Long-Term Exposure to Ambient Air Pollution and Myocardial Infarction-A systematic review and meta-analysis CURRENT

Background and Objective: An increasing amount of epidemiological original studies suggested that long-term exposure to particulate matter (PM 2.5 and PM 10 ) could be associated with the risk of myocardial infarction(MI), but the results were inconsistent. We aimed to synthesized available cohort studies to identify the association between ambient air pollution (PM 2.5 and PM 10 ) and MI risk by a meta-analysis. Methods: PubMed and Embase were searched through September 2019 to identify studies that met predetermined inclusion criterion. Reference lists from retrieved articles were also reviewed. A random-effects model was used to calculate the pooled relative risk ( RR ) and 95% confidence intervals ( CI ). Results: Twenty-two cohort studies involving 6,567,314 participants and 865,98 patients with MI were included in this systematic review. The pooled results showed that higher levels of ambient air pollution (PM 2.5 and PM 10 ) exposure were significantly associated with the risk of MI. The pooled relative ratio ( RR) for each 10-μg/m 3 increment in PM 2.5 and PM 10 were 1.20 (95% CI : 1.11–1.29), and1.03 (95% CI :1.00-1.07) respectively. Exclusion of any single study did not materially alter the combined risk estimate. Conclusions: Integrated evidence from cohort studies supports the hypothesis that long-term exposure to PM 2.5 and PM 10 as a risk factor for MI.


Introduction
The incidence and prevalence of cardiovascular diseases have increased in the last decades and became one of the main causes of death among adults [1][2][3] . Myocardial infarction (MI) is an acute and severe cardiovascular disease that generally can endanger the life of patients and has become a serious public health problem [4] . The causes of MI are complex and are related to lifestyle, diet structure, genetic factors and environmental factors, including air pollution [5][6][7] . Studies have shown that reducing modifiable risk factors may contribute to the prevention and control of MI [8][9][10] , which is of considerable public health importance.
Air pollution especially particulate matter (PM) has been increasingly investigated as an environmental risk factor for MI morbidity and mortality recently. However, these studies had modest sample sizes and reported inconclusive results [11][12][13] . These inconsistent and controversial results indicate the need to quantitatively synthesize and interpret the available evidence to provide more explicit information for policy decisions and clinical use. Meta-analysis is a statistical tool that can be used to integrate results of multiple independent studies considered to be 'combinable' for a more precise estimation [14,15] . Although previous meta-analyses have examined associations between air pollution exposure and MI, their studies most included cross-sectional literatures [13] , and many new researches have been recently published. Therefore, more meta-analyses with cohort studies are urgently needed.
Taken into consideration of the inconsistent conclusions and the limitation of existing epidemiological studies and the flaw of previous meta-analyses, we therefore performed a systematic review and meta-analysis of cohort epidemiological studies to examine the potential associations between air pollution exposure and the risk of MI. Given the heavy economic and health burden of curing MI, the results of our study may provide additional practical and valuable clues for the prevention of MI.

Materials And Methods
Ethical approval is not required for this systematic review.

Literature search strategy
We conducted this meta-analysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [16] and the checklist of items in the Meta-Analysis of Observational Studies in Epidemiology (MOOSE) [17] . A systematic literature search of PubMed and Embase was conducted through September 2019 by using the following search terms with no restrictions: 'air pollution' or 'particulate matter' or 'air pollutants' or 'PM 10 'or 'PM 2.5 ' in combination with 'myocardial infarction' or 'heart attack' or 'acute coronary syndrome'. The language was restricted to English. Additionally, reference lists of the retrieved original articles and relevant review articles were also scrutinized to identify further pertinent studies.

Study selection
Studies meeting the following criteria were included in the meta-analysis: (1) the study design was cohort; (2) the exposure of interest was ambient air pollution; the endpoint of interest was incidence of MI; and (3) the relative risk (RR) and the corresponding 95% confidence interval (CI) of MI relating to ambient air pollution were reported or could be calculated from the data provided. Animal studies, clinical trials, reviews, letters and commentaries were excluded. Only studies with detailed information on both ambient air pollution and the incidence of MI was included.
Data extraction and quality assessment Two investigators (WF and SC) extracted the following information from the studies: first author, publication year, country, study period, age, number of cases, size of cohort and time windows of exposure. Discrepancies were resolved by discussion with a third investigator (SC).
The Newcastle-Ottawa Scale was used to evaluate the qualities of cohort studies [18] , which is a ninepoint scale allocating points based on the selection of participants, comparability of groups, and exposure/outcome. This scale awards a maximum of nine points to each study: four for selection of participants and measurement of exposure, two for comparability of cohorts or cases and controls on the basis of the design or analysis, and three for assessment of outcomes and adequacy of follow-up.
Studies scoring 0-3 points, 4-6 points, and 7-9 points were categorized as low, moderate, and high quality of studies, respectively. When studies had several adjustment models, we extracted those that reflected the maximum extent of adjustment for potentially confounding variables. Each study was rated independently by two authors (WF and JM). Discrepancies were resolved by discussion with a third investigator (SC).

