to workers with urinary 1-OHP?1.87?g/mmol creatinine (Q1), workers within the Q2, Q3, and Q4 of 1-OHP subgroups had significantly increased decline in FEF25-75 (P=0.022,0.002, and P<0.001, respectively). The significantly increased trends in decline of FEF25-75 were found from Q1 to Q4 quartile groups of 1-OHNa,1-OHPh,9-OHPh,1-OHP, and ?OH-PAHs (all Ptrend?0.05).We further stratified the whole subjects by smoking status and working sites, respectively. Among the nonsmokers, baseline urinary levels of 1-OHNa,2-OHNa, 2-OHFlu,9-OHFlu,1-OHPh, and ?OH-PAHs were associated with significant decline in FEV1/FVC (all ?>0 and FDR P?0.10); baseline urinary levels of 9-OHPh and 1-OHP were significantly associated with FEF25-75 decline [P(95%CI)=349.39mL (67.01-631.78mL) and 335.05mL (11.84-658.26 mL), respectively]. While among the smokers, urinary 2-OHFlu and 9-OHFlu were positively associated with decline in FEV1/FVC; urinary 9-OHFlu, 1-OHPh,2-OHPh,9-OHPh, and 1-OHP were found to be significantly associated with FEF25-75 decline (all ?>0 and FDR P?0.10). Since the atmosphere PAHs levels in difference working places were dramatically different, we classified the study participants according to their working sites:385 subjects (31.0%) working in the office areas were considered as office workers, while 858 workers (69.0%) who occupied adjacent to or on the top-, side-, and bottom of the coke-oven were referred as coke-oven workers. Among the coke-oven workers, the baseline urinary concentrations of 1-OHNa,2-OHFlu,9-OHFlu,1-OHPh, 2-OHPh, and ?OH-PAHs were positively associated with decline in FEV1/FVC (all ?>0 and FDR P?0.10); urinary 1-OHNa,9-OHFlu,1-OHPh,2-OHPh,9-OHPh,1-OHP and ?OH-PAHs were associated with significant decline in FEF25.75 (all ?>0 and FDR P?0.10). While among the office workers, only urinary 9-OHFlu was found to be associated with FEV1/FVC decline [?(95%CI)=1.15%(0.02%-2.30%)]; urinary 1-OHPh and 1-OHP were positively associated with decline in FEF25.75 [P(95%CI)=338.09mL (66.45-609.72mL) and 417.67mL (48.31-787.02mL), respectively].We used the backward elimination that included all ten individual PAH metabolites (In-transformed) in the multivariate regression models to estimate predominate PAH contributors to declines in FEVi/FVC ratio and FEF25-75, respectively. The baseline urinary levels of 2-OHFlu and 9-OHFlu were associated with significant decline in FEV1/FVC among all subjects [P(95%CI)=1.56%(0.04%-3.08%) and 0.93%(0.16%-1.69%), respectively], and only the urinary 9-OHFlu was the significant determiner for FEVi/FVC ratio decline among the smokers [P(95%CI)=1.20%(0.34%-2.06%)]. While the baseline urinary 1-OHP was the significant determiner for FEF25-75 decline among the whole subjects and smokers [P(95%CI)=451.24mL (228.04-674.45mL) and 554.92mL (233.32-876.51mL), respectively]. Among the nonsmokers, the baseline level of urinary 1-OHNa was the significant determiner for increasing decline in FEVi/FVC [p(95%CI)= 3.68% (1.17%-6.19%)];and urinary 9-OHPh was the significant determiner for decline in FEF25-75 [p(95%CI)=502.57mL (192.38-812.76mL)]. For coke-oven workers, the baseline urinary 2-OHFlu level was the significant determiner for decline in FEV1/FVC [p(95%CI)=3.01% (1.25%-4.77%)]; and 2-OHPh was the significant determiner for decline in FEF25-75 [p(95%CI)=592.75mL (288.42-897.07mL)]. For office workers, the urinary 9-OHFlu was the significant determiner for FEV1/FVC decline [p(95%CI)=1.12%(0.01%-2.24%)], and 1-OHPh was the significant determiner for FEF25.75 decline [P(95%CI)=342.11mL (75.21-609.01mL)].Conclusions:In the present study, we found that the baseline exposure levels of 1-OHNa, 2-OHNa,2-OHFlu,9-OHFlu,1-OHPh,2-OHPh and ?OH-PAHs could lead to significantly elevated 4-year decline in FEV1/FVC ratio among coke-oven workers, while these effects were most obvious among workers with the highest levels of the above urinary PAH metabolites. The baseline levels of urinary 1-OHNa,1-OHPh,2-OHPh,9-OHPh,1-OHP, and EOH-PAHs were associated with significantly elevated decline in FEF25-75. When using backward selection method to adjustment for 10 urinary PAH metabolites, we found that the most significant determiners for FEV1/FVC decline were 2-OHFlu and 9-OHFlu for the whole subjects,2-OHFlu for workers who working adjacent to or on the coke-oven, and 9-OHFlu for smokers and office workers, respectively. The baseline urinary 1-OHP was the significant determiner for FEF25.75 decline among the whole subjects and smokers. The results indicated that long-term exposure to high concentrations of occupational PAHs could cause decrease in pulmonary function among coke-oven workers, and suggested that urinary fluorene metabolites were the most significant determiners for FEVi/FVC decline and urinary 1-OHP was the most significant determiner for FEF25.75 decline. The findings provide epidemiological evidences for the prevention and control of respiratory diseases among coke-oven workers.Part II The associations between heavy metals exposure and lung function decline among coke-oven workersObjectives:This study aimed to investigate quantitative relationships of urinary heavy metals with lung function declines among coke-oven workers.Methods:A total of 1243 coke-oven workers, who were enrolled in the baseline study (2010) and successfully follow-up in 2014, were involved in the study. The concentrations of baseline urinary As, Cd, Cr, Ni, and Pb among the cohort workers were detected by Inductively Coupled Plasma