Associations Between Air Quality Changes and Biomarkers of Systemic Inflammation During the 2014 Nanjing Youth Olympics: A Quasi-Experimental Study

Transcription

1 American Journal of Epidemiology The Author Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please Vol. 185, No. 12 DOI: /aje/kww209 Advance Access publication: April 28, 2017 Original Contribution Associations Between Air Quality Changes and Biomarkers of Systemic Inflammation During the 2014 Nanjing Youth Olympics: A Quasi-Experimental Study Huichu Li, Lian Zhou, Cuicui Wang, Renjie Chen, Xiaoying Ma, Bin Xu, Lilin Xiong, Zhen Ding, Xiaodong Chen, Yun Zhou, Yan Xu*, and Haidong Kan* * Correspondence to Dr. Haidong Kan, Department of Environmental Health, School of Public Health, Fudan University, P.O. Box 249, 130 Dong-An Road, Shanghai , China ( haidongkan@gmail.com); or Yan Xu, Jiangsu Provincial Center for Disease Prevention and Control, 172 Jiangsu Road, Nanjing , China ( cdcxy@vip.sina.com). Initially submitted January 19, 2016; accepted for publication April 27, There is increasing interest in quasi-experimental research to evaluate whether actions taken to improve air quality will benefit public health. We conducted a quasi-experimental study to evaluate inflammatory response to changes in air quality during the 2014 Nanjing Youth Olympics in China. We repeatedly measured 8 biomarkers of systemic inflammation in 31 healthy adults and obtained hourly air pollutant concentrations from a nearby fixedsite monitoring station. We used linear mixed-effect models to examine the associations between air quality changes and blood biomarkers. Air pollutant concentrations decreased apparently during the Youth Olympics. Concomitantly, we observed significant decreases in levels of soluble cluster of differentiation 40 (CD40) ligand and interleukin 1β (geometric means ratios were 0.45 and 0.24, respectively) from the pre-olympic period to the intra-olympic period. Afterwards, levels of C-reactive protein and vascular cell adhesion molecule 1 increased significantly (geometric means ratios were 2.22 and 1.29, respectively) in the post-olympic period. Fine particulate matter and ozone were significantly associated with soluble CD40 ligand, P-selectin, interleukin 1β, intercellular adhesion molecule 1, and vascular cell adhesion molecule 1. Other pollutants showed positive but nonsignificant associations. Our study indicated that reduced air pollution, especially fine particulate matter and ozone, during the 2014 Nanjing Youth Olympics was associated with alleviated systemic inflammation in healthy adults. air pollution; cardiovascular biomarkers; Olympics; systemic inflammation; quasi-experimental study Abbreviations: CD40, cluster of differentiation 40; CI, confidence interval; CRP, C-reactive protein; GMR, geometric means ratio; ICAM-1, intercellular adhesion molecule 1; IL-1β, interleukin 1β; MCP-1, monocyte chemotactic protein 1; PM 2.5, particulate matter with an aerodynamic diameter less than or equal to 2.5 μm; scd40l, soluble CD40 ligand; TNF-α, tumor necrosis factor α; VCAM-1, vascular adhesion molecule 1. Ambient air pollution has been associated with adverse cardiovascular events in a number of epidemiologic studies (1 3). However, the biological pathways responsible for the hazardous effects of air pollutants have not been well characterized. Recently, many epidemiologic studies have suggested systemic inflammatory response as one of the main mechanisms underlying the associations (4, 5), but the results have varied according to selected biomarkers, target population, and air pollution characteristics (6 10). Furthermore, most of the existing findings were obtained from observational studies, which were subject to residual confounding due to a number of individual characteristics and environmental factors. Nevertheless, it was almost impractical to perform a complete experimental trial to establish a link between air pollution and health outcomes. Alternatively, there is increasing interest in utilizing a quasi-experimental design to examine the health impacts of air pollutants or an air quality intervention, in which subjects are not randomly assigned to exposure conditions (11 15). However, evidence from highly polluted developing countries, such as China, is still limited. Facing severe air pollution problems, the Chinese government is making great efforts to improve air quality. Around 1290

