There is growing evidence of a link between ED, metabolic syndrome (MS) and cardiovascular disease (CVD). The study was to explore the prevalence of MS using three different definitions (World Health Organization (WHO), International Diabetes Foundation (IDF) and Adult Treatment Panel III (ATP III)), and to compare the association of CVD in ED outpatients using these definitions. This study enrolled 254 participants with a mean age of 55.3±0.9 years (range, 21 to 81 years) with ED as diagnosed by International Index of Erectile Function score. All participants underwent MS evaluation based on the three criteria. Differences of MS prevalence, demographical characteristics, biochemical profiles, pro-inflammatory and inflammatory markers, echocardiographic characteristics and the association with Framingham cardiac risk score (FCRS) were compared. The presence of diabetes mellitus (DM) in the WHO group and high waist girth in the IDF group were significant because of the necessity of respective criteria. The MS prevalence in the WHO, IDF and ATP III groups was 30.7, 34.3 and 36.6%, respectively (P ¼ 0.367). The degrees of agreement among each definition were substantial to perfect. No significant findings in echocardiographic characteristics, biochemical, inflammatory and pro-inflammatory markers were noted. The FCRS showed borderline nonsignificant difference (17.9 ± 0.4, 16.8 ± 0.4 and 16.9 ± 0.4, P ¼ 0.079); however, the FCRS was more closely correlated with the WHO than with the IDF and ATP III (Spearman's correlation coefficients were 0.522, 0.531 and 0.462, respectively; P ¼ 0.021). In patients o 55 years of age and those who smoke, the Spearman's correlation in the WHO group was significantly higher than in the IDF and ATP III groups. The prevalence of the MS using different definitions in ED patients was not different. The WHO-defined MS was more closely associated with CVD.