Magnesium intake and incidence of stroke: Meta-analysis of cohort studies

Background and aims: Prospective cohort studies are inconsistent regarding the association between magnesium intake and the risk of stroke. The objective was to perform a meta-analysis to summarise the relationship between magnesium intake and risk of stroke in observational studies. Methods and results: We searched the PubMed and EMBASE databases for studies conducted from 1966 through August 2011. Prospective studies that provided relative risk (RR) estimates with 95% confidence intervals (CIs) for the association between magnesium intake and the risk of total stroke incidence or mortality were included. Data were independently abstracted by two investigators using a standardised protocol. Study-specific risk estimates were combined by using a random effects model. A total of eight studies, with 8367 stroke cases among 304,551 participants, were included in the meta-analysis. The summary RR indicated a significant association between the highest magnesium intake and reduced risk of total stroke (summary RR: 0.89; 95% CI: 0.82, 0.97); our dose-response analysis showed a borderline inverse association between magnesium intake and total stroke risk (an increment of 100 mg day(-1); summary RR: 0.98; 95% CI: 0.95, 1.00). Subgroup analyses suggested a significant inverse association between highest magnesium intake and the risk of ischaemic stroke (summary RR: 0.88; 95% CI: 0.80, 0.98). Conclusion: The present meta-analysis of prospective cohorts suggests that higher magnesium intake is associated with reduced risk of total and ischaemic stroke. However, well-designed randomised controlled trials are needed to draw a definitive conclusion. (C) 2012 Elsevier B. V. All rights reserved.