2 Air Quality Improvement and Health Effects 1291 the times of some important events (Olympic Games, etc.), temporary actions implementing unprecedented air pollution control measures have provided an opportunity to investigate the biological response to drastic changes in air quality using the quasi-experimental method. Few studies have evaluated the association of health improvements with these air quality interventions. A prior investigation around the time of the 2008 Beijing Olympics showed that air quality improvement resulted in appreciable reductions in biomarkers of inflammation in a group of healthy young adults (16). This finding was important, but it still needed confirmation in more locations and populations in China. Therefore, we conducted a quasi-experimental study to evaluate the associations between air quality changes and levels of proinflammatory biomarkers in a panel of healthy middle-aged adults during the 2014 Nanjing Youth Olympic Games. METHODS Pollution control measures The 2014 Youth Olympic Games were held in Nanjing, China, from August 16 to August 28, Strict air pollution control measures were implemented from August 1 to September 6, 2014, in Nanjing and its 22 peripheral cities. These measures included the following: 1) 3,119 construction projects were suspended; 2) 554 major industrial factories were temporarily shut down, and 741 factories were ordered to curtail production; 3) 162 coal-fired power plants were ordered to use low-sulfur coal; 4) motor vehicles with high emission levels were banned from entering Nanjing; and 5) open burning (biomass, etc.) was prohibited in both urban and rural areas. The above measures were implemented under intensive supervision. Study design and participants We planned to recruit healthy nonsmoking participants from a community health center in Jiangning District, Nanjing. We included persons who were middle-aged and resided within a 5-km radius of the health center. We enrolled 31 participants in this study; all were physicians. This sample size was similar to that in our previous intervention study on air pollution (17). We adopted a longitudinal panel design with 1 clinical visit in the pre-olympic period (July 15 August 1), 2 visits in the intra-olympic period (August 6 August 31), and 2 visits in the post-olympic period (September 6 September 30), according to the schedule of pollution control measures. The 5 repeated clinical visits were separated by at least 2 weeks. We collected fasting blood samples at the same time of day (8:00 AM 10:00 AM)toreduce the influences of circadian variations and dietary patterns. Baseline data, including information on sex, age, history of disease, income, height, and body weight, were collected at enrollment. The institutional review board at the Jiangsu Provincial Center for Disease Control and Prevention approved this study. All participants provided written informed consent at enrollment. Biomarker test Peripheral blood samples (5 ml) were drawn by a nurse, centrifuged for serum extraction, and stored at 80 C within 30 minutes. We measured serum levels of 8 inflammatory biomarkers that have been used to predict cardiovascular risk and have been associated with air pollution in observational studies (18). These biomarkers included soluble cluster of differentiation 40 (CD40) ligand (scd40l), monocyte chemotactic protein 1 (MCP-1), tumor necrosis factor α (TNF-α), C-reactive protein (CRP), intercellular adhesion molecule 1 (ICAM-1), P-selectin, vascular cell adhesion molecule 1 (VCAM-1), and interleukin 1β (IL-1β). We measured the biomarkers using enzyme-linked immunosorbent assays or the Millipore MILLIPLEX MAP human cytokine chemokine kit (Millipore Corporation, Billerica, Massachusetts). All biomarker tests were conducted