Background: Stroke is one of the world's leading causes of death. Many studies have checked the relationship between short-term exposure to particulate matter (PM) and stroke, but few have focused on the effect of long-term exposure to PM2.5 (particulate matters with an aerodynamic diameter of <= 2.5 mu m). This study aimed to quantitatively examine the relationship of long-term exposure to PM2.5 with stroke incidence and mortality. Methods: We identified relevant studies by searching the PubMed, EMBASE and MEDLINE. After the systematical review of pertinent studies, random-effect meta-analysis was conducted to investigate the association between long-term exposure to PM2.5 and stroke. Results: Our meta-analysis included 16 cohort studies with more than 2.2 million people and above 49 149 endpoint events (incident stroke and death from stroke). The pooled hazard ratio (HR) for each 5 mu g/m(3) increment in PM2.5 was 1.11 (95% CI: 1.05, 1.17) (CI for confidence interval) for incidence of stroke and 1.11 (95% CI:1.05, 1.17) for mortality of stroke. In the region-specific analysis, significant association between PM2.5 and incidence of stroke was found in North America (HR =1.09, 95% CI:1.05, 1.14) and Europe (HR =1.07, 95% CI:1.05, 1.10), while the pooled result of Asia showed no significance (HR= 2.31, 95% CI:0.49, 10.95). Conclusions: Long-term exposure to PM2.5 is an important risk factor for stroke. Since air quality is intimately related to everyone, policies aimed at reducing particulate matters will benefit public health a lot.