|| Checking for direct PDF access through Ovid
Background: Although HDL-C is known to be inversely associated with cardiovascular risk, no causal relationship has been demonstrated. Since HDL-C comprises a group of different subfractions, they might have different effects on atherosclerosis, and its effects may be modified by the presence of diabetes mellitus.Background: Although High-Density Lipoprotein (HDL-C) is known to be inversely associated with cardiovascular risk, no causal relationship has been demonstrated. Since HDL-C comprises different subfractions, they might have different effects on atherosclerosis associated with other risk factors. In another hand, the presence of diabetes has been described either as a mediator or as a moderator of cardiovascular risk factors.Hypothesis: HDL-C subfractions have different atherogenic effects on subclinical atherosclerosis according to diabetes status.Methods: We evaluated 3,930 individuals enrolled at the baseline of the Brazilian Longitudinal Study of Adult Health aged 35 to 74 years (45.6% men) without previous cardiovascular disease and not in use of lipid-lowering agents. HDL2-C and HDL3-C were obtained by vertical auto profile method (Atherotech). Diabetes was defined as a previous medical history of diabetes or a fasting blood glucose > 126 mg/dl or a 2-hour post-load glucose test > 200 mg/dl or a glycated hemoglobin > 6.5%.Multiple linear regression models analyzed the relationship between each HDL-C subfraction and the common carotid artery intima-media thickness (cIMT).Results: The proportion of participants with the diagnosis of diabetes was 19.8%. The mean (± standard deviation) values obtained were cIMT (0.796±0.195 mm), total HDL-C (55.1±14.4 mg/dl), HDL2-C (14.9±6.7), and HDL3 (40.1±8.4 mg/dl). Total HDL-C and subfractions, as well as HDL2-C/HDL3-C ratio, were negatively associated with cIMT after adjustment for age, sex, and race (all p<0.001) and further for smoking habit, alcohol use, physical activity, LDL-C, body-mass index, waist circumference, fasting plasma glucose, triglycerides, systolic blood pressure and use of antihypertensive drugs (HDL-C: p = 0.003, HDL2-C: p = 0.01; HDL3-C: p = 0.003; HDL2-C/HDL3-C ratio: p = 0.02). When stratified by diabetes status, both HDL2-C (p= 0.03) and HDL2-C/HDL3-C ratio (p= 0.01) showed a negative association with cIMT in people with diabetes after adjusting for confounding variables. This association did not occur in individuals without diabetes (p = 0.11 and p = 0.30, respectively).Conclusion: HDL2-C and HDL3-C subfractions, as well as the HDL2/HDL3-C ratio, are inversely associated with cIMT after adjustment for traditional risk factors. This association of HDL2-C/HDL3-C ratio and HDL2-C is modified by the presence of diabetes, where it is more pronounced.