Impact of Air Pollution and Statins Use on Stroke among Hypertension
Patients: A Population-Based Cohort Study in Taiwan
Tsung-Ta Wu, Zong-Cheng Pan, Wen-Chao Ho, Meng-Hung Lin, Kang-Chih Fan, Pau-Chung Chen,
Trong-Neng Wu, Fung-Chang Sung, Ruey-Shiung Lin
There is growing evidence of the association between air pollution and the incidence and mortality of cardiovascular diseases especially stroke, which disease have had high risk to suffer among people who had hypertension. Stains are widely used for hyperlipidemia and cardiovascular disease by their cholesterol-lowering effect. Due to pleiotropy of statins, there are many study reported that statins are related to improve outcome and survival after stroke and may reduce the risk of stroke by their anti-inflammatory and neuroprotective effect. The objective of this study is to assess the modification of air pollution on using statins related to attack of stroke among hypertension patients. The study design was a retrospective cohort and the medical records of subjects including stroke events and statins use were collected by Longitudinal Health Insurance Database 2000 (LHID2000). Air pollution data including SO2,
CO, O3, NO2, PM10 and PM2.5 were collected by high-density Taiwan Environmental
Protection Administration monitoring stations and used in estimating exposure by Geographic Information Systems (GIS). Cox proportion regression models were used to estimate the relationship between air pollution and stroke occurring with statins using among hypertension patients. The results show that air pollution increases the risk of first attack of stroke and statins may reduce the risk of stroke occurring among hypertension patients. There are potential antagonistic effects between air pollution and statins use on stroke occurring among hypertension patients who were statins user. It is worth to be further studied to clarify this relationship among general population.
Keywords: Stroke; Air Pollution; Statins; Hypertension Patients; Longitudinal Health Insurance Database 2000 (LHID2000)
Introduction
Cerebrovascular diseases are the third of leading causes of death in Taiwan. There is growing evidence of the association between air pollution and the incidence of cardiovascular diseases especially stroke, which disease has had high risk to suffer among people who having hypertension. Stains are widely used for hyperlipidemia and cardiovascular disease by their cholesterol-lowering effect. Due to pleiotropy of statins, there are many studies reported that statins are related to improving outcome and survival after stroke. Statins may reduce the risk of stroke by their anti-inflammatory and neuroprotective effect. There are no study to research the association between air pollution and stroke occurring with statins use among hypertension patients.
Objective
The objective of this study is to estimate the effects of air pollution on stroke among hypertension patients with using statin.
Methods
The study design was a retrospective cohort study. We used outpatient visits and admissions records which included information on patient characteristics, such as sex, date of birth, dates of visits, date of admission, date of discharge, and diagnoses for outpatient visits and admissions(using the International Classification of Diseases, ninth revision [ICD-9], classification system). Our cohort was defined patients who had first-time diagnosis of hypertension (ICD-9 codes 401, 402, 403, 404, and 405) and age >18 years between January 1, 2000 and December 31, 2009 (n=124137). We collected and analyzed their first-time occurring of stroke (ICD-9 codes 430, 431, 432, 433, 434, 435, 436, 437) after first-time diagnosis of hypertension and statin use records for calculating the defined daily dose (cDDD). Air pollution data including NO, NO2, NOx, CO, SO2, O3, PM2.5 and PM10 were
collected by high-density Taiwan Environmental Protection Administration monitoring stations and used in estimating every district’s air pollutants concentration in Taiwan by
Geographic Information Systems (GIS). Air pollution was assessed twelve months prior to first-time occurring of stroke. Cox proportion regression models were used to estimate the relationship between air pollution and stroke occurring among hypertension patients stratified by statin status. Hypothesis testing was two-sided at the 0.05 significance level and was performed using SAS software version 9.4.
