BACKGROUND AND PURPOSE: Elevated levels of lipoprotein(a) [Lp(a)] have been reported in association with symptomatic coronary and carotid artery disease. Relevancy of Lp(a) as a risk predictor of presymptomatic atherosclerosis in general populations is not well established. METHODS: Serum Lp(a) distribution and its relation to sonographically assessed carotid atherosclerosis were examined in a random sample of 885 men and women aged 40 to 79 years (Bruneck Study). RESULTS: Logistic regression analysis revealed a binary-type association between Lp(a) and carotid artery disease, with the threshold level of Lp(a) for an enhanced atherosclerosis risk defined at 32 mg/dL. The strength of relation increased with advancing severity of carotid atherosclerosis (odds ratios for Lp(a), 1.8 for nonstenotic and 4.7 for stenotic carotid artery disease; P < .001). Lp(a) was unaffected by environmental factors except for a significant decrease in women taking hormone replacement therapy (P < .05). In a multivariate approach, Lp(a) turned out to be an independently significant predictor of carotid atherosclerosis (P < .001). No differential effect of Lp(a) on atherosclerosis (effect modification) was observed for sex, age, low-density lipoprotein cholesterol, apolipoprotein A-I and B, fasting glucose, diabetes, or hypertension. However, the Lp(a)-atherosclerosis relation was significantly modified by fibrinogen (P < .01) and antithrombin III (P < .05). CONCLUSIONS: The present study demonstrates a strong and independent association between elevated Lp(a) levels and carotid atherosclerosis in a large randomized population and provides evidence of a potential role of Lp(a) in the evolution of carotid stenosis. Apart from atherogenicity of Lp(a) cholesterol, interference with fibrinolysis of atheroma-associated clots and fibrin deposits in the arterial wall may achieve pathophysiological significance.