To investigate which of the three recently proposed definitions of the metabolic syndrome (MetS) is related to the excessive risk of ischaemic non-embolic stroke in elderly individuals, and thus may be more appropriate to implement in clinical practice. In a population-based case-control study of subjects aged older than 70 years (163 patients vs. 166 controls), we evaluated the association of first-ever acute ischaemic non-embolic stroke with the MetS defined by using recent definitions as proposed by the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III), the International Diabetes Federation (IDF) and the National Heart, Lung and Blood Institute/American Heart Association (NHLBI/AHA). By applying the NCEP ATP III, NHLBI/AHA and IDF definitions, the prevalence of MetS in the patient group was 46%, 57.1% and 69.9%, respectively, compared with 15.7%, 18.1% and 30.7% in the control group (p < 0.001 for all comparisons). After adjusting for multiple risk factors, the odds ratio (OR) for ischaemic stroke was 2.59 [95% confidence interval (CI): 1.24-5.42, p = 0.012] for NCEP ATP III-defined MetS and 3.18 (95% CI: 1.58-6.39, p = 0.001) for NHLBI/AHA-defined MetS. However, the association of IDF-defined MetS with ischaemic stroke was not significant (OR 1.18, 95% CI: 0.50-2.78, p = 0.71). The implementation of the IDF (unlike NCEP ATP III and NHLBI/AHA) MetS definition substantially increases the number of elderly subjects labelled as having MetS without contributing to the identification of those at high risk of stroke.