Result
Table 1 showed the demographic variables of hypertension patients. A total of 86,893 hypertension patients were included as our study cohort. Among these patients, 25,906(29.8%) patients had used statin (≥28 cDDDs), and 10,025(11.5%) patients had first-time occurring of stroke in the following period. Among hypertension patients with statin use, female was 52.5% and age group 40‒49 has high percentage (31.16%). Table 2 exhibited twelve months air pollution patterns prior to first-time occurring of stroke. The mean concentration and interquartile range (IQR) of CO were 0.58 ppm and 0.23 ppm. Table 3 showed that the Spearman's rank correlation coefficient between air pollutants and O3 had negative
correlations with NO, NO2, NOx, and CO. The effects of air pollution and statin use on
occurring of stroke demonstrated in table 4. There were significant effects of NO, NO2, NOx,
CO, SO2, PM2.5 and PM10 on increasing the risks of first-time occurring of stroke. CO had
highest adjusted hazard ratio (Adjusted HR, 2.835; 95% CI, 2.757 to 2.916). Statin use had significantly protection effects on first-time occurring of stroke (Adjusted HR<1). Furthermore, NO, NOx, CO, SO2, PM2.5 and PM10 had significant interaction with statin use
(P-value<0.05). In subgroup analysis, all air pollutants had significantly adverse effects on stroke among hypertension patients for both statin user and statin nonuser. The effects of CO, SO2, PM2.5 and PM10 on first-time occurring of stroke among statin users were more harmful
than which among statin nonusers. The risks of NO and NOX on first-time occurring of stroke
among statin users were lesser than which among statin nonusers (Table 5). Discussion and Conclusion
Based on results, CO may have strongest effect on stroke occurring among hypertension patients, especially for statin user. Further investigations on biological mechanism and pathogenesis are in need. In addition, air pollution and statin use have interaction to first-time occurring of stroke. It should more research about different trends of interaction between different pollutants and statin use to stroke among hypertension patients. This study is first study to discuss the relationship between air pollution and occurring of stroke with statins use among hypertension patients. The strengths of this study are the large sample sizes, and complete medical and medication records. However, our research limitations are less information like other environmental risk factors and personal living habits.
In conclusion, air pollution may increase the risk of first-time occurring of stroke among hypertension patients even for statin user. It is necessary to protect public from the adverse effects of air pollution, especially such high risk group like hypertension. This issue should be more concerned and further researched.
Table1. Patient demographics of the hypertension cohort Patients With Statin Use (≥28 cDDDs; n= 25,906)
Patients Without Statin Use (<28 cDDDs; n= 60,987)
No % No %
First-time occurring of stroke 2,074 8.01 7,951 13.04
Sex Male 12,306 47.50 31,913 52.33 Female 13,600 52.50 29,074 47.67 Age, year 18-29 675 2.61 2,952 4.84 30-39 3,032 11.70 9,118 14.95 40-49 8,073 31.16 16,168 26.51 50-59 7,122 27.49 12,574 20.62 ≥60 7,004 27.04 20,175 33.08 Income, NT$ 0 6,007 23.19 13,721 22.50 1-15840 3,352 12.94 9,051 14.84 15841-25000 10,748 41.49 26,575 43.57 ≥25001 5,799 22.38 11,640 19.09 Urbanization level Ⅰ 9,029 34.85 17,616 28.88 Ⅱ 11,655 44.99 27,630 45.30 Ⅲ 3,573 13.79 10,673 17.50 Ⅳ(rural area)) 1,649 6.37 5,068 8.31
Table 2. Hypertension Patients average concentrations of the ambient air pollutants and meteorological factors.
