Short and long term exposure to air pollution increases the risk of ischemic heart disease

Previous studies have suggested an increased risk of ischemic heart disease related to air pollution. This study aimed to explore both the short-term and long-term effects of air pollutants on the risk of ischemic heart disease after adjusting for meteorological factors. The Korean National Health Insurance Service-Health Screening Cohort from 2002 to 2013 was used. Overall, 2155 participants with ischemic heart disease and 8620 control participants were analyzed. The meteorological data and air pollution data, including SO2 (ppm), NO2 (ppm), O3 (ppm), CO (ppm), and particulate matter (PM)10 (μg/m3), were analyzed using conditional logistic regression. Subgroup analyses were performed according to age, sex, income, and region of residence. One-month exposure to SO2 was related to 1.36-fold higher odds for ischemic heart disease (95% confidence interval [95% CI] 1.06–1.75). One-year exposure to SO2, O3, and PM10 was associated with 1.58- (95% CI 1.01–2.47), 1.53- (95% CI 1.27–1.84), and 1.14 (95% CI 1.02–1.26)-fold higher odds for ischemic heart disease. In subgroup analyses, the ≥ 60-year-old group, men, individuals with low income, and urban groups demonstrated higher odds associated with 1-month exposure to SO2. Short-term exposure to SO2 and long-term exposure to SO2, O3, and PM10 were related to ischemic heart disease.

The heterogeneous study design, regional and ethnic differences and types of measured pollutants might have contributed to these controversial results.
Few studies have reported a wide range of air pollution exposure periods from short-term to long-term exposure. Moreover, when exploring the effect of air pollution, meteorological factors should be concurrently considered because the concentration and composition of air pollutants might be influenced by these factors, and the risk of cardiovascular disease could be associated with meteorological factors, such as ambient temperature 19 . For instance, the solubility of air pollutants is increased at lower temperatures, and the photolysis reaction could change the composition of air pollutants. Therefore, this study analyzed the effect of air pollutants on ischemic heart disease according to exposure periods prior to the development of ischemic heart disease. To evaluate this effect, we calculated the mean levels of air pollutants for time periods from 3 to 730 days of exposure. To minimize the confounding effects of meteorological factors, they were concurrently analyzed for their association with ischemic heart disease.

Materials and methods
Participant selection. This study was approved by the Ethics Committee of Hallym University (2017-I102) and was exempt from requiring informed consent 20 . All procedures were followed in accordance with the relevant guidelines. The Korean National Health Insurance Service-Health Screening Cohort (NHIS-HEALS), meteorological, and air pollution data were used (S1 description) [20][21][22] .
Participants with ischemic heart disease were selected from 514,866 patients with 497,931,549 medical claim codes (n = 2239) [20][21][22] . Among these participants, individuals were excluded if they were diagnosed with ischemic heart disease before 2004 (n = 80) to track previous exposure to meteorological factors in the last 2 years. The control group included those without a history of ischemic heart disease from 2002 through 2013 (n = 512,627) [20][21][22] . The control participants were 1:4 matched for age, sex, income, region of residence, and index date 20 .
The index date of ischemic heart disease participants was set as the time of diagnosis of ischemic heart disease [20][21][22] . Some ischemic heart disease participants were excluded because there were not enough matched control participants (n = 4). Collectively, 2155 ischemic heart disease participants were 1:4 matched with 8620 control participants (Fig. 1) [20][21][22] .
Variables. Independent variables. Daily mean temperature (°C), daily highest temperature (°C), daily lowest temperature (°C), relative humidity (%), ambient atmospheric pressure (hPa), SO 2 (ppm), NO 2 (ppm), O 3 Figure 1. A schematic illustration of the participant selection process used in the present study. Of 1,125,691 participants, 2155 ischemic heart disease participants were matched with 8620 control participants for age, group, sex, income group, and region of residence. Then, ischemic heart disease patients and control participants were matched according to the meteorological data and air pollution data before the index date.
Statistical analyses. The general characteristics of the ischemic heart disease and control groups were compared using the chi-square test 20 . The mean meteorological and air pollution data 30 days and 365 days before the index date were compared using independent t-tests.
To analyze the odds ratios (ORs) with the 95% confidence intervals (CIs) of meteorological and air pollution data for ischemic heart disease participants compared to control participants, a crude model (simple model), adjusted model (all insertion model), and final model (backward selection of variables) were calculated using conditional logistic regression 20 . As ORs of independent variables were calculated as continuous variables, they were displayed as SO 2 per 0.01 ppm, NO 2 per 0.1 ppm, O 3 per 0.01 ppm, CO per 1 ppm, and PM 10 per 10 μg/ m 3 . In these analyses, age, sex, income, and region of residence were stratified. In the analyses of 3 days, 5 days, 10 days, 15 days, 30 days, 60 days, 90 days, 180 days, 270 days, 365 days, 540 days, and 730 days of exposure, we selected 30 days as the short-term exposure and 365 days as the long-term exposure 20 . The results of other days of exposure are displayed in the supplemental file (Supplementary Tables S2-S12). To select final models, Akaike information criterion and Baysian information criterion of air pollutants were analyzed (Supplementary Table S13). The correlations between meteorological and air pollutants were provided as supplemental tables (Supplementary Tables S14-S15).
For the subgroup analysis, we divided participants by age, sex, income, and region (< 60 years old and ≥ 60 years old; men and women; low income [income 1-3] and high income [income [4][5]; urban and rural, respectively) in the final model 20 .
Two-tailed analyses were performed, and significance was defined as P values less than 0.05. SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) was used for statistical analyses 20 .

