Abstract P3034: Plasma branched-chain amino acid levels and cardiovascular disease risk factors: a longitudinal analysis of a lifestyle trial

Abstract

Purpose: Branched-chain amino acid (BCAAs) levels have been associated with a higher risk of cardiovascular diseases (CVD), but studies of CVD risk factors have mainly been cross-sectional. We examined the longitudinal association between changes in BCAA levels and CVD risk factors in a lifestyle trial. Method: We used data on 708 male and female participants, aged 25-78, of the U.S. PREMIER study. Data and biospecimens were from the NHLBI Biologic Specimen and Data Repository Information Coordinating Center. Participants were divided into control, lifestyle intervention, or lifestyle plus DASH diet groups. Biomarkers were measured with NMR spectroscopy. Multiple linear regression models assessed changes in BCAA levels in relation to changes in cardiometabolic risk factors over 6 months. Estimates were expressed per SD change in BCAA levels and adjusted for age, sex, race, marital status, region, and body mass index changes. Results: Increases in BCAA levels were associated with increases in insulin resistance (HOMA-IR) (Beta = 0.22, SE = 0.07, p = 0.005), inflammation (GlycA) (Beta = 0.03 mmol/l, SE = 0.004, p <0.001), Apolipoprotein B (Beta = 0.02 g/l, SE = 0.006, p <0.001), and VLDL-cholesterol levels (Beta = 0.03 mmol/l, SE = 0.007, p <0.001).. Associations for other lipoprotein measures differed for individual BCAAs. Increases in valine were associated with LDL-cholesterol (Beta = 0.07 mmol/l, SE = 0.01, p <0.001), with no significant association for leucine and isoleucine. Isoleucine was inversely associated with HDL-C (Beta = -0.03 mmol/l, SE = 0.01, p <0.001) and Apolipoprotein A (Beta = -0.02 g/l, SE = 0.009, p = 0.004). Also, leucine was inversely associated with HDL-C (Beta = -0.04, SE = 0.01, p <0.001), Apolipoprotein A (Beta = -0.03, SE = 0.009, p <0.001) and triglyceride (Beta = -0.05 (mmol/l), SE = 0.01, p = 0.004) while valine had a direct association with HDL-C (Beta = 0.03, SE = 0.01, p <0.001), Apolipoprotein A (Beta = 0.04, SE 0.009, p <0.001) and triglyceride (Beta = 0.06, SE = 0.01, p = 0.001). BCAA levels were not significantly associated with fasting glucose and blood pressure. Conclusion: Increases in BCAA levels were associated with insulin resistance, inflammation, and unfavorable lipid profiles. These findings support BCAA levels as a target for CVD prevention. Further research is warranted to elucidate possible differences in the effects of individual BCAAs on lipid metabolism.