High serum levels of lipoprotein(a) and homocyst(e)ine are considered independent risk factors for atherothrombotic disease. In a prospective study in patients undergoing coronary artery bypass grafting, the preoperatively determined lipoprotein(a) and homocyst(e)ine levels were related to the frequency of 1-year graft occlusion. A cohort of 610 patients who underwent coronary artery bypass surgery was followed through the first postoperative year. Shunt angiography was performed in 581 patients (95%) at a mean of 12.1 +/- 1.5 months after the operation. The serum levels of lipoprotein(a) (n = 570) and homocyst(e)ine (n = 565) in patients with occluded internal mammary artery (IMA) grafts were not significantly different from the levels in those with open IMA grafts. Also, the serum lipoprotein(a) and homocyst(e)ine levels in patients with > or = 1 occluded vein graft were not significantly different from those in patients with all vein grafts patent. This study also determined the incidence of graft occlusion in quartiles of the lipoprotein(a) and homocyst(e)ine levels, respectively, and tested for linear trends. No significant trends in the incidence of graft occlusion were found, but the number of patients with vein graft occlusions was higher in the lowest quartile of lipoprotein(a) than that in the upper 3 quartiles (odds ratio, 1.82, 95% confidence interval, 1.21 to 2.74, p = 0.0025). Controlling for background variables in multivariate models only slightly modified the results. Thus, apart from an unexplained excess of vein graft occlusions in the lowest quartile of lipoprotein(a) levels, no association between the preoperative serum lipoprotein(a) or homocyst(e)ine levels and the frequency of 1-year graft occlusion could be demonstrated.