Submicron particle exposure and stroke hospitalization: An individual-level case-crossover study in Guangzhou, China, 2014–2018

Abstract

BackgroundShort-term exposure to ambient PM2.5 and PM10 (particulate matter with aerodynamic diameters ≤2.5 μm and 10 μm, respectively) has been linked with hospitalization and mortality from stroke. However, the effect of PM1 (≤1 μm) exposure on the risk of hospitalization from stroke and its subtypes has rarely been investigated, in particular, on the basis of fine-scale exposure assessment at the individual level. MethodsWe collected data on hospital admissions due to stroke and its subtypes in Guangzhou, China from January 1, 2014 to December 31, 2018. Daily exposures to PM1, PM2.5, and PM10 were assessed from satellite-derived estimates at a 1-km2 spatial resolution based on residential addresses. A time-stratified case-crossover analysis combined with a conditional logistic regression model was performed to examine the associations of stroke hospitalization risks with short-term exposure to size-fractional particles. We conducted stratified analyses by sex, age, season, and ambient temperature. ResultsA total of 178,586 stroke hospitalizations were recorded during the study period, among which 141,709 cases were ischemic stroke and 25,255 cases were hemorrhagic stroke. The mean concentrations on the day of hospitalization were 20.0 μg/m3 (control days: 19.9 μg/m3) for PM1, 37.6 μg/m3 (37.4 μg/m3) for PM2.5, and 59.3 μg/m3 (59.0 μg/m3) for PM10. Short-term exposure to size-fractional particles was significantly associated with increased risks of hospital admission for overall stroke and ischemic stroke, whereas null or negative associations were observed for hemorrhagic stroke. Compared with PM2.5 and PM10, PM1 was associated with greater excess risks of stroke hospitalizations. For each 10-μg/m3 increase in PM1, PM2.5, and PM10 exposure at lag 03-day, the odds ratios were 1.016 (95% confidence interval: 1.008, 1.024), 1.007 (1.003, 1.011), and 1.007 (1.004, 1.010) for overall stroke hospitalization, and were 1.023 (1.014, 1.033), 1.010 (1.005, 1.014), and 1.009 (1.006, 1.013) for ischemic stroke, respectively. These associations were robust to co-pollutant adjustments and did not vary by sex and age, while significantly elevated risks were identified in cold months (October to March of the next year) and low-temperature days (<23.8 °C) only. ConclusionsShort-term exposure to particulate matter air pollution, particularly PM1, was associated with increased risks of hospitalization for overall stroke and ischemic stroke.