Pollutant Mean Minimum Maximum IQR Percentile
25th 50th 75th NO, ppb 8.27 0.58 45.40 7.54 3.94 6.04 11.49 NO2, ppb 19.75 2.61 33.43 6.70 16.29 19.56 22.99 NOX, ppb 28.00 3.26 78.72 13.34 20.27 26.15 33.62 CO, ppm 0.58 0.17 5.14 0.23 0.45 0.54 0.67 SO2, ppb 4.29 0.77 12.15 1.23 3.39 3.64 4.62 O3, ppb 29.14 17.62 41.48 4.49 26.76 29.73 31.25 PM2.5, μg/m3 34.38 15.42 60.78 12.03 28.46 33.03 40.50 PM10, μg/m3 59.97 27.78 94.60 24.44 48.99 58.01 73.43 RH, % 72.98 62.31 86.55 2.40 71.51 72.81 73.91 Temp, °C 24.01 19.75 26.47 0.95 23.51 23.84 24.45
Abbreviations: RH, relative humidity; Temp, temperature; IQR, interquartile range.
Table 3. Correlation coefficients of ambient air pollutants average concentration during exposure period. NO NO2 NOX CO SO2 O3 PM2.5 PM10 RH Temp NO 1.00 0.89 0.96 0.91 0.09 -0.84 -0.47 -0.53 -0.24 -0.28 NO2 1.00 0.97 0.93 0.27 -0.80 -0.16 -0.25 -0.37 -0.02 NOX 1.00 0.95 0.17 -0.83 -0.33 -0.41 -0.31 -0.15 CO 1.00 0.12 -0.81 -0.28 -0.38 -0.28 -0.15 SO2 1.00 0.06 0.40 0.48 0.00 0.40 O3 1.00 0.46 0.55 0.25 0.38 PM2.5 1.00 0.94 -0.06 0.77 PM10 1.00 0.05 0.75 RH 1.00 -0.18 Temp 1
Abbreviations: RH, relative humidity; Temp, temperature All P-value < 0.001
Table 4. The adjusted hazard ratios and interaction tern between air pollution and statin use for stroke among hypertension patients.
Pollutants Model 1
a
Model 2b
Air pollution (aHR, 95% CI) Statin use (aHR, 95% CI) Interaction tern (P-value)
NO, ppb 1.040 (1.037-1.043) 0.535 (0.509-0.561) 0.0003 NO2, ppb 1.116 (1.110-1.121) 0.530 (0.505-0.556) 0.4149 NOX, ppb 1.033 (1.031-1.035) 0.531 (0.505-0.557) 0.0007 CO, ppm 2.835 (2.757-2.916) 0.550 (0.524-0.578) 0.0114 SO2, ppb 1.127 (1.114-1.140) 0.552 (0.526-0.580) <.0001 O3, ppb 0.811 (0.806-0.816) 0.531 (0.506-0.558) 0.1232 PM2.5, μg/m3 1.036 (1.034-1.039) 0.565 (0.538-0.594) 0.0109 PM10, μg/m3 1.004 (1.003-1.006) 0.553 (0.527-0.581) 0.0005
Abbreviations: aHR, adjusted hazard ratio.
a
Model 1: without interaction tern between air pollution and statin use.
b
Model 1: with interaction tern between air pollution and statin use and show P-value of interaction tern All model adjusted for sex, age, income, and urbanization level.
Table 5. Association between air pollution and statin use for stroke among hypertension patients.
Pollutants All Cases (aHR, 95% CI) Statin Users (aHR, 95% CI) Statin Nonusers (aHR, 95% CI)
NO, ppb 1.071 (1.067-1.074) 1.063 (1.055-1.071) 1.073 (1.069-1.078) NOX, ppb 1.071 (1.068-1.073) 1.066 (1.060-1.071) 1.072 (1.069-1.075) CO, ppm 3.257 (3.165-3.351) 3.711 (3.477-3.962) 3.104 (3.006-3.205) SO2, ppb 1.140 (1.127-1.154) 1.205 (1.175-1.235) 1.119 (1.104-1.135) PM2.5, μg/m3 1.038 (1.035-1.041) 1.043 (1.037-1.050) 1.035 (1.032-1.039) PM10, μg/m3 1.004 (1.002-1.005) 1.008 (1.004-1.011) 1.002 (1.000-1.004)
Abbreviations: aHR, adjusted hazard ratio.
