Association between oral contraceptives and risk of hemorrhagic stroke: a meta-analysis of observational studies.

OBJECTIVES: To estimate the risk of hemorrhagic stroke associated with current use of oral contraceptives (OCs), and to further depict how the risk was affected by study characteristics.

METHODS: We searched PubMed, Embase, and Cochrane Library for relevant articles published up to February 2017 that examined the association between OC use and risk of hemorrhagic stroke. Two investigators independently reviewed articles based on inclusion criteria. The Newcastle-Ottawa scale was employed to evaluate quality of studies. Random-effects meta-analysis model was used to generate summary risk estimates.

RESULTS: Fifteen independent studies (5 cohort studies and 10 case-control studies) with 4271 hemorrhagic stroke cases were included in this meta-analysis. The overall summary odds ratio (OR) for hemorrhagic stroke (HS) in relation to current OC use was 1.39 [95% confidence interval (CI), 1.05-1.83]. Subgroup analyses on hemorrhagic stroke types showed that OC use was associated with subarachnoid hemorrhage (SAH) (OR 1.68; 95% CI 1.21-2.12), but not with intracerebral hemorrhage (ICH) (OR 0.92; 95% CI 0.33-2.54). The risk of hemorrhagic stroke was increased slightly among current OC users with high-dose estrogen (OR 1.60; 95% CI 1.12-2.27). The risk of hemorrhagic stroke was further increased in OC users with additional risk factors including current smoking, hypertension, and history of migraine.

CONCLUSIONS: This meta-analysis of observational studies suggests that current use of OCs could contribute to a small increased risk of hemorrhagic stroke, and the increased risk is related to subarachnoid hemorrhage, but not intracerebral hemorrhage.