Results
Regarding meteorological and air pollution data, SO 2 for 30 days, temperature (mean, highest, and lowest), relative humidity, SO 2 , NO 2 and CO for 365 days were different between the ischemic heart disease group and the control group (all P < 0.05, Table 1).
For 30 days of exposure, the OR was 1.36 (95% CI 1.06-1.75, Fig. 2) for SO 2 (0.01 ppm) in the ischemic heart disease group compared with the control group using the final model (  Fig. 4) for PM 10 (10 μg/m 3 ) in the ischemic heart disease group compared with the control group using the final model (Table 3).
In the subgroup analyses using the final model, in the age groups < 60 years old and ≥ 60 years old, the odds for O 3 and PM 10 were higher in ischemic heart disease participants than in the control group after 1 year of exposure. In the age ≥ 60 years group, the odds for SO 2 were higher in ischemic heart disease participants than in the control group after 30 days and one year of exposure. In men, individuals with low income, and urban groups, the odds for SO 2 for 30 days of exposure and odds for O 3 and PM 10 for one year of exposure were higher in ischemic heart disease patients than in the control group. In the high-income and rural groups, the odds for SO 2 for one year of exposure were higher in ischemic heart disease patients than in the control group (Table 4).

Discussion
Both short-term and long-term exposure to air pollutants were related to ischemic heart disease in the present study. The types of air pollutants that impacted ischemic heart disease differed according to exposure period. Short-term (30 days) exposure to SO 2 was related to higher odds of ischemic heart disease. For long-term (365 days) exposure, higher levels of SO 2 , O 3 , and PM 10 were associated with ischemic heart disease. In addition, www.nature.com/scientificreports/ the effects of air pollutants on ischemic heart disease were different according to the demographic factors of age and sex and the socioeconomic factors of income level and region of residence. Short-term exposure to SO 2 was associated with higher odds of ischemic heart disease in this study. In line with the present results, previous studies have reported elevated mortality related to short-term exposure to SO 2 18,29 . Although some prior studies demonstrated an association between exposure to PM 10 , SO 2 , and NO 2 with the disease burden of ischemic heart disease (years of life lost), one time-series study reported that SO 2 , but not other air pollutants or PMs, was related to the increased mortality of ischemic heart disease (excess risk of death = 3.18%, 95% CI 1.19-5.17) 18 . They found that gaseous pollutants, such as SO 2 , had higher impacts on the risk of ischemic heart disease than PMs 18 . Oxidative stress and the inflammatory response have been suggested as possible pathophysiologic mechanisms for the impact of SO 2 on ischemic heart disease 30 . Sulfate exposure for 2-7 days was associated with oxidative stress markers of urinary creatinine-indexed 8-epi-prostaglandin F2α in the Framingham heart study 30 .
Short-term exposure to other air pollutants, including NO 2 and PM 10, did not show an association with ischemic heart disease in the present study. Previous epidemiologic studies have suggested that the source or components of PMs are crucial for their hazardous impact on ischemic heart disease 11 . The PM 2.5 from windblown soil or biomass combustion was not associated with ischemic heart mortality 11 . The present study could www.nature.com/scientificreports/ not differentiate the sources of PM 10 because its heterogeneous composition could attenuate its adverse impacts on ischemic heart disease. Long-term exposure to SO 2 , O 3 , and PM 10 was related to higher odds of ischemic heart disease in this study. A large amount of previous epidemiologic data supports the long-term effects of air pollutants on the risk of ischemic heart disease 31,32 . The mortality of ischemic heart disease was 1.03-fold higher in patients who were exposed to a high level of PM 2.5 in the form of diesel traffic-related elemental carbon (95% CI 1.00-1.06) from Table 2. Crude and adjusted odd ratios (95% confidence interval) of the meteorological and pollution matter (mean of 30 days before the index date) for ischemic heart disease. *Conditional logistic regression model, Significance at P < 0.05. a Stratified model for age, sex, income, and region of residence. b Adjusted model was adjusted for obesity, smoking status (current smoker compared to nonsmoker or past smoker), frequency of alcohol consumption (≥ 1 time a week compared to < 1 time a week), CCI score, mean temperature, highest temperature, lowest temperature, relative humidity, atmospheric pressure, SO 2 , NO 2 , O 3 , CO, and PM 10 . c Final model was adjusted for obesity, smoking status (current smoker compared to nonsmoker or past smoker), frequency of alcohol consumption (≥ 1 time a week compared to < 1 time a week), CCI score, relative humidity, SO 2 , NO 2 , O 3 , CO, and PM 10 using the backward selection method.

