Relationship Between Asian Dust and Ischemic Stroke
A Time-Stratified Case-Crossover Study
Background and Purpose–Particulate matter is increasingly recognized as a cause of human diseases, including cardiovascular diseases. However, the association between Asian dust (AD), a windblown sand dust originating from mineral soil in the deserts of China and Mongolia, and the incidence of cardiovascular diseases is unclear. The aim of the present study was to elucidate whether AD is associated with the incidence of ischemic stroke.
Methods–This study investigated the onset time of 7429 consecutive patients with ischemic stroke who were admitted to seven stroke centers in Fukuoka, Japan, between June 1999 and March 2010. Meteorologic variables and air pollutants, including AD and suspended particulate matters, were investigated during the identical period. A time-stratified case-crossover study was conducted to assess the association between AD and stroke incidence using a conditional logistic regression analysis.
Results–Although AD was not significantly associated with the overall incidence of ischemic stroke, there was a relationship specific to the stroke subtypes. AD was significantly associated with the incidence of atherothrombotic brain infarction after adjusting for expected confounders, including meteorologic variables and other air pollutants. No association was detected in other types of ischemic stroke.
Conclusion–These data suggest that AD does not increase the overall incidence of ischemic stroke, but it is specifically associated with the risk of atherothrombotic brain infarction.
Stroke is the third leading cause of death and is a major disease responsible for functional dependency in Japan. The underlying risk factors for stroke such as hypertension, diabetes, dyslipidemia, and atrial fibrillation all have been established based on a large number of epidemiological studies. However, the factors that trigger stroke in patients at high risk are still uncertain.1
Air pollution has been shown to increase mortality and hospital visits or admission attributable to cardiovascular diseases.2–4 Particulate matter and gaseous pollutants are thought to be a trigger of cardiovascular diseases, including stroke.3–7 Recent epidemiological studies have demonstrated the possibility that Asian dust (AD) may be related to diseases and health conditions.8 AD is a windblown sand dust originating from mineral soil in the deserts of China and Mongolia. The sand dust is suspended in the air and transported overseas to distant areas, such as Taiwan, Korea, and Japan. AD is larger than particulate matter and therefore is considered to have little effect on health conditions. However, AD picks up various fungi and air pollutants during long-range transport in the air. In the present study, we designed a time-stratified case-crossover study to elucidate the association between AD and the incidence of ischemic stroke.
Subjects and Methods
Patient Registry Data: Fukuoka Stroke Registry
Patient data were collected from the Fukuoka Stroke Registry. Fukuoka is situated in the Kyushu Island and faces China and the Korean peninsula across the Sea of Japan. This city is nearest to the Asian continent; therefore, AD frequently is observed there.
The study included the data of patients aged 20 years and older with ischemic stroke who were admitted to the participating 7 hospitals within 24 hours of onset between June 1999 and March 2010. The time of stroke onset was confirmed on admission. Stroke was defined as a sudden onset of nonconvulsive and focal neurological deficit persisting >24 hours. The diagnosis of stroke was confirmed by brain imaging, including computed tomography and/or magnetic resonance imaging and classified into 4 subtypes: lacunar, atherothrombotic, cardioembolic, and unclassified type of infarction.
Data on air pollutants were obtained from the atmospheric environmental database in the National Institute for Environmental Studies. Air pollutants included suspended particulate matter, photochemical oxidants, nitrogen dioxide, and sulfur dioxide. Data on meteorologic variables and ground-level observations of AD events were obtained from the Japan Meteorologic Agency. The daily mean of ambient temperature and relative humidity were obtained using hourly measurements at Fukuoka district meteorologic observatory located in Fukuoka city. The occurrence of an AD event was generally determined by visibility-based observation at the same observatory.
A time-stratified case-crossover design was applied to examine the association between exposure to AD events and the risk of hospitalization for stroke. Within-subject comparisons were made between a case period and control periods in this design. The study estimated the odds ratios and 95% confidence intervals of hospitalization associated with AD events during the case day (lag0) to a few previous days using conditional logistic regression, because the effect of AD events was assumed to persist over the course of a few days. The study controlled for 4-day averaged ambient temperature and relative humidity from lag0 to lag3 in the basic model. Besides, we added suspended particulate matter, oxidants, nitrogen dioxide, and sulfur dioxide were added separately in the basic model to adjust for the effect of co-pollutants. The analyses were repeated for stroke subtypes. All analyses were performed using the SAS statistical software package version 9.1.3 (SAS Institute, Cary, NC).
