The associations between macronutrient intake and plasma parameters associated with increased risk for coronary heart disease (CHD) were evaluated in 80 overweight premenopausal women. We hypothesized that higher carbohydrate intake would be associated with a more detrimental plasma lipid profile. Dietary data were collected using a validated food frequency questionnaire (FFQ). Plasma total cholesterol (TC), triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) were determined from two fasting blood samples. In addition, selected apolipoproteins (apo) and LDL peak size were measured. Values for TC, TG and HDL were not in the range of risk classification; however, the mean values of LDL-C, 2.7 +/- 0.7 mmol/L, were higher than the current recommendations. Carbohydrate intake was positively associated with TG and apo C-III (P < .01) concentrations, and negatively associated with LDL diameter (P < .01). Participants were divided into low (<53% of energy) or high (> or = 53% energy) carbohydrate intake groups. Individuals in the <53% carbohydrate group consumed more cholesterol and total fat, but also had higher intake of polyunsaturated and monounsaturated fatty acids (SFAs). In contrast, subjects in the > or =53% group consumed higher concentrations of glucose and fructose than those in the low-carbohydrate (LC) group. In addition, subjects consuming <53% carbohydrate had lower concentrations of LDL-C and apo B (P < .01) and a larger LDL diameter (P < .05) than the > or =53% group. These results suggest that the lower LDL-C in the LC group may be related to both the amount of carbohydrate and the type of fatty acids consumed by these subjects.