Characteristics
Odds ratio for ischemic heart disease (95% CI)  11 . Multiple pathophysiologic mechanisms, including the systemic inflammatory response, prothrombotic pathway activation, oxidative stress, vascular dysfunction and remodeling, autonomic dysfunction, and epigenetic factors, have been proposed to mediate the impact of air pollutants on ischemic heart disease 12 . For instance, coronary artery calcification was proposed as one of the pathophysiologic mechanisms for the effect of air pollutants, including PMs and O 3, on the risk of ischemic heart disease 24,33 . The coronary artery calcium score, which is considered an atherosclerotic marker, was associated with elevated levels of PM 2.5   31 . Although the pathophysiologic mechanism of the effect of O 3 on ischemic heart disease remains elusive, oxidizing activities could induce inflammation in the coronary artery, which might result in atherosclerotic plaque formation and narrowing of the arterial lumen with increasing wall thickness 32 .
To support this hypothesis, it was reported that long-term exposure to O 3 was related to increased thickness of the common carotid artery (5.6 µm, 95% CI 1.4-9.7) and carotid plaque burden (OR 1.2, 95% CI 1.1-1.4) 32 .
The relative humidity for one year was negatively associated with ischemic heart disease in this study. Previous studies have suggested the contributions of meteorological factors of temperature variability with the risk of ischemic heart disease, although no prior research investigated the association of humidity with ischemic heart disease 34,35 . The potential impact of humidity on the solubility of gaseous pollutants and moisture content, which may decrease the amount of air pollutant exposure, could mediate the decreased rate of ischemic heart disease in high humidity conditions. The impacts of air pollutants were prominent in the old age group and men in this study. Previous studies have also suggested the higher susceptibility of older populations and men to the impacts of air pollutants on ischemic heart disease 18,31 . Preformed or subclinical atherosclerotic changes of the coronary artery could more easily progress due to the impacts of air pollutants on inflammation and atherosclerosis, although the synergistic or additive effects of air pollutants could not be determined in the current study. The high prevalence of ischemic heart disease in men compared to that in women could strengthen the statistical power in this population. In addition, populations with poor socioeconomic status showed a relationship between short-term exposure to SO 2 and ischemic heart disease in this study. Several previous studies investigated socioeconomic disparities and the impact of a higher burden of air pollution on morbidities and mortalities in minorities, although the results had some heterogeneity according to the air pollution models 36,37 . Populations with poor socioeconomic status have been reported to be exposed to more air pollution because of the lack of availability of air conditioning and increased industrial exposure [38][39][40][41] . In addition to high exposure to air pollution, poor socioeconomic groups were reported to have increased susceptibility to air pollution due to underlying health statuses and reduced access to medical care 38 . The association of short-term exposure to SO 2 and ischemic heart disease in the urban subgroup might be attributed to the higher level of air pollution in urban areas than in rural areas.
The analysis of a large, representative national cohort population strengthened the statistical power of the present study. The large study population enabled the selection of a control population that was matched for age, sex, income, and region of residence. Possible confounders were comprehensively considered in this study. In addition to past medical histories, lifestyle factors of smoking, alcohol consumption, and obesity were adjusted. Moreover, meteorological factors were concurrently considered along with air pollutants. The meteorological factors and air pollution data were collected and validated by the Korea Meteorological Administration. Based on these verified data, this study investigated both the short-term and long-term effects of air pollution on ischemic heart disease. However, a few limitations existed in the current study. Possible collinearity between air pollutants www.nature.com/scientificreports/ might exist, although we adjusted for these variables in the final models (Supplementary Tables S14-S15). The level of exposure to PM 2.5 was not available in this cohort. Because the exposure to air pollutants was based on the registered region of residence, the migration of participants during follow-up periods could not be accounted for in the present study. In addition, indoor exposure to air pollutants could not be individually assessed. For the diagnosis of ischemic heart disease, we could not differentiate the types or severity of disease because this study was based on health claims data. Last, because this study was based on Koreans, ethnic differences should be considered when interpreting this study.

Conclusions
Short-term exposure to SO 2 and long-term exposure to SO 2 , O 3 , and PM 10 were associated with an increased risk of ischemic heart disease. The older, male, low-income, and urban groups demonstrated an apparent association between short-term exposure to SO 2 and ischemic heart disease.

Data availability
Releasing of the data by the researcher is not allowed legally. All data are available from the database of the National health Insurance Sharing Service (NHISS; https ://nhiss .nhis.or.kr/). NHISS allows data access, at a particular cost, for any researcher who promises to follow the research ethics. Data of this article can be downloaded from the website after promising to follow the research ethics. Table 4. Adjusted odd ratios (95% confidence interval) of the meteorological and pollution matter for ischemic heart disease according to age and sex in the final model. *Conditional logistic regression model; Significance at P < 0.05. a Stratified model for age, sex, income, and region of residence. b Final model was adjusted for obesity, smoking status (current smoker compared to nonsmoker or past smoker), frequency of alcohol consumption (≥ 1 time a week compared to < 1 time a week), CCI score, relative humidity, SO 2 , NO 2 , O 3 , CO, and PM 10 using the backward selection method.