P148 Background: Severe hyperhomocyst(e)inemia is associated with premature atherothrombosis and stroke in children with homocystinuria. Epidemiological studies indicate that mild-to-moderate hyperhomocyst(e)inemia is an independent risk factor for coronary artery disease. However, data are conflicting concerning the risk of stroke associated with mild-to-moderate hyperhomocyst(e)inemia. Aim: To describe and quantify the risk of stroke associated with elevated plasma homocyst(e)ine by meta-analysis. Methods: Relevant articles were obtained by search of Medline, Science Citation Index, and abstracts of conference proceedings. Search terms were Homocysteine, stroke, genetics and MTHFR. Exposure was defined as hyper-homocyst(e)inemia, relative to control values in each study population. Outcome was defined as ischemic stroke confirmed by neuroimaging (NI), and stroke without NI. Inclusion criteria: (1) Case-control (C-C) studies with reported odds ratios (OR) (2) Cohort (COH) studies with reported risk ratios (RR). Exclusion criteria: (1) Studies without clearly-defined OR/RR (reviews, C-C without OR, cross-sectional studies, case series, case reports) (2) Studies which defined outcome as carotid atherosclerosis or intima-media thickening. Statistical analyses for between-study heterogeneity and pooled risk estimates were performed using Stata software©. Results: Of 35 studies analyzed in detail, 15 (2 COH, 13 C-C) met inclusion criteria. Of 13 C-C studies (1306 cases, 6793 controls), the adjusted OR ranged 0.99–4.7, two had adjusted OR Conclusions: On average, mild-moderate hyperhomocyst(e)inemia is independently associated with a 69% increase in the odds and 86% increase in the risk of stroke.