Patient Characteristics and Concentration of Air Pollutants
There were 7429 patients with ischemic stroke included in this study. The mean age of the study subjects was 72.4±11.9 years, and 3007 patients (40.5%) were female. The prevalence of hypertension, diabetes, dyslipidemia, and atrial fibrillation was 74.9%, 31.4%, 34.7%, and 28.8%, respectively. There were 137 days in the study period when an AD event was observed (AD day).
AD and Incidence of Ischemic Stroke
The case-crossover analysis showed that AD was not significantly associated with the overall incidence of ischemic stroke (Supplementary Table I). However, the relationship was specific to the stroke subtypes. The results after adjustment for temperature and relative humidity revealed a significant association between AD and incidence of atherothrombotic infarction (Figure). The association was strongest on days 0 to 2. There was no significant association between AD and the incidence of other stroke subtypes (Figure). Adding each pollutant in the basic model did not substantially alter the results (Table).
Few studies have investigated whether AD is associated with the occurrence of cardiovascular disease including ischemic stroke. A limited number of previous studies performed in Taipei, Taiwan, produced conflicting results.9–11 The present case-crossover study found that AD increased the risk of atherothrombotic brain infarction but not that of other types of ischemic stroke. The pathogenesis for ischemic stroke is diverse depending on subtypes, thus the association should be analyzed separately in each subtype.
The mechanisms by which AD increases the risk of atherothrombotic infarction are unknown. Meteorologic variables and the levels of air pollutants were significantly altered during AD events in comparison with those in the periods without AD (Supplementary Table II). The current study also investigated the effects of various air pollutants, including suspended particulate matter, nitrogen dioxide, and sulfur dioxide, on the risk of ischemic stroke because the exposure to air pollutants, such as particulate matter smaller than 10 μm in diameter, carbon monoxide, nitrogen dioxide, and sulfur dioxide have been suggested to increase the admission of ischemic stroke.5–7 Suspended particulate matter on AD days were marginally associated with the incidence of atherothrombotic infarction on that day (day 0; odds ratio, 1.14; 95% confidence interval, 1.00–1.29) or on the next 2 days (days 0–1; odds ratio, 1.18; 95% confidence interval, 0.98–1.41; data not shown). Particulate matter induces inflammation in the respiratory tract and facilitates a coagulative state.3,4,12,13 Therefore, the suspended particulate matter during AD events may have short-term adverse effects on the risk of atherothrombotic infarction. Although the magnitude of the association was modest (odds ratio <1.5), the overall impact of AD on the population would be significant because of its high exposure prevalence.
Another possibility is that unidentified substances are additionally involved in the association, because AD was still identified as a predictor for the occurrence of atherothrombotic brain infarction after adjustment for suspended particulate matter (Table). Recent studies have revealed that microorganisms during AD events are divergent and different from those during the non-AD period,14 and that microbial components adhering to AD induce allergic lung inflammation.15 Therefore, a systemic response to these adhered microorganisms originating from the distant desert may accelerate inflammation and thrombosis, leading to the incidental atherothrombotic infarction. There might be certain populations that are more susceptible to the adverse effects of AD, such as alcoholics (Supplementary Table III). AD-related substances may be more easily absorbed in such patients or cause inflammation in the respiratory tract. Further studies are needed to elucidate the mechanisms.
There are some limitations in the present study. There was no detailed information about patient behavior. Consequently, the degree of exposure to the AD and other air pollutants or residual confounders such as dehydration, infectious diseases, and exercise could not be assessed.
The authors are grateful to all the Fukuoka Stroke Registry collaborators for their participation and all the clinical research coordinators (Hisayama Research Institute For Lifestyle Diseases) for their help in collecting clinical data.
Sources of Funding
This study was supported by a grant-in-aid for Scientific Research on Innovative Areas from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (21120505).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.672501/-/DC1.
- Received July 28, 2012.
- Accepted August 14, 2012.
- © 2012 American Heart Association, Inc.
- Guiraud V,
- Amor MB,
- Mas JL,
- Touze E
- Pope CA III,
- Muhlestein JB,
- May HT,
- Renlund DG,
- Anderson JL,
- Horne BD
- Sun Q,
- Hong X,
- Wold LE
- Tsai SS,
- Goggins WB,
- Chiu HF,
- Yang CY
- Maheswaran R,
- Haining RP,
- Brindley P,
- Law J,
- Pearson T,
- Fryers PR,
- et al
- Wellenius GA,
- Schwartz J,
- Mittleman MA
- Bell ML,
- Levy JK,
- Lin Z