Statistical analyses
We used RR to measure the association between ambient air pollution and the risk of MI and the random effects model was used to calculate an overall pooled RR for the main analysis. Q statistic with a significance level at P < 0.10 and I 2 statistic were used to test the heterogeneity.
The I 2 statistic measures the percentage of total variation across studies due to heterogeneity rather than chance. It was calculated according to the formula by Higgins [19] . We used I 2 to quantify the heterogeneity, with 25%, 50%, and 75% indicating low, moderate and high degrees of heterogeneity, respectively.
In our meta-analysis, the following formula was used to calculate the standardized risk estimates for each study: Subgroup analyses were conducted to determine the possible influence of some factors such as state and publication years. We conducted a sensitivity analysis to explore potential sources of heterogeneity and to investigate the influence of various exclusion criteria on the pooled risk estimate. The Begg's rank correlation and the Egger's linear regression tests were used to assess the potential publication bias [20,21] . Using Duval and Tweedie's nonparametric trim-and-fill method to adjust potential publication bias [22] . All analyses were performed using STATA statistical software (version 12.0; College Station, TX, USA) and all tests were two-sided with a significance level of 0.05.

Literature search
Figure1 shows the process of study identification and inclusion. Initially 2977 and 3109 citations were retrieved from PubMed database and Embase, respectively. After excluding 236 duplicates, 5850 potentially relevant studies from electronic database were identified. Of these, we excluded 5698 papers because they were experimental, biomechanics, reviews or irrelevant studies. After full-text review of the remaining 152 articles, 28 articles were excluded because of insufficient data to calculate the risk estimates, and 102 were excluded because they were not a risk factor. Finally, 22 studies  were included.

Characteristics of the included studies
The main characteristics of the 22 studies on MI and long-term PM exposure in our meta-analysis were summarized in Table I. These studies were published between 2004 and 2019. Among them, eight studies were from Europe, two studies were from Asia and twelve studies were from America.
The size of the cohorts ranged from 1120 to 4,404,046 with a total of 6,567,314 subjects. The exposure measure was PM 2.5 in seventeen studies and PM 10 in eleven studies; six publications investigated the association of MI with exposure to both PM 2.5 and PM 10 .The end point was MI incidence (thirteen studies), MI mortality (ten studies), only one study MI hospital, and two studies included MI incidence and mortality. Fifteen studies were published after 2010, and seven study before 2010. The quality assessment scores ranged from 6 to 9, with an average score of 7 points, representing satisfactory quality of the studies.

Results of meta-analysis
We employed Meta-analysis to assess the association of PM 2.5 and PM 10  Association between PM 10 and the risk of MI Eleven studies investigated the association of PM 10 exposure with the risk of MI (Fig. 3). The pooled estimates of these studies indicated that a 10 mg/m 3 increase in PM 10 was associated with a higher risk of MI (RR = 1.03, 95%CI: 1.00 to 1.07), and the heterogeneity is (P = 0.061, I 2 = 43.3%). We conducted subgroup analyses by state, and publication year (before 2010 versus after 2010). In general, these subgroup analyses showed no statistically significant difference in results.

Sensitivity
Sensitivity analyses was conducted to find potential sources of heterogeneity in the association between air pollution (PM 2.5 and PM 10 ) and MI risk, to examine the influence of various exclusions on the combined RR, and assess the robustness of all results. The pooled RR did not materially change, for PM 2.5 and PM 10 , the both overall combined RR did not materially change, with a range from1.18

Publication bias
Visual inspection of the funnel plot showed significant asymmetry (PM 2.5 Fig. 4 and Fig. 5 PM 10 Fig. 6 and Fig. 7). For PM 2.5 and PM 10 , the Egger test indicated of publication bias, but the Begg test did not (PM 2.5 Egger, Z = 4.248 p = 0.000, Begg, t = 1.04 p = 0.315; PM10 Egger, Z = 3.11 p = 0.002, Begg, t = 1.83 p = 0.104). We used the trim-and-fill method to evaluate the impact of any potential publication bias, the results showed that one and five potentially missing studies would be needed to obtain funnel plot symmetry for the association of MI and air pollution (PM 2.5 , PM 10 ), respectively.
After using the trim-and-fill method, both the corrected RR for PM 2.5 was 1.18 (95% CI: 1.08 to 1.29; random-effects model, p = 0.012), for PM 10 was 1.04 (95% CI: 1.02 to 1.07; random-effects model, p = 0.002), which suggested the both pooled RRs were not substantially changed by the correction for potential publication bias.

