Autoantibodies against oxidized low-density lipoprotein (LDL) have been reported to be associated with atherosclerosis. However, data are not consistent. We compared the titres of autoantibodies to malondialdehyde-modified LDL in three groups, a case group with angiographically documented severe coronary stenosis (> 80% stenosis in at least 1 vessel, n = 47), a hospital control group with minor stenosis on the coronary angiography (< 50% stenosis in all three major vessels, n = 47) and a healthy population control group with no history of coronary heart disease (n = 49). Age ranged from 26 to 68 years. Subjects were frequency-matched for gender distribution and storage time of the blood samples. No relevant differences in autoantibody titre between case and control groups were found. The mean autoantibody titres (± S.D.) were 1.44 ± 1.82, 1.46 ± 1.40 and 1.62 ± 1.95 for cases, hospital controls and population controls, respectively. No correlations were found between autoantibody titre and age, number of cigarettes smoked and LDL or total cholesterol. Autoantibody titres were correlated with body mass index (r = 0.2) and high-density lipoprotein (HDL) (r = - 0.2). Odds ratios (OR) were calculated by tertiles of autoantibody titres for the hospital control group and the population control group, respectively. Age-adjusted OR (95% confidence interval) for medium and high compared to low autoantibody titre were 0.76 (0.27-2.14) and 1.09 (0.39-2.95) for the comparison between cases and hospital controls and 1.09 (0.39-3.07) and 0.90 (0.32-2.56) for the comparison between cases and population controls. Adjustment for gender, body mass index, smoking habits and HDL yielded essentially the same results. This study does not support an association between autoantibody titres to oxidized LDL and the extent of coronary stenosis.