following the manufacturers instructions. Air pollution measurement We obtained hourly air quality data for the period July 15 September 30, 2014, from a fixed-site monitoring station about 5 km away from the community health center, in the same district. This monitoring station is regulated by the Nanjing Environmental Protection Bureau and is not in the direct vicinity of industrial, traffic, or other emission sources, suggesting that its measurements may well represent background air pollution levels in this district. Methods based on tapered element oscillating microbalance, ultraviolet fluorescence, chemiluminescence, light absorbance, and ultraviolet fluorescence were used to measure levels of particulate matter with an aerodynamic diameter less than or equal to 2.5 μm (PM 2.5 ), sulfur dioxide, nitrogen dioxide, carbon monoxide, and ozone. We calculated daily 24-hour average concentrations of PM 2.5,sulfurdioxide, nitrogen dioxide, and carbon monoxide and maximum 8-hour average (10:00 AM 6:00 PM) concentrations of ozone during the study period. To allow for adjustment for weather conditions, we collected data on daily ambient temperature and relative humidity from a meteorological station located in the district. Statistical analysis Data for all biomarkers were naturally log-transformed before statistical analyses because their distributions were almost right-skewed. We used a linear mixed-effect model to evaluate the associations between air quality and biomarkers (16). This model utilizes intra- and interindividual variations and allows adjustment for correlation between repeated measurements within each subject. We firstly evaluated the differences in the levels of each biomarker from the pre- Olympic period to the intra-olympic period and from the intra-olympic period to the post-olympic period, using the biomarker concentrations as the response variables and periods as the explanatory dummy variables in the models. For example, 0 refers to the intra-olympic period, 1 refers to the pre-olympic period, and 2 refers to the post-olympic period. We thereafter applied the linear mixed-effect model to investigate the associations between continuous exposure to various air pollutants (as fixed-effect variables) and

3 1292 Li et al. biomarkers in single-pollutant models. In all models, we incorporated several sociodemographic variables (sex, age, body mass index, and educational level), day of the week, and weather conditions (temperature and relative humidity) as fixed-effect terms. Finally, we added a random intercept for each participant to account for the autocorrelations in both air pollution exposure and health measurements during the study period. We evaluated the associations between air pollutant concentrations and each biomarker in different time windows (0 6 hours, 7 12 hours, hours, 1 day (25 48 hours), 2 days (49 72 hours), and 0 7 days) prior to blood drawing. We introduced the 7-day moving averages of weather variables into the models to adjust for their potentially lagged confounding effects. We did not control for time trends, because all clinical follow-ups were completed within 3 months. As a sensitivity analysis, we fitted 2-pollutant models to evaluate the robustness of our results. All statistical tests were 2-sided, and a P value less than 0.05 was considered statistically significant. All analyses were conducted in R software, version (R Foundation for Statistical Computing, Vienna, Austria) with the package lme4. All estimates are presented as ratios of geometric mean values (geometric means ratios) and 95% confidence intervals for a biomarker associated with alterations in the time period, a 10-μg/m 3 change in PM 2.5, sulfur dioxide, nitrogen dioxide, and ozone levels, or a 1-mg/m 3 change in carbon monoxide level. RESULTS Descriptive results The subjects were 46 years of age, on average, with a standard deviation of 5 years. No one missed any scheduled follow-up visits, and all subjects claimed to have stayed in the district for at least 1 week before the scheduled blood drawing. Table 1 gives descriptive statistics for air