Mass Spectrometer (ICP-MS). We then conducted multivariate linear regression models to estimate the dose-response relationships between the baseline urinary concentration of above 5 heavy metals and declines in lung function parameters, with adjustment for gender, working years, body mass index (BMI), waist-to-hip ratio, smoking status, drinking status, physical activity and ?OH-PAHs. The false discovery rate (FDR) method was used to control for multiple comparisons and a FDR-adjusted P<0.10 was considered noteworthy.We also classified the workers by the quartiles (25th,50, and 75th percentiles) of each heavy metal into four subgroups (Q1,Q2, Q3, and Q4 subgroups), and workers belonged to the first quartile of each urinary heavy metal (?25th percentiles, Q1 subgroup) were used as the reference group. The multivariate linear regression models with adjustment for above confounders were used to investigate the associations.Results:We found that each 10-fold increase in baseline urinary As generated 205.87mL (95%CI:77.92-333.81mL) increased decline in FVC,165.92mL (95%CI:68.73-263.12mL) increased decline in FEV1,4.42%(95%CI:0.69%-8.13%) increased decline in FVC%, and 4.05%(95%CI:1.13%-6.95%) increased decline in FEV1%; each 10-fold increase in urinary Cr was associated with 96.71mL (95%CI:13.45-179.99mL) accelerated decline in FEV1; each 10-fold increase in urinary Ni generated 88.70mL (95%CI:14.39-163.OOmL) elevated decline in FVC and 71.27mL (95%CI:14.83-127.72mL) elevated decline in FEV1; each 10-fold increase in urinary Pb generated 153.74mL (95%CI:58.81-248.68mL) increased decline in FEV1,and 3.15%(95%CI:0.28%-6.03%) increased decline in FEV1%.We further stratified the study participants by their working sites. Among the workers working adjacent to or on the coke-oven, each 10-fold increase in baseline urinary As was associated with 243.54mL (95%CI:77.76-409.33mL) increased decline in FVC,186.35mL (95%CI:62.22-310.48mL) increased decline in FEV1,6.06%(95%CI:1.31%-10.82%) increased decline in FVC%, and 5.99%(95%CI:2.37%-9.60%) increased decline in FEV1%; each 10-fold increase in urinary Pb generated 153.49mL (95%CI:28.18-278.82mL) accelerated decline in FEV1, however, we did not find significant associations between the baseline urinary concentrations of Cd, Cr and Ni and declines in above lung function measurements (all FDR P>0.10) among the workers working adjacent to or on the coke-oven. Among the office workers, none of these 5 urinary heavy metals were shown to have significant associations with declines in FVC, FEV1, FVC%, FEV1%,FEV1/FVC, and FEF25.75 (all FDR P>0.10). As a result, in the following analyses, we mainly focused on investigating the effects of heavy metals on lung function decline among workers working adjacent to or on the coke-oven.For the workers working adjacent to or on the coke-oven (coke-oven workers), we further classified these subjects by the quartiles of each urinary heavy metal. After adjustment for potential confounders, workers within the Q4 subgroups of As (>4.817 ?g/mmol creatinine) had significantly greater declines in FVC, FEV1,FVC%, and FEV1% than those within the Q1 subgroup, respectively (all P<0.05 for Q4 vs. Q1, Ptrend<0.02). Additionally, workers with urinary Pb>0.662?g/mmol creatinine (Q4) had a significantly higher decline in FEVi than those with urinary Pb<0.269?g/mmol creatinine (Q1) [P(95%CI)=122.52mL (2.46-242.58mL), P=0.046 for Q4 vs. Q1]. We further stratified the coke-oven workers by smoking status. Among the smokers, each 10-fold increase in baseline urinary As generated 282.85mL (95%CI:69.38-496.32mL) accelerated decline in FVC,185.93mL (95%CI:25.31-346.54mL) accelerated decline in FEV,,6.52%(95%CI: 0.62%-12.43%) accelerated decline in FVC%, and 6.42%(95%CI:1.96%-10.91%) accelerated decline in FEV1%; each 10-fold increase in Cd was associated with 4.67% (95%CI:0.46%-8.89%) increased decline in FEV1; while among nonsmokers, none of the five heavy metals had significant associations with FVC, FEV1, FVC%, and FEV1%(all FDR P>0.10).Conclusions:In the present study, we found that the baseline exposure levels of As could lead to significantly elevated declines in lung function parameters (FVC, FEV1,FVC% and FEV1%) among coke-oven workers, while these effects were most obvious among workers with the highest levels of As and smokers. There was a significant association between baseline urinary Pb and elevated decline in FEV1. Among smokers, we also demonstrated significant association between urinary Cd concentrations and increased decline in FEV1%. These results indicated that long-term exposure to high concentrations of As and Pb could cause significant lung function decline among coke-oven workers, and among smokers the baseline urinary levels of As and Cd had a linear dose-dependent manner with declines in lung function indices, suggesting that heavy metals exposure could generate lung injure at an early stage and smoking could enhance this effect.In summary, the study aimed to investigate dose-response relationships of urinary PAHs and heavy metals exposure with lung function declines among coke-oven workers. The results showed that the baseline levels of urinary OH-PAHs were associated with significantly increased declines in FEV1/FVC and FEF25-75;the baseline As concentrations had significant associations with accelerated declines in FVC, FEV1, FVC%, and FEV1%; the baseline Pb level was associated with significantly deeper decline in FEV1. |
PDF Full Text Request