Han MH, Yi HJ, Kim YS, and Kim YS. Effect of Seasonal and Monthly Variation in Weather and Air Pollution Factors on Stroke Incidence in Seoul, Korea. Stroke. 2015
Seasonal variation of stroke across large regions with varying climates has been reported, though results have been mixed in prior studies. In this manuscript, Han et al evaluated the trends in ischemic and hemorrhagic stroke incidence across month and season in the Seongdong district of Seoul, Korea. This was undertaken to evaluate if variation is related to meteorological and air pollution parameters.
This district is located in an urban section of Seoul, has 4 distinct seasons, and has a single tertiary-care hospital that evaluates nearly all local stroke patients. Its approximately quarter million population remained stable over the 10 year study period. The average temperature, diurnal temperature range, average humidity, as well as particulate matter (PM10) and nitrogen dioxide (NO2) pollutant levels were obtained from meteorological and air quality agencies.
3,001 patients were in this study (2,202 ischemic stroke and 799 intraparenchymal hemorrhage); the average age was ~65 years. Monthly incidence (of all stroke) was highest in September (17.7/100,000) and lowest in October (14/100,000). The highest incidence for ischemic stroke (IS) was September (especially for men (OR 1.391) and individuals age >60 years (OR 1.268)) and highest incidence for hemorrhage (ICH) was in January. The seasonal IS rate in the summer (OR 1.183) and autumn (OR 1.127) was higher than that in the winter. There was no seasonal variation in history of traditional vascular risk factors. An association of mean temperature with IS risk was demonstrated, with a 0.6% higher risk of IS for every 1℃ increase in monthly temperature (RR 1.006, p=0.003). In the older age group, average temperature was associated with IS incidence (RR 1.007, p=0.007) and negatively correlated with ICH (RR 0.985, p=0.004). No association was demonstrated with humidity. Interestingly, PM10 levels were negatively correlated with IS (RR 0.97, p=0.009) and the association was more significant in men (RR 0.943, p=0.002). However, PM10 (RR 1.106, p=0.002) and NO2 (RR 1.262, p=0.001) levels were positively correlated with ICH incidence in older individuals.
Notable limitations of the study include being single-center (serving a small geographic region) and that a limited number of air pollution and meteorological variables was assessed. Nevertheless, the results are interesting. Future work will have to evaluate the biological mechanisms leading to such variation in stroke incidence and further elucidate the impact of environmental pollutants.