OBJECTIVE Coronary heart disease has been described to be increased with both glucose intolerance and cigarette smoking. All three of these have also been reported to be associated with central adiposity (disproportionate deposition of fat on the trunk compared to the extremities). The purpose of this analysis was to determine the relationship of cigarette smoking to glucose intolerance and coronary heart disease, the relationship of cigarette smoking to risk factors such as adiposity, body fat distribution, and plasma lipoprotein and insulin levels, the relationship of cigarette smoking to these risk factors independent of disease status, and whether these risk factors could account for any of the relationship between cigarette smoking and disease status. METHODS The study design was cross-sectional. The study sample contained 219 middle-aged and elderly Japanese-American men: 77 with normal and 74 with impaired glucose tolerance and 68 with type II diabetes. There were 54 men with coronary heart disease. A detailed smoking history was obtained. Glucose tolerance status was established by medical history and a 75-g oral glucose tolerance test. Coronary heart disease was determined by medical history and a resting electrocardiogram. Adiposity and fat distribution measurements were body mass index (kg/m2), skinfold thicknesses, body circumferences, and cross-sectional fat areas by computed tomography. Levels of insulin, C-peptide, cholesterol (total, low-density lipoprotein [LDL], high-density lipoprotein [HDL], HDL2, HDL3, very-low-density lipoprotein [VLDL]), and triglyceride (total, VLDL) were measured in fasting blood specimens. RESULTS A central pattern of body fat was associated with both non-insulin-dependent diabetes mellitus and coronary heart disease. Smoking history was related to both adiposity and body fat distribution, and was strongly related to coronary heart disease but not to diabetes. Past smokers who had smoked up to a month ago were the heaviest while present smokers who were currently smoking or had smoked within the past month were the leanest. However, although present smokers had reduced amounts of fat, this was attributable to those present smokers without heart disease. Present smokers with heart disease were not as lean and had increased amounts of intra-abdominal fat. Past smokers had the greatest amount of central fat and this was attributable to those with heart disease. By two-way (smoking history and coronary heart disease status) analysis of covariance, smoking history was significantly related only to subcutaneous fat disposition on the chest and abdomen independent of coronary heart disease, while coronary heart disease status was strongly related to plasma levels of insulin C-peptide, VLDL, HDL, HDL2, and HDL3 cholesterol, and total and VLDL triglyceride, independent of smoking history. Further analysis showed that none of the body fat variables could account for the risk of coronary heart disease associated with smoking history. Higher fasting plasma C-peptide levels in past smokers accounted statistically for part of the risk of coronary heart disease associated with cigarette smoking. However, this effect was not mediated by any of the body fat measurements. CONCLUSIONS Disproportionately increased intra-abdominal fat is related to coronary heart disease but not to smoking history. Smoking history is related to coronary heart disease but not to diabetes. Weight gain is associated with smoking cessation and appears to be concentrated in the central subcutaneous regions, especially for those who have coronary heart disease. Weight gain associated with cessation of smoking appears to be unrelated to atherogenic changes in lipids, lipoproteins, or insulin. Other pathogenic processes must be considered in the association between smoking and coronary heart disease.