BACKGROUND: Studies on soy supplementation suggest a cardioprotective potential. OBJECTIVE: To examine the effects on LDL cholesterol and arterial function as a result of dietary enrichment with soy supplementation. DESIGN: A Randomized, double blind, parallel intervention trial. SETTING: Department of Endocrinology and Metabolism C, Aarhus University Hospital, and Department of Human Nutrition, The Royal Veterinary and Agricultural University, Denmark. SUBJECTS: In all, 100 hypercholesterolaemic but otherwise healthy subjects were included in the study of which 89 completed it. INTERVENTIONS: Subjects were randomly assigned to 24 weeks of daily intake of either a soy supplement, Abalon (30 g soy protein, 9 g cotyledon fibre and 100 mg isoflavones) or placebo (30 g of casein). The soy supplement and placebo were provided in two sachets daily that were stirred in water. Fasting plasma lipids, TNF-alpha, homocysteine, insulin sensitivity, homeostasis model assessment (HOMA-IR), serum insulin, serum glucose, blood pressure as well as Glucagon-like peptide 1 (GLP-1), glucose-dependent insulinotropic peptide (GIP) and plasma lipids to a fat-rich meal were recorded before and after the intervention. In a sub study in 32 subjects, arterial dilatory capacity, compliance, and distensibility were recorded before and after the intervention. RESULTS: In the main study, no difference in fasting plasma lipid levels or insulin sensitivity was found between soy-based supplement and placebo. A significant postprandial increase in GIP to the meal test was observed in the soy group (P < 0.05). In a substudy, no difference between the groups in changes in flow-mediated vasodilatation (P = 0.84) was detected, while the soy supplementation caused a reduction in LDL and total cholesterol. CONCLUSIONS: No significant effects on blood lipids were observed in the main study to a soy supplementation in hypercholesterolaemic subjects after 24 weeks. In the substudy, the soy supplementation, however, reduced LDL and total cholesterol but did not influence markers of arterial function.