Serum lipoproteins in African Americans and whites with non-insulin-dependent diabetes in the US population.

BACKGROUND: Despite the significant role that dyslipidemia is believed to play in the development of cardiovascular disease in diabetes, most studies examining diabetic dyslipidemia in the United States have not been population based, and very little data are available for African Americans with diabetes. We used data from a national survey to compare the effect of diabetes on lipid concentrations in African-American and white men and women. In addition, we examined other factors related to lipid concentrations and controlled for these factors in our analyses. METHODS AND RESULTS: The Second National Health and Nutrition Examination Survey included a representative sample of 4177 African Americans and whites in the US civilian noninstitutionalized population 20 to 74 years old. These persons were classified as having non-insulin-dependent diabetes mellitus (NIDDM) (n = 720) or as being nondiabetic (n = 3457) based on an oral glucose tolerance test and a medical history of diabetes. Subjects were given an interview and physical examination that included measurement of serum lipoproteins, body mass index, body fat distribution, dietary fat intake, alcohol consumption, frequency of smoking, and use of medications. By univariate analysis, a worse profile of mean cholesterol, triglycerides, and high-density lipoprotein cholesterol levels was generally apparent in NIDDM than in nondiabetic subjects, regardless of race or sex; a similar pattern was found for the prevalence of abnormal concentrations of these lipids. Lipid profiles appeared to be worse in whites with NIDDM than in African Americans. For mean total and low-density lipoprotein cholesterol, concentrations tended to be worse in women with NIDDM than in men. When other factors significantly affecting lipid levels were adjusted by multivariate analysis, we found that in all race/sex groups, total cholesterol was higher in NIDDM than in nondiabetic subjects but differences were not significant (P = 54), triglyceride concentrations were significantly higher in NIDDM subjects (P < .0001), and high-density lipoprotein cholesterol concentrations were lower in NIDDM subjects (P = .003). An interaction of diabetes with race was found for low-density lipoprotein cholesterol (P = .0001), where concentrations were substantially lower in NIDDM than in nondiabetic subjects among African Americans (P < .01) but slightly higher in NIDDM subjects among whites (P = .33). For other lipids, no differential effect of NIDDM was found by race or sex. CONCLUSIONS: In African-American and white men and women in the United States, NIDDM is associated with a pattern of dyslipidemia that may potentiate the atherosclerotic process. Diabetic treatment should include aggressive treatment of dyslipidemia to reduce this increased risk.