quality and weather conditions during the pre-olympic, intra-olympic, and post-olympic periods. There were no missing data on air pollution measurements. Generally, we observed drastic changes in PM 2.5 and ozone levels but modest changes in concentrations of sulfur dioxide, nitrogen dioxide, and carbon monoxide. From the pre-olympic period to the intra- Olympic period, levels of PM 2.5 and ozone decreased by 30.4% and 52.7%, respectively. Furthermore, from the intra-olympic period to the post-olympic period, levels of the 2 pollutants increased by 2.1% and 235.4%, respectively. Notably, sulfur dioxide levels increased by 39.5% during the post-olympic period as compared with the intra- Olympic period. PM 2.5 was weakly correlated with sulfur dioxide, nitrogen dioxide, and ozone (Pearson s r ) and moderately correlated with carbon monoxide (r = 0.53). Table 2 shows geometric mean levels of the 8 biomarkers in the pre-, intra-, and post-olympic periods. The levels of scd40l, MCP-1, CRP, VCAM-1, and IL-1β were lowest in the intra-olympic period. Regression results We used the linear mixed-effect model to evaluate the differences between biomarker levels in the 3 periods. As shown in Figure 1A, the levels of scd40l, MCP-1, CRP, P-selectin, VCAM-1, and IL-1β decreased in the intra-olympic period compared with the pre-olympic period, but only the reductions in scd40l (geometric means ratio (GMR) = 0.45, 95% confidence interval (CI): 0.25, 0.81) and IL-1β(GMR = 0.24, 95% CI: 0.06, 0.93) were statistically significant. Comparing the post-olympic period with the intra-olympic period (Figure 1B), we found significant increases in CRP (GMR = 2.22, 95% CI: 1.20, 4.09) and VCAM-1 (GMR = 1.29, 95% CI: 1.05, 1.60) levels, but the increases in scd40l, MCP-1, and IL-1β were not statistically significant. Levels of TNF-α and ICAM-1 did not change in the direction expected, but the between-period differences were not statistically significant. We further investigated the associations between air pollutant concentrations and levels of inflammatory biomarkers Table 1. Air Pollutant Concentrations and Weather Conditions (Mean Level (Standard Deviation) and Mean Difference) During Various Periods Around the Time of the Youth Olympics, Nanjing, China, 2014 All (July 15 September 30) Pre-Olympic (July 15 August 1) Time Period Intra-Olympic (August 6 August 31) Post-Olympic (September 6 September 31) Difference Intra-Olympic Period Pre-Olympic Period Post-Olympic Period Intra- Olympic Period) Mean 95% CI Mean 95% CI PM 2.5, μg/m (19.3) 53.6 (24.5) 37.3 (12.5) 38.1 (18.5) , , 9.6 Sulfur dioxide, μg/m (7.2) 15.2 (4.1) 12.9 (3.7) 18.1 (10.5) , , 9.5 Nitrogen dioxide, μg/m (7.2) 22.0 (6.6) 24.1 (5.9) 27.0 (8.1) , , 6.9 Ozone, μg/m (25.4) 40.6 (14.9) 19.2 (15.1) 64.4 (18.3) , , 54.6 Carbon monoxide, mg/m (0.2) 1.1 (0.2) 1.1 (0.2) 1.1 (0.2) , , 0.1 Temperature, C 25.3 (3.1) 28.8 (2.5) 24.9 (2.0) 23.2 (2.1) , , 0.6 Relative humidity, % 75.8 (6.6) 73.9 (6.8) 78.8 (5.4) 73.8 (6.6) , , 1.6 Abbreviations: CI, confidence interval; PM 2.5, particulate matter with an aerodynamic diameter less than or equal to 2.5 μm.

4 Air Quality Improvement and Health Effects 1293 Table 2. Geometric Mean Levels (Standard Deviation) of Biomarkers of Systemic Inflammation in 31 Healthy Adults During Various Periods Around the Time of the Youth Olympics, Nanjing, China, 2014 Time Period Biomarker Pre-Olympic Intra-Olympic Post-Olympic scd40l, ng/ml 4.23 (1.89) 3.38 (2.35) 3.60 (2.67) MCP-1, ng/ml 0.49 (0.25) 0.48 (0.25) 0.50 (0.22) TNF-α, pg/ml (10.66) (14.17) 9.84 (11.91) CRP, μg/ml 1.43 (7.17) 1.28 (6.56) 1.34 (4.19) ICAM-1, ng/ml (40.94) (34.70) (37.38) P-selectin, ng/ml (28.55) (18.32) (23.98) VCAM-1, ng/ml (118.69) (113.40) (108.97) IL-1β, pg/ml 0.07 (0.12) 0.05 (0.08) 0.06 (0.07) Abbreviations: CD40, cluster of differentiation 40; CRP, C-reactive protein; ICAM-1, intercellular adhesion molecule 1; IL-1β, interleukin 1β; MCP-1, monocyte chemotactic protein 1; scd40l, soluble CD40 ligand; TNF-α, tumor necrosis factor α; VCAM-1, vascular adhesion molecule 1. using the same models. Among various pollutants, we