Insulin resistance (IR) is frequently observed in patients with coronary heart disease (CHD). The relationship between IR and the angiographical characteristics of coronary atherosclerosis were investigated in 66 patients with coronary artery lesions. Insulin resistance was assessed by a 75-g oral glucose tolerance test and homeostasis model assessment (HOMA). The angiographical characteristics of coronary atherosclerosis (i.e., the severity of CHD) were defined by both Gensini's score (GS) (a higher degree of coronary artery stenosis or a proximal lesion was assigned a higher score than a distal lesion) and the number of significantly stenosed vessels. When GS was examined as a categorical variable classified by tertile values (Group A, n = 22: 1< or =GS< or =14; Group B, n = 22: 15< or =GS< or =32; and Group C, n = 22: 33< or =GS), patients with a high GS (Group C) had significantly (p<0.05) higher values of fasting plasma insulin, insulin response, and HOMA IR than patients with a low GS (Group A) (12.6+/-1.2 microU/ml vs. 6.9+/-1.2 microU/ml, 122.2+/-11.9 microU ml(-1) h(-1) vs. 72.9+/-12.9 microU ml(-1) h(-1), and 2.9+/-0.3 vs. 1.5+/-0.3, respectively). The values in Group B patients (9.4+/-1.2,microU/mI, 108.5+/-12.5 microU ml(-1) h(-1), and 2.1+/-0.3, respectively) were intermediate between those in Groups A and C. The area of insulin/area of glucose ratio was significantly (p<0.05) higher in Groups B and C than in Group A (0.54+/-0.06 microU/mg, 0.54+/-0.06 microU/mg, and 0.32+/-0.06 microU/mg, respectively). However, no significant differences were observed in variables of glucose tolerance, serum lipid, lipoproteins, fibrinogen, uric acid, and blood pressure among the 3 groups. Significant (p<0.05) positive associations were found between GS, the number of diseased coronary arteries, and fasting immunoreactive insulin, insulin response, the area of insulin/area of glucose ratio and HOMA IR by logistic regression analysis. After adjusting for the number of diseased coronary arteries, the association between GS and IR was not significant, suggesting that IR contributed to the severity of coronary atherosclerosis but not to the distribution of lesions. In conclusion, IR was associated with the severity of CHD as measured by both Gensini's score and the number of diseased coronary arteries, and increased the risk of CHD regardless of the location of the lesions.