During a mean follow-up period of 4.9 years, 308 deaths were identified, 200 of which were due to cardiovascular disease. Compared with patients without MetS, patients with MetS according to the AHA/NHLBI statement had a 1.26-fold higher risk (95% CI, 1.01–1.59) of all-cause mortality and a 1.41-fold higher risk (1.06–1.87) of cardiovascular mortality. Patients with increasing numbers of components of MetS had a gradually increased risk for all-cause and cardiovascular mortality (P < 0.05). When each component of MetS was considered as a dichotomized variable separately, only low high-density lipoprotein cholesterol (HDL-C) and elevated fasting blood glucose (FBG) were associated with all-cause and cardiovascular mortality. After using restricted cubic splines, we found a U-shaped association of HDL-C, body mass index and blood pressure, a positive association of FBG, and no association of triglycerides with the risks of all-cause and cardiovascular mortality.