found significant associations only for PM 2.5 and ozone. As for the lag structure, the associations were generally strongest at lags of hours or 1 day. scd40l, ICAM-1, P-selectin, VCAM-1, and IL-1β were significantly associated with PM 2.5 at a lag of 1 day and with ozone at all lags examined. Nevertheless, we did not find any statistically significant associations for MCP-1, TNF-α, and CRP at any lag. As A) Biomarker scd40l MCP-1 TNF-α CRP ICAM-1 P-selectin VCAM-1 GMR (95% CI) 0.45 (0.25, 0.81) 0.95 (0.82, 1.11) 1.06 (0.85, 1.33) 0.40 (0.15, 1.06) 1.18 (0.83, 1.68) 0.74 (0.50, 1.08) 0.86 (0.61, 1.21) Web Figure 1 (available at shows, a 10-μg/m 3 increase (1-day lag) in PM 2.5 was associated with increases in levels of scd40l (GMR = 1.25, 95% CI: 1.14, 1.37), P-selectin (GMR = 1.08, 95% CI: 1.01, 1.14), and IL-1β (GMR = 1.32, 95% CI: 1.07, 1.63). Web Figure 2 shows that a 10-μg/m 3 increase (1-day lag) in ozone was significantly associated with increases in scd40l (GMR = 1.17, 95% CI: 1.10, 1.24), ICAM-1 B) Biomarker scd40l MCP-1 TNF-α CRP ICAM-1 P-selectin VCAM-1 GMR (95 %CI) 1.24 (0.85, 1.81) 1.01 (0.92, 1.11) 0.91 (0.79, 1.04) 2.22 (1.20, 4.09) 0.82 (0.67, 1.00) 0.81 (0.64, 1.03) 1.29 (1.05, 1.60) IL-1β 0.24 (0.06, 0.93) IL-1β 2.14 (0.95, 4.84) Geometric Means Ratio Geometric Means Ratio Figure 1. Geometric means ratios (GMRs) for biomarkers of systemic inflammation in 31 healthy adults around the time of the Youth Olympics, Nanjing, China, A) Changes in inflammatory biomarker levels in the intra-olympic period compared with the pre-olympic period; B) changes in inflammatory biomarker levels in the post-olympic period compared with the intra-olympic period. CD40, cluster of differentiation 40; CRP, C-reactive protein; ICAM-1; intercellular adhesion molecule 1; IL-1β, interleukin 1β; MCP-1, monocyte chemotactic protein 1; scd40l, soluble CD40 ligand; TNF-α, tumor necrosis factor α; VCAM-1, vascular adhesion molecule 1. Bars, 95% confidence intervals (CIs).

5 1294 Li et al. (GMR = 1.05, 95% CI: 1.01, 1.08), P-selectin (GMR = 1.07, 95% CI: 1.02, 1.11), and VCAM-1 (GMR = 1.06, 95% CI: 1.02, 1.09) levels. Other pollutants showed generally positive but nonsignificant associations with biomarker levels (data not shown). We further evaluated the associations between air pollutants and inflammatory biomarkers using 2-pollutant models in a sensitivity analysis. As Web Figure 3 shows, the associations of PM 2.5 with P-selectin and IL-1β became statistically insignificant after adjustment for sulfur dioxide, nitrogen dioxide, carbon monoxide, and ozone, while the associations with scd40l remained significant in all 2-pollutant models other than the models with carbon monoxide. Similarly, the association between ozone and scd40l was almost unchanged when copollutants other than carbon monoxide and PM 2.5 were controlled for, whereas the associations with ICAM-1, P-selectin, and VCAM-1 turned to be insignificant after controlling for copollutants (Web Figure 4). Other air pollutants were still not significantly associated with any biomarkers in 2-pollutant models (data not shown). DISCUSSION Drastic variation in air quality around the time of the 2014 Nanjing Youth Olympics provided us with a unique opportunity to conduct this quasi-experimental study. We observed significant changes in levels of several inflammatory biomarkers, including scd40l, IL-1β, CRP, and VCAM-1, in the pre-olympic, intra-olympic, and post-olympic periods. Moreover, we found that among various air pollutants, PM 2.5 and ozone were significantly associated with levels of scd40l, ICAM-1, VCAM-1, P-selectin, and IL-1β. Our findings contribute to the scientific knowledge that shortterm alleviation of air pollution can result in appreciable health benefits. Globally, there have been other air pollution reductions resulting from governmental interventions. These improvements in air quality have been shown to lead to significant public health benefits, including decreases in total or causespecific mortality and decreases in children s hospital admissions for respiratory diseases and asthma (11 13, 17). Recently, researchers have investigated the biological pathways underlying these intervention-associated health benefits (16, 19). For example, in a quasi-experimental study carried out among healthy young people around the time of the 2008 Beijing Olympics, Rich et al. (16) observed significant decreases in levels of inflammatory biomarkers, including CRP, P-selectin, scd40l, and white blood cell counts, in accordance with reduced air pollution levels, but only the change in P-selectin was statistically significant. Our study identified more inflammatory biomarkers (scd40l, IL-1β, CRP, ICAM-1, and VCAM-1) whose levels were decreased in association with air quality improvement in a more sensitive group of participants (middle-aged adults) during the 2014 Nanjing Youth Olympics. It is important, from both scientific and regulatory standpoints, to determine the specific components of the air pollution mixture responsible for the observed associations. Following the implementation of air quality interventions during the Nanjing Youth Olympics, we observed drastic reductions in PM 2.5 and ozone concentrations but modest changes in levels of sulfur dioxide, nitrogen dioxide, and carbon monoxide. Although levels of most of the biomarkers we examined were reduced during the Olympics, the biomarkerpollutant associations varied considerably in magnitude and direction in single-pollutant analyses. We found that only the reductions in PM 2.5 and ozone were significantly associated with decreases in levels of inflammatory biomarkers. Therefore, we speculate that reductions in PM 2.5 and ozone might be mainly responsible for the health benefits due to air quality improvement during the Nanjing Youth Olympics. However, it was difficult to disentangle the associations between individual air pollutants and inflammation biomarkers because of the correlations among pollutants and the complicated implementation of pollution controls. In this quasi-experimental study, ambient concentrations of PM 2.5 and ozone were significantly associated with serum levels of scd40l, ICAM-1, P-selectin, VCAM-1, and IL-1β. All of these biomarkers have been previously associated with short-term changes in air pollution in both observational studies and experimental trials, although there is heterogeneity between various studies (6, 7, 17, 20). For example, in an elderly cohort, Bind et al. (21) found that CRP and ICAM-1 levels were significantly associated with 24-hour average ozone, and levels of VCAM-1 and ICAM-1 were associated with PM 2.5 at the same lag. In a controlled exposure experiment, Devlin et al. (22) demonstrated significant increases in CRP, interleukins, and TNF-α after exposure to ozone at 0.3 parts per million for 2 hours. In contrast, another experiment in Toronto, Ontario, Canada, revealed nonsignificant increases in CRP, TNF-α, and IL-1β after ozone and PM exposure (23). Few experimental or quasi-experimental studies have evaluated the independent associations between air pollutants and health outcomes using multipollutant models. In our study, we found a robust association of PM 2.5 and ozone with scd40l in 2-pollutant models, suggesting that this biomarker may play an important role in biological responses to PM 2.5 and ozone. Our results may have important biological relevance. A number of epidemiologic studies have consistently supported PM 2.5 and ozone as risk factors for cardiovascular disease, but the underlying mechanism has not been well established (24, 25). A systemic inflammation response is considered one of the most important pathways. Briefly, the inhalation of PM 2.5 and ozone may initiate inflammatory reaction cascades, which can in turn result in the abundant release of acute-phase protein, adhesion molecules, and cytokines and consequently lead to injuries to multiple tissues (26). Our results showed that PM 2.5 and ozone were consistently and robustly associated with increased levels of scd40l, a key protein involved in inflammation and coagulation response that is important in the development of cardiovascular disease (17). One advantage of this study was the use of a quasiexperimental design, which is efficient for causal inference. This design can also provide valuable insights into the mechanisms of air pollution by minimizing the threats to ecological validity and by facilitating the generalizability of study results through natural experiments. Second, we examined

