Objective · To study the metabolite profiles on patients of ischemic stroke using metabonomics approach. Methods · The serum samples from the 29 patients with ischemic stroke and 31 healthy controls were analyzed by gas chromatography time-of-flight mass spectrometry (GC-TOFMS) coupled multi-dimensional statistical methods to find differential metabolites in two groups. Results · Orthogonal partial least squares analysis (OPLS) model was generated based on identified metabolites and shown clear discrimination from patients and healthy controls. Some serum metabolite levels were significantly altered in patients. Six up-regulated metabolites included γ- aminobutyric acid, glutaric acid, glyceric acid, gluconic acid, lactobionic acid, and cholesterol, and nineteen down-regulated metabolites included citric acid, aconitic acid, malic acid, succinic acid, β-alanine, and glycerol-3 phosphate. Conclusion · Amino acid metabolism, glycometabolism and tricarboxylic acid (TCA) cycle are disturbed in patient of ischemic stroke. The metabonomic approach has great potential to understand the underlying mechanisms of stroke in ischemic patients.