Type 2 diabetes mellitus is associated with a marked increase of coronary heart disease (CHD). We aimed to assess the impact of elevated serum lipoprotein (a) (Lp[a]) concentrations on the risk of CHD in patients with type 2 diabetes mellitus. A consecutive series of 352 outpatients was investigated. We determined the serum lipid profile and checked the patients for a history of CHD and of its traditional risk factors. Furthermore, the patients were divided into 3 groups according to the degree of elevation of the serum Lp(a) concentration: serum Lp(a) concentrations greater than 50 mg/dL, between 30 and 50 mg/dL, and less than 30 mg/dL, a presumed high normal value; and the prevalence of CHD was compared among the 3 groups. The serum Lp(a) concentrations in the subjects varied widely from 0.4 to 163.6 mg/dL. Patients with CHD had significantly higher serum Lp(a) concentrations than those without CHD (P = .0045). Logistic regression analysis to identify factors associated with the presence of CHD revealed that elevated serum Lp(a) is a significant risk factor (P = .0246). The prevalence of CHD increased with increasing serum Lp(a) concentrations (P = .048). Patients with serum Lp(a) concentrations greater than 50 mg/dL had a significantly higher prevalence of CHD than those with serum Lp(a) concentrations less than 30 mg/dL: the odds ratio of an elevated serum Lp(a) concentration was 3.346 (P = .039). In conclusion, elevated serum Lp(a) is a significant risk factor; and the risk of CHD appears to increase with increasing serum Lp(a) concentrations. Serum Lp(a) concentration of 50 mg/dL might represent a threshold level in relation to the risk of CHD in patients with type 2 diabetes mellitus.