LAUSR

BACKGROUND AND PURPOSE Fine particulate matter (PM2.5) has been associated with increased risks of stroke, but it remains unclear which specific size ranges and chemical constituents dominate the effects of PM2.5 on stroke. We aimed to evaluate the associations of size-segregated particles and various constituents of PM2.5 with daily emergency-room visits for stroke. METHODS We conducted a time-series study to investigate the associations of 5 particle size ranges from 0.01 to 2.5 µm and 35 constituents of PM2.5 with the daily emergency-room visits for stroke in Shanghai, from 2014 to 2019. Over-dispersed generalized additive models were used to estimate the associations. The robustness of these associations was evaluated by additionally controlling for PM2.5 mass. RESULTS For size ranges from 0.01 to 0.3 µm, there were significant positive associations between particle number concentrations and daily emergency-room visits for stroke with the strongest associations occurring for the size range 0.05-0.1 µm. The size-dependent pattern was not changed by adjusting for PM2.5 and gaseous pollutants. The associations of daily emergency-room visits for stroke also varied considerably by various PM2.5 constituents. After controlling for the simultaneous exposure to PM2.5 and gaseous pollutants in two-pollutant models, we identified 11 out of 35 constituents that had robust associations, these being organic carbon, elemental carbon, chlorine, magnesium, ammonium, nitrate, sulfate, copper, manganese, lead and zinc. CONCLUSION Ultra-fine particles and some PM2.5 constituents (i.e., carbonaceous fractions, inorganic ions and some elements) may be mainly responsible for the excess risk of stroke induced by PM2.5.