Elevated levels of serum lipoprotein(a) [Lp(a)] are reported to be associated with risk of atherosclerosis and thrombosis. Little is known about the influence of Lp(a) on the progression of coronary artery disease. We evaluated the association of serum Lp(a) and the long-term changes of angiographic severity in patients who underwent repeated coronary angiography at intervals of more than 2 years. We evaluated 70 patients, and divided them into 3 groups by angiographic findings. Median Lp(a) concentration was significantly higher in the progression group (N=36) than in the no-change group (N=23) or the regression group (N=11) (32.4 vs 22, 19.3 mg/dl, p<0.05). Furthermore, the progression group had more patients whose Lp(a) levels were greater than 30 mg/dl (p=0.006), while in the regression group all patients were under 30 mg/dl. Stepwise logistic regression analysis for progression of lesions showed that Lp(a) > or =30 mg/dl remained significant, giving an estimated odds ratio (OR) of 2.46 (p= 0.005). In the subgroup analysis, OR in patients with mild lesions was reduced to 2.05 (p<0.05) while in patients with severe lesions OR was increased to 3.39 (p=0.003). The serum Lp(a) level has a close correlation with angiographic progression, and may be an important predictor for progression.