Background and Purpose-Experimental models of stroke provide consistent evidence of smaller stroke volumes in animals ingesting tea components or tea extracts. To assess whether a similar association of black or green tea consumption with reduced risk is evident in human populations, we sought to identify and summarize all human clinical and observational data on tea and stroke. Methods-We searched PubMed and Web of Science for all studies on stroke and tea consumption in humans with original data, including estimation or measurement of tea consumption and outcomes of fatal or nonfatal stroke. Data from 9 studies involving 4378 strokes among 194 965 individuals were pooled. The main outcome was the occurrence of fatal or nonfatal stroke. We tested for heterogeneity and calculated the summary effect estimate associated with consumption of Ն3 cups of tea (green or black) per day using random-effects and fixed-effects models for the homogeneous studies. Publication bias was also evaluated. Results-Regardless of their country of origin, individuals consuming Ն3 cups of tea per day had a 21% lower risk of stroke than those consuming Ͻ1 cup per day (absolute risk reduction, 0.79; CI, 0.73 to 0.85). The proportion of heterogeneity not explained by chance alone was 23.8%. Conclusions-Although a randomized clinical trial would be necessary to confirm the effect, this meta-analysis suggests that daily consumption of either green or black tea equaling 3 cups per day could prevent the onset of ischemic stroke.