Serum homocysteine is not independently associated with an atherogenic lipid profile: The Very Large Database of Lipids (VLDL-21) study

Background and aims: Hyperhomocysteinemia is an independent risk factor for cardiovascular disease, but the mechanism for this risk remains unclear. While reducing serum total homocysteine (tHcy) has been shown to decrease strokes, there is no evidence for an effect on myocardial infarctions in randomized controlled trials. This study aims to examine the relationship between tHcy and several lipid measures. Methods: Our analyses included 18,297 U.S. adults from the Very Large Database of Lipids who had an extended lipid panel including direct measurement of triglycerides (TG), and the cholesterol concentration of low-density lipoprotein (LDL-C), high-density lipoprotein (HDL-C), non-HDL-C, very low-density lipoprotein (VLDL-C), and remnant-lipoprotein cholesterol (RLP-C: IDL-C + VLDL₃-C). Additional measurements were tHcy, hemoglobin A1c (HbA1c), insulin, creatinine, and blood urea nitrogen (BUN). Subjects were categorized into tHcy quartiles. Linear regression models were performed using lipids and tHcy as dependent and independent variables respectively, and further adjusted with age, sex, HbA1c, insulin, creatinine, and BUN levels in multivariable regression. Results: In unadjusted analysis, levels of LDL-C (p < 0.001), non-HDL-C (p < 0.001) and HDL-C (p < 0.001) were 7-10% lower whereas levels of TG (p < 0.001), VLDL-C (p = 0.016) and RLP-C (p < 0.001) were 2-6% higher in the highest tHcy quartile. These associations between tHcy levels and lipids were eliminated (p-value range: 0.101-0.750) when controlling for age, sex, HbA1c, insulin, creatinine, and BUN. Conclusions: Although high levels of tHcy were associated with 2-6% higher TG-rich lipoproteins in unadjusted analysis, after adjustment for confounders our findings do not support the hypothesis that hyperhomocysteinemia is associated with an atherogenic lipid profile.