6 Air Quality Improvement and Health Effects inflammatory biomarkers, allowing for a relatively systematic evaluation of the inflammatory response associated with a change in air pollution. Third, all of the participants were healthy, middle-aged nonsmokers who were engaged in the same occupation, had similar levels of physical activity, and donated their fasting blood samples, which may reduce confounding effects when exploring the associations between air pollution and blood biomarkers. Two important limitations should be mentioned. Firstly, exposure measurement errors are inevitable because we used fixed-site monitoring data as surrogates for personal exposure to air pollution. This kind of exposure measurement error was almost randomly distributed (averaging out to zero) and would have biased the effect estimates towards the null (27). Secondly, the sample size was relatively small, which may add uncertainty and attenuate the generalizability of our results. In summary, implementation of air pollution control measures during the 2014 Nanjing Youth Olympics may have led to significant decreases in blood levels of several inflammation biomarkers in a panel of healthy, middle-aged adults. Furthermore, this quasi-experimental study demonstrated that among various air pollutants, PM 2.5 and ozone were significantly associated with reductions in levels of inflammatory biomarkers, especially scd40l. Our findings provide robust mechanistic evidence that short-term improvements in air quality would result in alleviation of systemic inflammation. They provide further encouragement for governmental efforts to control air pollution in heavily polluted countries such as China. ACKNOWLEDGMENTS Author affiliations: School of Public Health, Key Laboratory of Public Health Safety of the Ministry of Education and Key Laboratory of Health Technology Assessment of the Ministry of Health, Fudan University, Shanghai, China (Huichu Li, Cuicui Wang, Renjie Chen, Haidong Kan); Jiangsu Provincial Center for Disease Control and Prevention, Nanjing, China (Lian Zhou, Xiaoying Ma, Bin Xu, Zhen Ding, Xiaodong Chen, Yan Xu); School of Applied Meteorology, Nanjing University of Information Science and Technology, Nanjing, China (Lian Zhou); Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP 3 ), Fudan University, Shanghai, China (Renjie Chen, Haidong Kan); Nanjing Municipal Center for Disease Control and Prevention, Nanjing, China (Lilin Xiong); and Jiangning District Center for Disease Control and Prevention, Nanjing, China (Yun Zhou). H.L. and L.Z. contributed equally to this work. This work was supported by the Public Welfare Research Program of the National Health and Family Planning Commission of China (grants and ), the National Basic Research Program of China (973 Program; grant 2012CB955502), the China Medical Board Collaborating Program (grant ), and the Cyrus Tang Foundation (grant CTF-FD ). Conflict of interest: none declared. REFERENCES 1. Wu S, Deng F, Niu J, et al. Association of heart rate variability in taxi drivers with marked changes in particulate air pollution in Beijing in Environ Health Perspect. 2010;118(1): Mustafic H, Jabre P, Caussin C, et al. Main air pollutants and myocardial infarction: a systematic review and meta-analysis. JAMA. 2012;307(7): Shah ASV, Langrish JP, Nair H, et al. Global association of air pollution and heart failure: a systematic review and metaanalysis. Lancet. 2013;382(9897): Brook RD, Franklin B, Cascio W, et al. Air pollution and cardiovascular disease: a statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation. 2004;109(21): Franchini M, Mannucci PM. Short-term effects of air pollution on cardiovascular diseases: outcomes and mechanisms. J Thromb Haemost. 