RATIONALE: Chronic obstructive pulmonary disease (COPD) has been identified as a risk factor for cardiovascular diseases such as myocardial infarction. The role of COPD in cerebrovascular disease is, however, less certain. Although earlier studies have suggested that the risk for stroke is also increased in COPD, more recent investigations have generated mixed results. OBJECTIVES: The primary objective of our review was to quantify the magnitude of the association between COPD and stroke. We also sought to clarify the nature of the relationship between COPD and stroke by investigating whether the risk of stroke in COPD varies with age, sex, smoking history, and/or type of stroke and whether stroke risk is modified in particular COPD phenotypes. RESULTS: The MEDLINE and EMBASE databases were searched in May 2016 to identify articles that compared stroke outcomes in people with and without COPD. Studies were grouped by study design to distinguish those that reported prevalence of stroke (cross-sectional studies) from those that estimated incidence (cohort or case-control studies). In addition, studies were stratified according to study population characteristics, the nature of COPD case definitions, and adjustment for confounding (smoking). Heterogeneity was assessed using the I2 statistic. We identified 5,493 studies, of which 30 met our predefined inclusion criteria. Of the 25 studies that reported prevalence ratios, 11 also estimated prevalence odds ratios. The level of heterogeneity among the included cross-sectional studies did not permit the calculation of pooled ratios, save for a group of four studies that estimated prevalence odds ratios adjusted for smoking (prevalence odds ratio, 1.51; 95% confidence interval, 1.09-2.09; I2 = 45%). All 11 studies that estimated relative risk for nonfatal incident stroke reported increased risk in COPD. Adjustment for smoking invariably reduced the magnitude of the associations. CONCLUSIONS: Although both prevalence and incidence of stroke are increased in people with COPD, the weight of evidence does not support the hypothesis that COPD is an independent risk factor for stroke. The possibility remains that COPD is causal in certain subsets of patients with COPD and for certain stroke subtypes.