Abstract MP13: Plasma Homocysteine is Causally Associated With Ischemic Stroke in a Han Chinese Population: A Mendelian Randomization Study

Abstract

Introduction: Observational studies identified plasma homocysteine as a risk factor for ischemic stroke, but randomized controlled trials of folate supplementation reported inconsistent findings. The methylenetetrahydrofolate reductase ( MTHFR) gene encodes an enzyme critical for homocysteine metabolism and can serve as an unconfounded proxy for plasma homocysteine levels. Hypothesis: We assessed the hypothesis that there was a causal association between plasma homocysteine and risk of ischemic stroke by Mendelian randomization method. Methods: We examined the association between the MTHFR gene and ischemic stroke among 11,753 Chinese participants. The discovery stage included 999 ischemic stroke cases and 1,001 controls with no history of atherosclerotic disease. The MTHFR gene was sequenced using the SOLiD 4hq platform. Variants with minor allele count >10 were each tested for association with stroke using logistic regression models. Interactions between significant variants and alcohol drinking on stroke were also tested. Aggregate analysis of rare variants (minor allele count <20) used the sequence kernel association test. Variants achieving P<1х10 -4 were tested for replication among 4,724 ischemic stroke cases and 5,029 controls from the China Stroke Project. Results: An association between MTHFR rs1801133, a missense variant known to reduce enzyme activity and increase plasma homocysteine level, and ischemic stroke was identified in discovery, replication and joint analyses (P=2.10х10 -8 , 2.65х10 -11 , and 6.93х10 -17 , respectively). Each copy of the rs1801133 risk allele conferred respective odds ratios (95% confidence interval) of 1.44 (1.27, 1.64), 1.21 (1.15, 1.28), and 1.25 (1.18, 1.31). This variant also interacted with alcohol intake on stroke risk (P for interaction=7.82х10 -9 , 1.86х10 -10 , and 1.46х10 -18 , respectively). Odds ratios (95% confidence interval) of stroke associated with the risk allele were 1.66 (1.40, 1.96), 1.26 (1.18, 1.36), and 1.32 (1.23, 1.40) in non-drinkers and 1.17 (0.92, 1.49), 1.12 (1.01, 1.23), and 1.12 (1.03, 1.23) among alcohol drinkers in discovery, replication and joint analyses, respectively. No additional variants were identified. Conclusions: In conclusion, this is the first large scale genomic study to implicate the etiologic relevance of plasma homocysteine in ischemic stroke. An attenuating effect of alcohol intake on this relation was also observed.