Discussion
Myocardial infarction has a high medical burden for families and society, and remains a worldwide public health challenge [45] . Ascertaining the risk factors of MI could provide significant information for the prevention of MI. Our meta-analysis has quantitatively examined the association between long-term exposure to ambient air pollution (PM 2.5 and PM 10 ) and MI. The pooled analysis included 22 cohort studies with more than 6.5 million people showed an inverse association between exposures to air pollutants (PM 2.5 and PM 10 ) and the risk of MI, which was consistent with previous reviews and meta-analysis [13] . However, previous meta-analysis only included studies published before 2014, while more recent studies were not included which might report lower estimates or negative results.
Additionally, our meta-analysis included all cohorts studies, which data reliable, the effect of exposure can be fully and directly analyzed, and conclusions are relatively stable.
The mechanisms by air pollution exposure could contribute to the development of MI might include inflammation, induction of autophagy and down-regulation of membrane repair protein MG53.
Researchers found that inflammation plays an important role in the formation of coronary atherosclerosis and aggravation of plaque instability, and air pollutants can also promote MI via promoting inflammation [46] . The second potential mechanism is induction of autophagy. Several observational studies have shown that autophagy is a normal process for cells to achieve their own metabolism and organelle renewal. Autophagy can maintain the body's metabolism to reduce damage and protect the organism. However, excessive autophagy can lead to apoptosis of cardiomyocytes and aggravate the damage of ischemic-related sites [47] . Studies have found that autophagic levels for exposure to air pollutants are significantly higher than the control group, while, the corresponding protein expression levels, MI size decreases, and myocardial cell damage decrease in Farnesoid X recertor(FXR)knock out SD rats. Therefore, it is speculated that the exposure of air pollutants promotes the development of MI through FXR-induced autophagy [48] . The third possible mechanism is down-regulation of membrane repair protein MG53.Exposure to air pollutants can affect membrane repair through down-regulating the expression of MG53 protein, and aggravates the severity of ischemia and hypoxia in MI [49] .
We also found that long-term exposure to PM 2.5 has a more pronounced effect than PM 10 on MI risk in each 10 µg/m 3 increase, which is in line with previous related researches [13] . Compared with PM 10 , PM 2.5 can remain suspended for a longer time in the air and be inhaled into the respiratory tract and directly into the pulmonary alveoli. In addition, PM 2.5 has a larger superficial area and hence absorbs more chemical constituents than PM 10 . Therefore, PM 2.5 is probably more harmful on human health than PM 10 [50, 51] .
Considering the different diets, lifestyles of people, and the prevalence of MI in different regions, we also conducted subgroup analysis by region, and statistically significant differences across subgroups were found but not for PM 10 . In Asia, a 10 mg/m 3 increase in PM 2.5 exposure was positively associated with the risk of MI (RR = 1.38), which was inconsistent with previous studies. The possible reasons for the differences including inconsistency in study designs and potentially selective reporting of the results for pollutants. However, it is suggested that Asia region should pay attention to the relationship between air pollution and MI, and more works about air pollution and epidemiology remain to be done. We conducted a subgroup analysis by publication years but found the pooled result of studies before 2010 was not significantly different from that after 2010.
There are several strengths in this meta-analysis. Firstly, all the studies in our analysis were cohort studies, which is considered as stronger measure for demonstrating causation and identification of risk factors than other observational study designs [52] . Secondly, in this comprehensive literature review, we pooled data of 22 cohort studies from several geographical regions in 1 meta-analysis, thus increasing the statistical power and allowing an investigation of regional patterns. Thirdly, sensitivity analysis and consistent results from various subgroup analyses indicated that our findings were reliable and robust, although heterogeneity existed among the included studies. Furthermore, all these studies were published in the past decade, indicating that data on both the exposure and the outcome are recent and relevant.
Some limitations in the present meta-analysis should be of concern. First of all, the heterogeneity of the included studies was significant and existed through the whole analyses. But we explored the potential heterogeneity resources by subgroup analyses and sensitivity analysis. Secondly, the number of included studies is not enough, especially for the mortality and hospital of MI and subgroup-analysis. Third, other air pollutants may have interaction with PM 2.5 and PM 10

Ethical Approval and Consent to participate
Ethical approval is not required for this systematic review.

Consent for publication
Not applicable.

Availability of supporting data
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Additionally, this study is a meta-analysis, no new data is generated.

Acknowledgement
We thank all the authors of the studies included in our meta-analysis.

Competing interests
The authors declare that they have no competing interests.

Funding
No funding.

Authors' contributions
Wenning Fu, Jing Mao, Zuxun Lu and Shiyi Cao had full access to all the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis.
Wenning Fu and Li Zou independently extracted the related data information.    Association between exposure to PM2.5 and the risk of MI in a meta-analysis of cohort studies.

Figure 3
Association between exposure to PM10 and the risk of MI in a meta-analysis of cohort studies.

Figure 4
Funnel plot with 95% confidence limits of PM2.5 and the risk of MI.

Figure 5
Filled funnel plot of RR from studies that investigated the association between PM2.5 and the risk of MI.

Figure 6
Funnel plot with 95% confidence limits of PM10 and the risk of MI.

Figure 7
Filled funnel plot of RR from studies that investigated the association between PM10 and the risk of MI.