2007;5(11): Johannesson S, Andersson EM, Stockfelt L, et al. Urban air pollution and effects on biomarkers of systemic inflammation and coagulation: a panel study in healthy adults. Inhal Toxicol. 2014;26(2): Rückerl R, Hampel R, Breitner S, et al. Associations between ambient air pollution and blood markers of inflammation and coagulation/fibrinolysis in susceptible populations. Environ Int. 2014;70: Rückerl R, Greven S, Ljungman P, et al. Air pollution and inflammation (interleukin-6, C-reactive protein, fibrinogen) in myocardial infarction survivors. Environ Health Perspect. 2007;115(7): Chuang KJ, Chan CC, Su TC, et al. The effect of urban air pollution on inflammation, oxidative stress, coagulation, and autonomic dysfunction in young adults. Am J Respir Crit Care Med. 2007;176(4): Mirowsky JE, Peltier RE, Lippmann M, et al. Repeated measures of inflammation, blood pressure, and heart rate variability associated with traffic exposures in healthy adults. Environ Health. 2015;14(1): Peel JL, Klein M, Flanders WD, et al. Impact of Improved Air Quality During the 1996 Summer Olympic Games in Atlanta on Multiple Cardiovascular and Respiratory Outcomes. (Research report no. 148). Boston, MA: Health Effects Institute; Clancy L, Goodman P, Sinclair H, et al. Effect of air-pollution control on death rates in Dublin, Ireland: an intervention study. Lancet. 2002;360(9341): Hedley AJ, Wong CM, Thach TQ, et al. Cardiorespiratory and all-cause mortality after restrictions on sulphur content of fuel in Hong Kong: an intervention study. Lancet. 2002;360(9346): Bayer-Oglesby L, Grize L, Gassner M, et al. Decline of ambient air pollution levels and improved respiratory health in Swiss children. Environ Health Perspect. 2005;113(11): Wong CM, Lam TH, Peters J, et al. Comparison between two districts of the effects of an air pollution intervention on bronchial responsiveness in primary school children in Hong Kong. J Epidemiol Community Health. 1998;52(9): Rich DQ, Kipen HM, Huang W, et al. Association between changes in air pollution levels during the Beijing Olympics and biomarkers of inflammation and thrombosis in healthy young adults. JAMA. 2012;307(19): Chen R, Zhao A, Chen H, et al. Cardiopulmonary benefits of reducing indoor particles of outdoor origin: a randomized, double-blind crossover trial of air purifiers. J Am Coll Cardiol. 2015;65(21):

7 1296 Li et al. 18. Stoner L, Lucero AA, Palmer BR, et al. Inflammatory biomarkers for predicting cardiovascular disease. Clin Biochem. 2013;46(15): Huang W, Wang G, Lu SE, et al. Inflammatory and oxidative stress responses of healthy young adults to changes in air quality during the Beijing Olympics. Am J Respir Crit Care Med. 2012;186(11): Viehmann A, Hertel S, Fuks K, et al. Long-term residential exposure to urban air pollution, and repeated measures of systemic blood markers of inflammation and coagulation. Occup Environ Med. 2015;72(9): Bind MA, Baccarelli A, Zanobetti A, et al. Air pollution and markers of coagulation, inflammation, and endothelial function: associations and epigene-environment interactions in an elderly cohort. Epidemiology. 2012;23(2): Devlin RB, Duncan KE, Jardim M, et al. Controlled exposure of healthy young volunteers to ozone causes cardiovascular effects. Circulation. 2012;126(1): Brook RD, Urch B, Dvonch JT, et al. Insights into the mechanisms and mediators of the effects of air pollution exposure on blood pressure and vascular function in healthy humans. Hypertension. 2009;54(3): Bell ML, Dominici F, Samet JM. A meta-analysis of timeseries studies of ozone and mortality with comparison to the National Morbidity, Mortality, and Air Pollution Study. Epidemiology. 2005;16(4): Atkinson RW, Kang S, Anderson HR, et al. Epidemiological time series studies of PM 2.5 and daily mortality and hospital admissions: a systematic review and meta-analysis. Thorax. 2014;69(7): Cole MP, Freeman BA. Promotion of cardiovascular disease by exposure to the air pollutant ozone. Am J Physiol Lung Cell Mol Physiol. 2009;297(2):L205 L Hutcheon JA, Chiolero A, Hanley JA. Random measurement error and regression dilution bias. BMJ. 2010; 340:c2289.